UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Low- and high-level visual perception in adults with autism spectrum disorder Shafai, Fakhri 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2018_september_shafai_fakhri.pdf [ 5.18MB ]
Metadata
JSON: 24-1.0368844.json
JSON-LD: 24-1.0368844-ld.json
RDF/XML (Pretty): 24-1.0368844-rdf.xml
RDF/JSON: 24-1.0368844-rdf.json
Turtle: 24-1.0368844-turtle.txt
N-Triples: 24-1.0368844-rdf-ntriples.txt
Original Record: 24-1.0368844-source.json
Full Text
24-1.0368844-fulltext.txt
Citation
24-1.0368844.ris

Full Text

LOW- AND HIGH-LEVEL VISUAL PERCEPTION IN ADULTS WITH AUTISM SPECTRUM DISORDER by  Fakhri Shafai  M.Ed., Saint Mary's College of California, 2012  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Neuroscience)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  July, 2018  © Fakhri Shafai, 2018    ii The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled:   Low- and high-level visual perception in adults with autism spectrum disorder______________    submitted by   Fakhri Shafai                                     in partial fulfillment of the requirements for  the degree of   Doctor of Philosophy_______________________________________________   in                     Neuroscience_____________________________________________________   Examining Committee:   Ipek Oruc, Neuroscience________________________________________________________     Supervisor        Grace Iarocci, Psychology_______________________________________________________                                                                                                                                                                        Supervisory Committee Member   Supervisory Committee Member  Anthony Bailey, Psychiatry University Examiner  Frances Chen, Psychology University Examiner    Additional Supervisory Committee Members:  Jason Barton, Opthalmology_______________________________________________________                                                                                                                                                                        Supervisory Committee Member  Naznin Virji-Babul, Physical Therapy_______________________________________________                                                                                                                                                                        Supervisory Committee Member       iii Abstract   Descriptions of atypical visual processing have been associated with Autism Spectrum Disorder (ASD). Pinpointing the source of altered perception has been challenging. It is possible that the divergence occurs early in visual processing. There have been reports of higher prevalence of refractive errors for children with ASD, but few studies assessing refractive error status of adults on the spectrum. We contribute to this gap by providing complete optometric eye exams to assess refractive status in a group of adults with ASD and Controls. We find no significant differences between our ASD and Control groups for presence or severity of refractive errors, but higher rates of myopia for both groups compared to prevalence in the general population. Results from behavioural studies suggest altered visual functioning may occur further downstream, possibly in the earliest cortical visual areas. The Enhanced Perceptual Functioning (EPF) model proposes that behavioural results can be explained as a consequence of superior perception for “simple visual material”. This model suggests that low-level properties of images may be processed differently in ASD. We examine this assertion for aspects of orientation processing, which is associated with low-level perception. In three behavioural psychophysical tasks, we examine the status of orientation discrimination, veridical perception, and orientation detection in adults with and without ASD. We find no significant differences between our ASD and Control groups for any aspect of orientation perception tested. As social difficulties are a criterion for diagnosis of ASD, various aspects of face processing have been studied in this population, with mixed results. The social motivation hypothesis and amygdala dysfunction hypothesis both suggest face processing abilities may be associated with social competency difficulties in ASD. We examine two different aspects of face processing: identification and recognition of expression, and compare performance to measures of ASD symptom severity and   iv social competence. We find impaired identity and expression perception across all tasks for our ASD group, but only expression processing was associated with social competence. These results have implications for our understanding of visual perception differences in ASD and offer recommendations for future research directions and intervention tools.      v Lay Summary   Individuals with Autism Spectrum Disorder (ASD) show differences in how they process visual information. Determining the source of these differences has been an open area of research. In this study, we began by examining the eye for refractive errors and found that rates of myopia were similar for adults with ASD and Control participants. Next, we assessed one of the most basic aspects of visual processing: orientation perception. We found no evidence for qualitative or quantitative differences between the two groups. Finally, we examined face identification and expression recognition, and found that these were both impaired in our ASD group compared to Controls. Although face identification skills were not associated with social competence or autism severity, impairments in recognizing sad expressions were significantly related to difficulties in social competence. Our research suggests that training paradigms that prioritize facial expression recognition have the potential to improve outcomes for individuals with ASD.   vi Preface  I had the primary role for experimental design and data analysis for work described in Chapter 2. Data collection was conducted by Dr. Gloria Lee, Doctor of Optometry, at her clinic at Vida Eye Care in Vancouver, British Columbia.  I planned and developed the experimental design, data collection, data analysis, and manuscript preparation with Ipek Oruc (IO) for work described in Chapter 3. Additionally, Dr. Grace Iarocci (GI) and Kimberly Armstrong provided support for participant recruitment and confirmation of diagnosis of ASD using the Autism Diagnostic Observation Schedule (ADOS). I wrote and edited the paper, with feedback from IO and GI. A version of Chapter 3 has been published. Shafai, F., Armstrong, K., Iarocci, G., & Oruc, I. (2015). Visual orientation processing in autism spectrum disorder: No sign of enhanced early cortical function. Journal of Vision, 15(18), 1-15.  I had the primary role for experimental design, data collection, data analysis, and manuscript preparation for work described in Chapter 4. Dr. Grace Iarocci and Kimberly Armstrong also contributed to this work by confirming diagnosis of ASD via ADOS. A version of Chapter 4 is in preparation for publication. Dr. Ipek Oruc provided a supervisory role for all work described in Chapters 2-4, including experimental design, data collection, data analysis, and manuscript preparation.  The name of the specific UBC Research Ethics Boards, and the Certificate Numbers of the Ethics Certificates obtained for the research described in this thesis: UBC Behavioural Research Ethics Board H12-02873 and Vancouver Coastal Health V12-02873.     vii Table of Contents  Abstract ................................................................................................................................... iii Lay Summary ............................................................................................................................ v Preface ...................................................................................................................................... vi  Table of Contents ....................................................................................................................vii List of Tables ............................................................................................................................. x List of Figures .......................................................................................................................... xi List of Abbreviations ............................................................................................................. xiv Acknowledgements ................................................................................................................. xv Dedication.............................................................................................................................. xvii  Chapter 1: Introduction............................................................................................................ 1 1.1 Visual processing pathway .......................................................................................... 1 1.1.1 Initiation of the visual pathway and optometric measures ......................................... 2 1.1.2 Principles of basic vision ......................................................................................... 4 1.1.3 Low-level vision: early visual areas ......................................................................... 6 1.1.4 High-level vision: object- and face-selective areas ................................................... 9 1.2 Face processing in the human brain ........................................................................... 10 1.2.1 Neuropsychology of face processing: lessons from acquired prosopagnosia ........... 11 1.2.2 Face processing network ........................................................................................ 14 1.2.3 Configural face processing ..................................................................................... 17 1.3 Autism Spectrum Disorder ........................................................................................ 20 1.3.1 Diagnosis of ASD .................................................................................................. 22 1.3.2 Measures of intelligence, autism severity, and social competence .......................... 23 1.3.3 Sensory issues in ASD ........................................................................................... 25 1.4 Vision in ASD ........................................................................................................... 26 1.4.1 Visuo-cognitive theories of visual processing in ASD ............................................ 27 1.4.2 Optometric concerns in ASD ................................................................................. 30 1.4.3 Low-level visual perception in ASD ...................................................................... 33 1.4.4 Face processing in ASD ......................................................................................... 37 1.5 Objectives ................................................................................................................. 51 1.5.1 Chapter 2 objectives .............................................................................................. 51 1.5.2 Chapter 3 objectives .............................................................................................. 52 1.5.3 Chapter 4 objectives .............................................................................................. 54 Chapter 2: Optometric assessment of adults with ASD......................................................... 57 2.1 Methods .................................................................................................................... 58 2.1.1 Participants ............................................................................................................ 58 2.1.2 Setup for the eye exam ........................................................................................... 58 2.1.3 Procedure .............................................................................................................. 59 2.1.4 Data analysis ......................................................................................................... 61 2.2 Results ....................................................................................................................... 62 2.3 Discussion ................................................................................................................. 88 Chapter 3: Visual orientation processing in ASD .................................................................. 93 3.1 General methods ........................................................................................................ 94   viii 3.2 Experiment 1: Precision ............................................................................................. 96 3.2.1 Methods ................................................................................................................. 96 3.2.2 Results and discussion ........................................................................................... 99 3.3 Experiment 2: Veridical perception.......................................................................... 101 3.3.1 Methods ............................................................................................................... 101 3.3.2 Results and Discussion ........................................................................................ 104 3.4 Experiment 3: Sensitivity ......................................................................................... 106 3.4.1 Methods ............................................................................................................... 107 3.4.2 Results and Discussion ........................................................................................ 110 3.5 General Discussion .................................................................................................. 112 Chapter 4: Association of social competence with face processing abilities in adults with ASD ........................................................................................................................................ 121 4.1 General methods ...................................................................................................... 123 4.1.1 Participants .......................................................................................................... 125 4.1.2 Experimental setup .............................................................................................. 126 4.2 Experiment 1: perception for facial identity ............................................................. 126 4.2.1 Stimuli ................................................................................................................. 127 4.2.2 Procedure ............................................................................................................ 129 4.2.3 Data analysis ....................................................................................................... 133 4.2.4 Results and discussion ......................................................................................... 133 4.3 Experiment 2: memory for facial identity ................................................................. 135 4.3.1 Experimental setup .............................................................................................. 135 4.3.2 Stimuli ................................................................................................................. 135 4.3.3 Procedure ............................................................................................................ 136 4.3.4 Data analysis ....................................................................................................... 137 4.3.5 Results and discussion ......................................................................................... 137 4.4 Experiment 3: face expression discrimination .......................................................... 140 4.4.1 Experimental setup .............................................................................................. 141 4.4.2 Stimuli ................................................................................................................. 141 4.4.3 Procedure ............................................................................................................ 143 4.4.4 Data analysis ....................................................................................................... 145 4.4.5 Results and discussion ......................................................................................... 146 4.5 Correlation analysis ................................................................................................. 150 4.5.1 Results and discussion ......................................................................................... 151 4.6 General discussion and conclusions ......................................................................... 167 Chapter 5: Conclusion .......................................................................................................... 174 5.1 Optometric concerns in ASD ................................................................................... 175 5.2 Low-level visual processing in ASD ........................................................................ 179 5.3 Object vs. Face-processing in ASD .......................................................................... 182 5.4 Overall conclusions ................................................................................................. 189 5.4.1 Enhanced perceptual functioning models ............................................................. 190 5.4.2 Amygdala theory of autism .................................................................................. 192 5.4.3 Social motivation hypothesis ............................................................................... 195 5.4.4 Implications for general models of ASD .............................................................. 197 5.5 Strengths and limitations ......................................................................................... 199   ix 5.6 Future directions ...................................................................................................... 201 Bibliography .......................................................................................................................... 203   x List of Tables   Table 2.1 Optometric values for participants with ASD…………………………………………65  Table 2.2 Optometric values for control participants…………………………………………….70  Table 2.3 Types and severity of refractive error for participants with ASD and controls……….72  Table 2.4 Severity of astigmatism based on the CYL values for participants with ASD and   Controls………………………………………………………………………...………………...75  Table 2.5 Classification for each type of astigmatism as defined by the AXIS value…………...77  Table 2.6. Classification of myopia in the right eye for specific age ranges…………………….81  Table 3.1. Participant demographics for Experiment 1………………………………………….97  Table 3.2. Participant demographics for Experiment 2…………………………………..…….102  Table 3.3. Participant demographics for Experiment 3………………………………………...107  Table 4.1. Participant demographics for all face experiments…………………………………126    xi List of Figures   Figure 2.1  Number of eyes with SphEq values ......................................................................... 74 Figure 2.2  The number of eyes with and without CYL astigmatism severity in diopters ........... 76 Figure 2.3  The number of eyes with types of AXIS astigmatism .............................................. 78 Figure 3.1  Experiment 1: Protocol for measuring precision of orientation perception using an orientation discrimination task. ................................................................................................. 91 Figure 3.2 Results of Experiment 1 ........................................................................................... 93 Figure 3.3  Experiment 2: protocol for measuring accuracy of perceived orientation using an adjustment task.. ....................................................................................................................... 96 Figure 3.4 Results of Experiment 2. (A) bias and (B) precision ................................................. 97 Figure 3.5  Experiment 3: Protocol for measuring orientation sensitivity using a detection task ............................................................................................................................................... 101 Figure 3.6 Results of Experiment 3 ......................................................................................... 103 Figure 3.7. Data from participants with ASD who have a Block Design test peak.................... 110 Figure 4.1 The face stimuli……………………………………………………………………..119  Figure 4.2 The house, or portico, stimuli……………………………………………………….120  Figure 4.3 Protocol for the 5-AFC identification task………………………………………….121  Figure 4.4 Results from Experiment 1………………………………………………………….125  Figure 4.5 Results from Experiment 2: CFMT…………………………………………………128  Figure 4.6 Results from Experiment 2: Face-specific memory………………………………...129  Figure 4.7 Correlation between performance on the upright and inverted versions of the CFMT   …………………………………………………………………………………………………..131    xii Figure 4.8 Illustration of the stimuli and procedure for Experiment 3…………………………133 Figure 4.9 Expression discrimination results…………………………………………………...138 Figure 4.10 Normalized expression discrimination thresholds from Experiment 3……………139 Figure 4.11 Mean MSCS scores for ASD (red) and controls (blue) in each of seven distinct domains of social competence………………………………………………………………….140 Figure 4.12 Mean AQ scores for ASD (red) and controls (blue)……………………………….141 Figure 4.13 Correlation between social competence and face-specific perception…………….142 Figure 4.14 Correlation between ASD symptom severity and face-specific perception……….143 Figure 4.15 Correlation between social competence and face-specific memory……………….144 Figure 4.16 Correlation between ASD symptom severity and face-specific memory…………145 Figure 4.17 Correlation between social competence and expression discrimination…………..146 Figure 4.18 Correlation between social competence and expression discrimination thresholds for angry, happy, and sad expressions for the ASD participant group……………………………..147 Figure 4.19 Correlation between social competence and expression discrimination thresholds for angry, happy, and sad expressions for the Control participant group…………………………..149 Figure 4.20 Correlation between social competence subdomains and expression discrimination thresholds for sad expression for the ASD participant group…………………………………..150 Figure 4.21 Correlation between social competence subdomains and expression discrimination thresholds for sad expressions for the Control participant group………………………………152 Figure 4.22 Correlation between social competence subdomains and expression discrimination threshold for angry expressions for the Control participant group……………………………..154 Figure 4.23 Correlation between ASD symptom severity and expression discrimination threshold………………………………………………………………………………………...155   xiii Figure 4.24 Correlation between ASD symptom severity and expression discrimination thresholds for angry, happy, and sad expressions for our ASD participant group……………..157 Figure 4.25 Correlation between ASD symptom severity and expression discrimination thresholds for angry, happy, and sad expressions for the Control participant group…………...158    xiv List of Abbreviations   ADDL: autism and developmental disorders lab ADOS: autism diagnostic observation schedule AFC: Alternative forced-choice ASD: autism spectrum disorder ATR: against-the-rule axis AQ: autism spectrum quotient AXIS: axis of astigmatism BDT-peak: block design test peak BFRT: Benton face recognition task CEFT: children’s embedded figures test CFMT: Cambridge face memory test CPD: cycles per degree CYL: cylinder power DP: developmental prosopagnosia DSM-III: diagnostic and statistical manual of mental disorders, third edition DSM-5: diagnostic and statistical manual of mental disorders, fifth edition EEG: electroencephalography EPF: enhanced perceptual functioning ERP: event-related potential FDR: false discovery rate FFA: fusiform face area FIQ: full scale intelligence quotient fMRI: functional magnetic resonance imaging IFC: Interval forced-choice IQ: intelligence quotient LFI!: Let’s face it! LGN: lateral geniculate nucleus MEG: magnetoencephalography MSCS: multidimensional social competence scale OBL: oblique axis OFA: occipital face area pSTS: posterior superior temporal sulcus RMS: root mean square ROI: region of interest SPH: spherical power SphEq: spherical equivalent V1: visual area 1 V2: visual area 2 WASI-II: Wechsler abbreviates scale of intelligence- second edition WC: weak coherence WTR: with-the-rule axis    xv Acknowledgements   My deepest gratitude goes to my supervisor, Dr. Ipek Oruc for her mentorship, guidance, and support throughout this endeavor. She encouraged my growth as a scientist and provided me the opportunity to develop new skills while examining aspects of visual perception in ASD. Her supervision and patience during research design, data collection, and analysis throughout this program were indispensable. Her example of dedication, perseverance, and creative problem-solving ability will continue to inspire me in my career and life and general. Thank you for everything, Ipek!  Many thanks to Dr. Grace Iarocci, Kimberly Armstrong, and the members of the Autism and Developmental Disorders Lab (ADDL) at Simon Fraser University for their collaborative efforts during this research project. The design, recruitment, data collection, or interpretation of our results could not have been possible without support from Dr. Iarocci and her lab.  I would also like to extend my gratitude to the members of my supervisory and comprehensive exam committee; Drs. Jason Barton, Grace Iarocci, Naznin Virji-Babul, and Todd Woodward. In addition, I would like to thank the final oral examination committee members Drs. Anthony Bailey and Frances Chen. I am grateful for your patience, guidance, and time throughout this process. Many thanks to my external examiner, Dr. Armando Bertone. His valuable feedback and questions provided me with an opportunity to improve my thesis.  The Autism Research Training (ART) program provided me with the valuable opportunity to meet with ASD researchers in a variety of disciplines and develop peer-connections throughout Canada. The ART mentorship allowed me to work closely with the top scientists in my field. Many thanks to my mentors Drs. Grace Iarocci, and James Tanaka. I appreciate your valuable insight into my research directions and career advice.   xvi  My deepest gratitude goes to our ASD participants and their families who have agreed to contribute to this research with their time and efforts. This research would not be possible without the interest and support of individuals with ASD. Many thanks to those participants who told others about our study. I am also grateful to all the individuals who acted as Controls in our experiments, without your participation we would be unable to draw meaningful conclusions from the data.  I would like to thank all of my friends who have provided me emotional and moral support throughout the years. My friends that are current and past members of the Vis-Cog group were the best peers, support group, and VSS roommates anyone could ask for: Andrea Albonico, Esther Alonso Prieto, Aenne Brielmann, Jolande Fooken (and Jan!), Manuela Malaspina, and Kim Meier. My classmates in the Department of Neuroscience graduate program became my first friends in Canada and introduced me to the many things to love about Vancouver: Eitan Anenberg, Sonia Brodie (and Fraser!), Nate Holms, Cassie MacRae, and Paul Metzak. Finally, my friends outside the program, from childhood to more recent friendships, have been incredible supports during difficult times: Robyn and Josh Coquelle, Ben Ebert, Katelyn Eng, Richard Faulder, Amy Gonzales, Michelle Herrera, David and Danielle Holquin, Will McCosker, Kim Mc Nelly, Chris Newbigging, Carey and Dave Ragni, Nicola Ray, Emily Roldan (and Greg!), Erin van Ark, and Maeve Wickham. Thanks to you all!  Finally, I would like to extend my heartfelt gratitude to my parents, sister, brother, sister-in-law, and my extended family for their constant support. Their faith in me never wavered and they always believed I could complete this program, even when I did not share the sentiment. My parents and siblings provided encouragement and gentle guidance throughout this program. Their love and support made the completion of this program possible.     xvii Dedication          This dissertation is dedicated to Patrick James Kirbach and his family. Patrick is the first person with autism spectrum disorder that I ever had the pleasure to work with and he forever changed my life. It is for him that I seek to understand and eventually treat the sensory sensitivities that impact so many people on the spectrum.  Thank you for letting me be a part of Patrick’s journey, Rosemary and Lee!   1 Chapter 1: Introduction   This thesis seeks to advance current understanding of visual processing in adults with Autism Spectrum Disorder (ASD) and to further examine and characterize any potential differences in the way individuals with this developmental disorder process visual stimuli of varying complexity. An additional aim of this research is to investigate any potential relationship between altered visual processing and the behavioural abnormalities associated with ASD. In this chapter, I will provide an introductory background to the stages of visual function and characteristics of ASD that will be the focus of studies of visual processing in ASD in later chapters.  1.1 Visual processing pathway  Beginning with the optical elements of the eyeball that refract and focus light on the retina, the photoreceptor layer of the retina responds to light absorption by initiating a signal that travels to the retinal ganglion cells and on through the optic nerve (Hubel & Wiesel, 1960). This signal is partially crossed at the optic chiasm, before the optic tracts project to the lateral geniculate nucleus (LGN), where all the axons synapse before spreading out as optic radiations that continue to send the signal to the primary visual cortex located in the occipital cortex (DeYoe & Van Essen, 1988). From minute details to intricate scenes, the human visual system proficiently carries out tasks such as detection, discrimination, and recognition of images with remarkable speed and accuracy.   A number of human visual areas have been recognized as possibly playing a role in processing stimuli of various levels of complexity (Tootell, Dale, Sereno, & Malach, 1996). The precise location and functional role of each area remain under investigation. Two main principles   2 have been put forth in order to describe the manner by which each area contributes to processing of visual stimuli: hierarchical processing and functional specialization. The principle of hierarchical processing proposes that information relayed to the visual cortex is processed gradually in steps (DeYoe & Van Essen, 1988; Grill-Spector & Malach, 2004). Visual information is first handled as simple, localized forms such as spatial frequency or orientation angles before being transformed into more complex representations in a sequence of processes.   The principle of functional specialization suggests that neural pathways are specialized to process different characteristics of visual stimuli (Friston et al., 1997; Zeki et al., 1991). This principle proposes parallel processing streams are dedicated to distinct functional tasks. The two main hierarchical processing pathways are the dorsal stream, or "where" pathway, and the ventral stream, also called the "what" pathway. The dorsal stream is referred to as the "where" pathway because these are the areas of the cortex that increase activity during tasks requiring spatial localization or visually guided actions (Goodale & Milner, 1992; Mishkin, Ungerleider, & Macko, 1983). Tasks that rely on object and form recognition, however, most often increase activity in the "what" pathway of the ventral stream (Grill-Spector & Malach, 2004; Milner & Goodale, 2008). This thesis focuses on the functions of the ventral pathway in individuals with ASD.   1.1.1 Initiation of the visual pathway and optometric measures  The earliest attempts to determine how visual information was processed relied on animal models and postmortem analysis. Importantly, these gross anatomical studies of the eye revealed that light is refracted by both the cornea and lens onto the surface of the retina. A lens is a transparent structure consisting of layers of protein and water and is capable of changing shape   3 via ciliary muscles (Smerdon, 2000). Images become inverted due to projection as they pass through the vitreous humor before reaching the photoreceptors of the retina at the back of the eye. When a person fixates on a stimulus, the image is focused at a region of the retina known as the fovea. This is the location of highest acuity, or highest spatial resolution, of vision (Rossi & Roorda, 2010).  In optometric settings, acuity is often measured using a "Snellen" acuity chart. It is reported as a ratio with "normal" vision being 20/20 (in the United States) or 6/6 (worldwide) as this chart was designed to be viewed from 20 feet or 6 meters away (Peters, 1961). The denominator changes as an indication of the spatial resolution of one's vision at a certain distance, where lower values denote better visual acuity. For example, a person with 20/15 (or 6/4.5) visual acuity has better than "normal" vision and can see as clearly at 20 feet as an individual with 20/20 can see at 15 feet (or 4.5 meters). A person with 20/30 (or 6/9) visual acuity, however, has worse than "normal" vision and has to stand 20 feet away to see something that a person with normal acuity can see from 30 feet (or 9 meters). Visual acuity can also be measured at near distances, which is useful for assessing an individual’s ability to accommodate and focus at closer distances, like those necessary for reading or working on a computer (Crick & Khaw, 1997).  A refractive error occurs when the shape of the eye and/or lens prevents light from focusing directly on the retina. This can be caused by the eye being too short or long, changes in the shape of the cornea, or aging of the lens (Resnikoff, Pascolini, Mariotti, & Pokharel, 2008). The incidence of refractive errors in children varies worldwide, with some regions having higher rates of refractive errors than children from other parts of the world (Zhao et al., 2000). It is hypothesized that both genetics and environmental factors contribute this variation (Kleinstein,   4 Jones, Hullett, & et al., 2003). Regardless of the underlying cause, people with refractive errors experience symptoms such as blurred vision, squinting, and eye strain. Myopia, or nearsightedness, is when objects close up can be focused, while objects further away appear blurry. Hyperopia, or farsightedness, describes when objects in the distance are seen more clearly than objects close by. An astigmatism is a defect in the spherical shape of the cornea or the lens and does not allow light to focus evenly on the retina, resulting in distorted images. Presbyopia describes the natural aging of the lens where it is no longer able to change shape enough to focus on nearby objects clearly, usually causing refractive errors by late middle-age (Koretz, Cook, & Kaufman, 1997; Koretz, Cook, & Kaufman, 2001). Refractive errors can often be treated via prescription eyeglasses or contact lenses (Emsley, 1925; Schiefer, Kraus, Baumbach, Ungewiß, & Michels, 2016). In some cases, refractive surgery can change the shape of the cornea to allow for better visual acuity without the use of prescription lenses.   In the context of vision research, it is common practice amongst visual scientists to ensure that subjects have "normal" or "corrected-to-normal" vision in order to minimize potential confounds due to optometric issues. Refractive errors must be corrected before studies can proceed. This ensures that any findings cannot be attributed to low-vision due to uncorrected refractive errors.   1.1.2 Principles of basic vision   The retina is comprised of ten layers (Smerdon, 2000), with three major cell types being relevant to the present line of research: photoreceptor cells, bipolar cells, and ganglion cells. The layer just below contains two types of photoreceptor cells: rods and cones. Cones are utilized in settings with an abundance of light, whereas rods function primarily in low-light conditions   5 (Crick & Khaw, 1997). Once a photoreceptor is exposed to light and becomes hyperpolarized, a response relative to the magnitude of the signal is sent to the bipolar cells. The bipolar cells then relay the signal to the retinal ganglion cells, each with a corresponding receptive field. The receptive field is the particular region in the visual field that will modify the firing of that neuron if triggered by a visual stimulus. Each receptive field is arranged with a disc called the "center" and larger peripheral ring called the "surround." The center and surround are functionally distinct and respond oppositely to light, such that if the center is activated by light then the surround is inhibited, and vice versa (Enroth-Cugell & Robson, 1966). A retinal ganglion cell's receptive field is composed of input from any photoreceptors connected to it via synapse and can be described as being either "on-center" or "off-center." An "on-center" cell increases firing when light hits the center of the field and is inhibited when light hits the surround. Alternatively, "off-center" cells are stimulated when light hits the surround and inhibited when it hits the center. Collectively, the combined activity of both on-center and off-center retinal ganglion cells contributes to edge detection in the early visual areas of the cortex (Zeck, Xiao, & Masland, 2005). The axons of the retinal ganglion cells go on to make up a large portion of the optic nerve (Hubel & Wiesel, 1960), which then relays the information from the receptive field to the visual cortex via the lateral geniculate nucleus.  Single-unit recordings have been an important tool for studying the function of particular neurons within the cortex. Using a microelectrode system, it is possible to measure whether an electrophysiological response is occurring in the neuron of interest. Action potentials generated by neurons can be detected as a change in voltage over time. This method has been utilized in animal studies to probe the function of a region in response to visual stimuli of varying complexities (Barraclough & Perrett, 2011; De Valois, William Yund, & Hepler, 1982; DeYoe   6 & Van Essen, 1988; Enroth-Cugell & Robson, 1966; Hubel & Wiesel, 1968). Despite the wealth of information provided by these studies, their applicability to human visual cognition remains limited. In human patients with Parkinson’s disease or epilepsy, microelectrodes may be used to pinpoint the location for subsequent treatments (Mukamel & Fried, 2012; Rey et al., 2015). These rare opportunities to study a specific neuron's activity in the awake patient have allowed the mapping of the functional specialization of single cells in the visual cortex and address questions unique to the human visual system.  Mapping between spatial locations of the retina and the cortex utilizes a technique called visual field topography. Positions that are next to one another on the retina's visual field will project to nearby regions of the visual cortex, often referred to as retinotopic representation. The topographical cortical representation on the retina can be measured once a subject fixates and visual stimuli are displayed at a particular location (DeYoe & Van Essen, 1988; Grill-Spector & Malach, 2004). The representation of the fovea is larger in the cortex compared to that of the periphery, often referred to as cortical magnification, resulting in less cortical space being devoted to outlying portions of the visual field (Daitch & Green, 1969). In line with the hierarchical principle, activation occurs at the earliest visual areas and then spreads to later visual areas.  1.1.3 Low-level vision: early visual areas  The earliest visual areas activated by an image are those which process the most basic properties of visual stimuli, often referred to as Visual Area 1 (V1) and Visual Area 2 (V2). These early visual areas are organized in symmetric bands on the unfolded cortical hemispheres (Grill-Spector & Malach, 2004), and are considered the primary destinations for all visual stimuli   7 coming to the occipital lobe as most axons of the LGN project to them. Primary sensory information are broken down into feature-based sensory cues such as orientation and spatial frequency (DeYoe & Van Essen, 1988). Other image properties (e.g. shape, texture, etc.) are processed further downstream by cells tuned to those particular characteristics.   Single-unit recording studies of visual function in animals in the 1960's indicated that neurons in the early visual areas were selectively tuned to respond to orientation, spatial frequency, and motion direction aspects of a visual stimulus (De Valois et al., 1982; Enroth-Cugell & Robson, 1966; Hubel & Wiesel, 1960). Individual neurons show selectivity for multiple aspects of visual processing in these earliest areas, including orientation and spatial frequency (DeYoe & Van Essen, 1988). For instance, a neuron may have an electrophysiological response to a line presented at a horizontal orientation while another neuron may only respond to lines presented in a vertical orientation (Vázquez, Cano, & Acuña, 2000). The same neuron may only fire when presented with stimuli within a certain range of spatial frequencies, or the number of cycles of luminance modulation per degree of visual angle (Tootell, Silverman, & De Valois, 1981). Thus, the electrophysiological response to various stimuli of choice can be measured in the same isolated neuron, with a much higher spatial and temporal resolution than that allowed by fMRI or other neuroimaging techniques.  To probe the functionality of the earliest visual areas in human observers, behavioural studies, including visual psychophysical paradigms, have been used. Psychophysics is a branch of psychology devoted to studying the relationship between physical stimuli and their induced perceptions (Fechner, 1860). Stimulus-response relationships are often measured in terms of thresholds. Determining the minimum amount of signal strength for a particular aspect of a   8 stimulus necessary to evoke a response allows one to probe the visual system in terms of its limits.  As coding for orientation has been associated with the earliest levels of visual processing, a number of studies have investigated the abilities of the visual system to discriminate between different orientations. In the mid 1800s researchers found that processing is superior at cardinal angles (horizontal and vertical) compared to oblique angles (e.g. 45 degrees diagonal; Appelle, 1972; Fechner, 1860). This phenomenon, termed the oblique effect, has been demonstrated for various aspects of orientation processing, including detection, discrimination, and Vernier acuity (Camisa, Randolph, & Sandra, 1977; Campbell, Kulikowski, & Levinson, 1966; Emsley, 1925; Essock, 1980; Westheimer & Beard, 1998). Discrimination thresholds enable researchers to infer the precision of orientation perception around various base orientations (Vázquez et al., 2000; Vogels & Orban, 1986). Consistently, the oblique effect has been observed as higher precision around cardinal angles (i.e. lower discrimination thresholds) compared to oblique orientations.   Another aspect of the lowest levels of visual processing examined by psychophysicists is spatial frequency. Neurons in the earliest visual areas frequently demonstrate a preference for spatial frequency, firing selectively for stimuli with narrow range of spatial frequencies (DeYoe & Van Essen, 1988). In experiments that investigate neuronal selectivity to a limited range of spatial frequencies, a common stimulus chosen is the Gabor patch. These images are localized in both space and spatial frequency and are assumed to resemble the visual receptive fields of V1 and V2. By varying contrast, it is possible to measure the contrast threshold for detection of the grating of a specific spatial frequency. The Michelson contrast of a grating is defined as the maximum luminance minus the minimum luminance divided by twice the mean luminance. Contrast sensitivity is the reciprocal of the contrast threshold, and if a subject's contrast   9 sensitivity is tested over a range of spatial frequencies, it is possible to obtain a contrast sensitivity function (Campbell & Robson, 1968; Daitch & Green, 1969). The maximum contrast sensitivity occurs around middle frequencies (e.g. 2-4 cycles per degree) for individuals with normal vision, and sensitivity falls for lower and higher frequencies resulting in a characteristic inverted-U shaped curve (Campbell & Robson, 1968; Enroth-Cugell & Robson, 1966).  1.1.4 High-level vision: object- and face-selective areas  Single-case studies on patients with brain lesions and functional magnetic resonance imaging (fMRI) studies have led researchers to conclude that occipitotemporal areas of the brain are selective for more complex aspects of visual images, demonstrated by increased firing rates and preference for high-level stimuli such as faces. These regions are associated with coding for higher-level aspects of visual stimuli, such as the identity of a face. Subjects show greater activation in object-selective areas when viewing pictures of objects compared to textures or scrambled objects (Grill-Spector, 2003; Grill-Spector & Malach, 2004; Malach et al., 1995). This collection of areas in the ventral visual pathway is anteriorly and laterally positioned "downstream" from the early visual areas.  Along the ventral pathway, object-selective areas have a greater response to stimuli requiring recognition of objects whereas dorsal activation is greatest in the context of object manipulation (Goodale, Milner, Jakobson, & Carey, 1991; Grill-Spector & Malach, 2004). These relatively dispersed form recognition areas are the focus of intense debate as to their functional specificity for a variety of high-level visual images, including the human face.  As with lower-level stimuli, single-unit cell recordings in primate brains also inform our understanding of face and object processing. Action potentials are recorded for an individual unit   10 while primates are viewing images of faces or objects. Just as neurons in the early visual areas can be selectively tuned to respond to stimuli with a certain orientation, these studies have revealed that neurons in higher level visual areas can be tuned to selectively respond to object stimuli (I. Fujita, Tanaka, Ito, & Cheng, 1992; Vogels, 1999) and face stimuli (Leopold, Bondar, & Giese, 2006; Pinsk, DeSimone, Moore, Gross, & Kastner, 2005; Romanski & Diehl, 2011; Zhu et al., 2013). Additionally, single-cell recordings in primate amygdala, a structure in the limbic system typically associated with processing extreme emotions, have indicated that this structure plays a role in processing faces or the eyes presented alone (Leonard, Rolls, Wilson, & Baylis, 1985; Sah, Faber, Lopez De Armentia, & Power, 2003; Todorov, 2012). It has also been demonstrated that a single unit in the posterior hippocampus of humans can respond selectively to an individual identity (Quiroga, Reddy, Kreiman, Koch, & Fried, 2005). These single-unit recording studies indicate that the individual neuron plays an important, but not necessarily solitary, role in processing of social stimuli; for a review see (for a review see Barraclough & Perrett, 2011). Multiple aspects of face processing in humans and primates, such as viewing angle, expression, and identity, are simultaneously analyzed by specialized clusters of neurons in the brain to integrate the information for social understanding and interaction.   1.2 Face processing in the human brain  The most socially salient of all visual stimuli is arguably the human face. It provides a medium for communicating a variety of socially-relevant information. For example, identification (who is that person?), social engagement (what is that person looking at?), and emotional state (how is the person feeling?) can all be conveyed by a person's face. It is commonly believed that humans have expert face-processing skills and perform impressive face identification and discrimination   11 tasks across a wide variety of settings (Farah, Wilson, Drain, & Tanaka, 1998). The qualitative manner by which faces are processed, and whether it is distinct from the processing of other types of visual stimuli, remain under debate.   1.2.1 Neuropsychology of face processing: lessons from acquired prosopagnosia  Neuropsychology is a branch of scientific inquiry that seeks to pair specific psychological processes with the structure and function of the human brain. This experimentally driven field relates behaviour and cognition to brain functioning (Behrmann, 2003). Neurological disorders that impact behaviour and/or cognitive abilities offer neuropsychologists an opportunity to combine various techniques to probe the functional specialization of brain structures in light of the subsequent issues unique to a particular patient population.  One of the richest, albeit rare, sources for studying neuropsychological processing of human faces is found in a disorder known as prosopagnosia, or face blindness. Agnosia describes the inability to recognize a sensory stimulus, most often due to some form of brain damage. Individuals with acquired prosopagnosia had intact face processing abilities prior to a stroke, injury, or illness that caused one or more occipitotemporal lesion (Barton, 2008; Barton, Cherkasova, Press, Intriligator, & O'Connor, 2004; Damasio, Damasio, & Van Hoesen, 1982; Kanwisher & Yovel, 2006). The individual becomes severely impaired with face recognition following the brain lesion. While extremely rare, these individuals provide the unique opportunity to relate the visual impairment to damage of specific regions in the brain.    Despite having typical cognitive abilities and visual processing in other aspects of vision, individuals with this disorder are unable to learn new faces or recognize familiar ones (Barton, Cherkasova, Press, et al., 2004). They are often able to recognize a face as a face, but whose face   12 it is remains elusive. In some cases, the impairment extends to mild object agnosia, but a primary deficit in prosopagnosia is that face processing is diminished while low-level vision and cognitive function are intact (Barton, 2011). Many individuals with this disorder develop visual strategies to recognize familiar people, such as hairstyle, birthmarks, or body features. Others may rely on their other senses, such as identifying a person's voice (Cousins, 2013; Liu, Pancaroglu, Hills, Duchaine, & Barton, 2016). Regardless of these strategies, many individuals with prosopagnosia report that their face-selective impairment has negatively impacted personal and/or professional relationships (Feigin, Barker-Collo, McNaughton, Brown, & Kerse, 2008).  A number of visual tests have been developed to assist researchers in identifying individuals with prosopagnosia, including famous faces tests (Barton, Cherkasova, & O'Connor, 2001; De Renzi, Faglioni, Grossi, & Nichelli, 1991), the Benton Facial Recognition Test (Benton & Van Allen, 1968), the Warrington Recognition Memory Test for Faces (Warrington, 1984), and the Cambridge Face Memory Test (Duchaine & Nakayama, 2006). Famous face tests are difficult to standardize as each individual tested may not have been previously exposed to the famous individuals or the famous faces tested may only be appropriate for patients in a certain age range. In addition, different experiments have used in-house famous face tests, making it difficult to draw conclusions when comparing multiple studies. The Benton Face Recognition Task (BFRT) uses images where the hair and clothes have been cropped to reduce the likelihood that those features could be used to differentiate between individuals. Subjects are presented with a target face and six test faces below before being asked to identify which test face matches the target (Benton & Van Allen, 1968). The BFRT, while still widely used, has been questioned by some researchers as it allows for feature matching as the target and test faces are shown at the same time (Duchaine & Nakayama, 2006). The Warrington Recognition Memory Test for Faces   13 has been criticized for using images that include non-face information that can be used to identify faces (Duchaine & Nakayama, 2006).  The Cambridge Face Memory Test (CFMT) was designed to utilize the strengths of the previous measures of face recognition and memory while avoiding the previously mentioned pitfalls (Duchaine & Nakayama, 2006). This measure also presents six target faces to participants, but does not include the exemplar target face when participants are later asked which of three test faces was the target. The CFMT presents the test faces in novel views and with noise, allowing for greater difficulty and avoiding ceiling effects in controls (Duchaine & Nakayama, 2006). Additionally, the CFMT has the option to test upright and inverted orientations, allowing for upright face advantage scores to be calculated. In typical individuals, the upright CFMT score average is 80.4 percent while the overall mean for individuals with prosopagnosia is 50.7 percent, and near chance for the novel images with noise at 34.9 percent (Duchaine & Nakayama, 2006). This test is available online and has been completed by thousands of participants, allowing robust norms to be established.   With the ability to pinpoint the location of a lesion via neuroimaging, it is possible to compare the nature of the individual's behavioural impairment to the damaged area and infer the function of that region. These studies have revealed that areas in the occipitotemporal cortex provide a critical contribution to various aspects of face recognition.  An important limitation of studying face processing in acquired prosopagnosia is that brain lesions vary in extent and location and are thus distinct for each patient. Lesions often go beyond the bounds of a particular structure of interest, making it difficult to determine if the behavioural results are due to the loss of a structure or if the surrounding area is also involved in a particular aspect of face processing. Assigning functional tasks to a property of face processing   14 is especially complicated in patients having more than one lesion, as it becomes difficult to know whether the impairment is a downstream effect of one lesion over another. These issues present challenges to our ability to generalize the structural findings in acquired prosopagnosia to face processing in typical individuals, but nevertheless provide some regions of interest for researchers looking to understand the network of structures necessary for processing the human face.   1.2.2 Face processing network  It is generally agreed that three major cortical regions make up what is known as the "face core network." This network consists of the Occipital Face Area (OFA), the Fusiform Face Area (FFA), and the posterior superior temporal sulcus (pSTS) (Gauthier, Tarr, et al., 2000; Haxby, Hoffman, & Gobbini, 2000; Rossion, 2008a). The right hemisphere tends to show a stronger response to face stimuli (Fox, Iaria, & Barton, 2009; Rossion, Hanseeuw, & Dricot, 2012; Sergent, Ohta, & Macdonald, 1992).  According to an influential model, the OFA is involved in "early stage" processing that then feeds into the FFA and pSTS (Pitcher, Walsh, & Duchaine, 2011). It is located in the lateral occipital area, and bilateral lesions in this area can impede normal face perception (Rossion, Caldara, et al., 2003). Functional localization studies have shown this region has higher activity when presented with faces compared to objects (Pitcher et al., 2011), and may be involved in processing the features of a face (Nichols, Betts, & Wilson, 2010). It has been suggested that the OFA is the first step within a hierarchical network used for face perception (Gauthier, Skudlarski, Gore, & Anderson, 2000; Haxby et al., 2000), although the FFA and pSTS may   15 provide feedback and modify the OFA’s response to stimuli (Haxby et al., 2000; Pitcher et al., 2011).   The FFA is the most studied structure within the face core network and is found on the ventral surface of the temporal lobe in the fusiform gyrus. It was first identified in neuroimaging studies as the region that had higher levels of activation when subjects were shown faces compared to object stimuli (Kanwisher, McDermott, & Chun, 1997; Kanwisher & Yovel, 2006; Sergent et al., 1992). Lesion studies of prosopagnosia patients have largely indicated that damage to the FFA is behind the acquired loss of ability to distinguish between faces (Barton, 2008; Barton, Cherkasova, Press, et al., 2004; Wada & Yamamoto, 2001). This suggests that the FFA is a specialized region within the brain that plays an instrumental role in allowing for individuation of human faces.   Empirical evidence from face perception studies suggests that the FFA is not solely dedicated to processing of human faces, but individuates between highly similar stimuli, a skill that is vital to face recognition, but not exclusively so. Proponents of the "domain-general" hypothesis suggest that face perception is strongly associated with this area because individuation of faces is a skill that is constantly used throughout the lifespan (Ishai, Schmidt, & Boesiger, 2005), but this does not necessarily mean that the region only performs individuation for faces. For instance, the FFA may be specialized for recognizing distinctions between objects with which one has expertise (e.g. a car expert looking at cars) (Bilalić, Langner, Ulrich, & Grodd, 2011; Gauthier, Skudlarski, et al., 2000). While subjects often display the greatest level of FFA activation when viewing faces, activity in the FFA is higher for experts viewing objects related to their expertise than novices, suggesting that higher activation levels for faces may be   16 the result of years of practice rather than an inherent purpose dedicated to faces (Bilalić et al., 2011; Gauthier, Skudlarski, et al., 2000).   The pSTS is the final member of the "face core network" and is located in the superior temporal sulcus. It may be involved in the dynamic aspects of face processing, such as expression and eye movement (Hoffman & Haxby, 2000; Puce, Allison, Bentin, Gore, & McCarthy, 1998). It is proposed to play an important role in emotion processing (Fox, Moon, Iaria, & Barton, 2009; Said, Moore, Engell, Todorov, & Haxby, 2010), although case studies have indicated that lesions in this area may disrupt both face identity and expression processing (Fox, Iaria, et al., 2009) . It has been suggested that specific regions of the STS code information related to face identification while other regions code information for face expressions (Winston, Henson, Fine-Goulden, & Dolan, 2004).    In addition to the three areas of the "face core network", other brain structures have been proposed to make up an "extended face network" (Haxby et al., 2000). The amygdala is located within the temporal lobe and both provides and receives input between sensory areas and structures associated with long-term memory (Sah et al., 2003). It has been suggested that it plays a role in responding to the emotions displayed in a person's facial expressions. One study of a patient with bilateral lesions in the amygdala found that identity recognition abilities were not impacted, but recognition of fear was severely diminished by damage to this region (Adolphs, Tranel, Damasio, & Damasio, 1994). More recently, it has been suggested that the amygdala is activated when viewing extreme, atypical, or unexpected stimuli (Todorov, 2012). Thus, intense face emotions elicit strong amygdala signals because it is primed to notice extremes in stimuli. This is consistent with arguments that the amygdala responds more to unfamiliar faces than to faces of close friends or family. Familiar faces elicit less of a response in   17 the amygdala because they are regularly encountered in the person's daily life (Gobbini & Haxby, 2007).   Event-related potentials (ERP) can be measured using scalp electrodes and provide excellent temporal resolution for responses to visual stimuli. One of the most robust findings associated with the neural processing of faces is a component of the ERP known as N170, which refers to the negative potential produced roughly 170ms after a subject is presented with a visual stimulus. There is a larger amplitude in response to human faces or human eyes in isolation compared to N170 amplitudes for other objects (Shlomo Bentin, Allison, Puce, Perez, & McCarthy, 1996; Oruc et al., 2011). This response is strongest in electrodes over the occipitotemporal area, consistent with face-selective activity being concentrated in the fusiform and inferior-temporal gyri (Dalrymple et al., 2011; Rossion & Jacques, 2008; Rossion, Joyce, Cottrell, & Tarr, 2003), especially for electrodes located to the right side of the brain (S. Bentin & Deouell, 2000). The temporal resolution afforded by electrophysiological studies can be combined with neuroimaging data to obtain a more complete understanding of the location and time course for processing of faces.  1.2.3 Configural face processing  One widely accepted model of face recognition strategy is known as configural processing. Configural face perception suggests that faces are processed globally as a single gestalt, or "as a whole" instead of in a part-based manner with a focus on the local features (Behrmann, Richler, Avidan, & Kimchi, 2015; Rossion, 2008b; Tanaka & Farah, 1993; Van Belle, De Graef, Verfaillie, Rossion, & Lefèvre, 2010). The position of two eyes above a nose, which is centered   18 above the mouth, refers to the first-order relationship between the features of a face. This configuration contributes to visual detection of the face as a distinct class of stimuli.   A number of experimental paradigms designed to examine mechanisms of face processing have supported the configural hypothesis (Maurer, Le Grand, & Mondloch, 2002; McKone, Kanwisher, & Duchaine, 2007). One such class of experiments has yielded a classic behavioural finding referred to as the Composite Face Effect, in which combining the upper half of one individual's face with the lower have of another individual's face results in subjects being more likely to perceive it as a new identity (Hole, 1994; A. Young, W. , Hellawell, & Hay, 1987). This task asks participants whether one half (e.g. the upper face) of the two presented faces is the same or different while ignoring the other half (e.g. the lower face). Participants usually have difficulty in selectively attending to just one half of the faces, indicating the configural processing mechanisms are causing faces to be processed as a whole (Behrmann et al., 2015). However, if the faces are misaligned, the two individual identities are easier to distinguish. This is taken as evidence that subjects "fuse" the upper and lower half of a face to perceive it as a whole, thus allowing them to confuse the combined faces as a new identity (Rossion, 2013). The combination of features in the context of a whole face induces the visual system to utilize holistic face processing mechanisms.  Another paradigm that results in a classical behavioural response supporting configural processing strategies is the Face Inversion Effect (Van Belle et al., 2010; Yin, 1969) in which an individual's ability to perceive the identity of a face is significantly impaired when viewing inverted faces compared to upright. Inverted faces are considered to be good controls for upright faces as they contain identical information in terms of physical image properties, such as contrast, stimuli complexity, and spatial frequency content. In this atypical orientation, the   19 familiar first-order configuration is flipped and prevents initiating of the face processing pathway and subsequent configural analysis (Valentine, 1988). One example for this effect is called the Thatcher illusion, where subjects are unable to detect grotesque changes in a face if the image is inverted (Thompson, 1980) yet they are readily apparent once the image is rotated upright. The face-inversion effect is taken as evidence that upright faces are processed using a separate strategy from a more general-purpose processing strategy used for other objects and inverted faces (Farah et al., 1998; Farah, Wilson, Drain, & Tanaka, 1995; X. M. Guo, Oruc, & Barton, 2009; McKone et al., 2007; Moscovitch, Winocur, & Behrmann, 1997).  The Part-Whole Effect is another significant behavioural finding associated with configural processing. In this paradigm, subjects either observe two faces or individual features in isolation (e.g. the nose) and are asked to indicate whether the feature of interest is the same or different. Subjects are superior at discriminating changes in individual features (i.e. the eyes) when they are embedded in a full face rather than when they are shown in isolation. This effect is not observed for other stimuli classes such as non-face objects (Farah et al., 1998; Tanaka & Farah, 1993). Despite more task irrelevant information to analyze in the form of the rest of the face and features, individuals are more adept at recognizing changes in a specific feature if that change is presented within the context of the full face. This is not to imply the individual features parts are not decomposed or analyzed independently, but that there is an advantage to viewing the features in relation to the other features in a human face (Behrmann et al., 2015). This advantage is considered evidence that faces are processed in a holistic "gestalt" manner rather than as the sum of discrete parts.    20  Taken together, the composite face effect, the face inversion effect, and the whole/part advantage all support a configural face processing strategy in which there is a bias towards a holistic, global representation, rather than local features.   Eye gaze is another important aspect of face processing, as it provides information regarding another person’s focus of attention or emotional state and enables the observer to adjust attention accordingly (Hoffman & Haxby, 2000; Kendon, 1967). Attention to the eyes is modulated by early life experiences (Senju & Johnson, 2009; Senju et al., 2015) and gender (Bayliss, di Pellegrino, & Tipper, 2005). Gaze-following is associated with social cognition (Shepherd, 2010) and averted eye-gaze can impede configural face encoding (S. G. Young, Slepian, Wilson, & Hugenberg, 2014).   1.3 Autism Spectrum Disorder  Autism Spectrum Disorder (ASD) is a complex range of neurodevelopmental disorders in which individuals have characteristic impairments in two core areas, (a) social interaction and communication and (b) restricted, repetitive behaviors and interests (American Psychiatric Association, 2013). Initially believed to be a rare disorder, ASD is now estimated to affect one in every 68 children (Christensen et al., 2016). Twin and sibling studies have indicated that ASD is likely genetic in origin (Anney et al., 2010; Charman et al., 2017; Le Couteur et al., 1996; McKernan, Russo, Burnette, & Kates, 2017; Szatmari et al., 2007). The significantly higher co-occurrence of ASD in monozygotic compared to dizygotic twins is evidence of the heritability of this disorder (Rosenberg et al., 2009; Tick, Bolton, Happé, Rutter, & Rijsdijk, 2016). Additionally, rodent models of ASD have reliably induced ASD-like traits in rats following inhibition of certain genes (Servadio, Vanderschuren, & Trezza, 2015). Despite the evidence for   21 the genetic basis of ASD, it has been difficult to pinpoint the precise genes involved. As such, no biological tests have been developed to diagnose it. Of the dozens of genes that have been implicated in ASD, many are believed to play a role in synapse formation or pruning (Giovedí, Corradi, Fassio, & Benfenati, 2014).  This complex disorder is multifactorial in nature, most likely with a contribution from multiple genetic susceptibilities and environmental factors acting upon them (Rosenberg et al., 2009). Individuals present with a range of symptoms, impacting a range of behaviours and aspects related to communication and social functioning. The heterogeneity within this population makes it difficult to pinpoint the cause of ASD and further suggests ASD is caused by many factors (Amaral, Schumann, & Nordahl). Leo Kanner first described autism as a separate syndrome in 1943 with brief case history reports of eleven children. These individuals all had common traits such as preferring social isolation or "extreme aloneness," marked delays or abnormalities in language acquisition and usage, executive function, emotional detachment, and an "insistence on sameness" that was often observed as monotonous and repetitive interests, abnormal verbal utterances, and actions (Kanner, 1943). One year later, Austrian doctor Hans Asperger published case histories of a group of children who had behaviour resembling some, but not all aspects of those described by Kanner. While the children Asperger described also preferred solitude and had specific interests in restricted fields, they did not have delays in language acquisition and had above average intelligence (Wolff, 1991).  These are the earliest descriptions of what would later be known as a “spectrum” disorder, meaning there is a wide degree of variation in autism affects individuals. These variations have led to the distinction of “high-functioning” and “low-functioning” autism.   22 Individuals who are high-functioning tend to have average or above average intelligence, intact verbal abilities, and the ability to function independently with little support. Low-functioning individual on the spectrum, however, often have cognitive deficits, lack verbal communication skills or may be entirely nonverbal, have difficulty managing their behaviour, and require support for daily activities.   1.3.1 Diagnosis of ASD  Clinical diagnosis of ASD requires evaluation by observation of the individual, most often a young child, and interviews with family and educators. It is standard practice for clinical psychologists to use the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) to define and classify specific behaviors and communication delays associated with this developmental disability. It provides diagnostic classification, criteria sets, and descriptive texts to assist mental health professionals in evaluating individuals (American Psychiatric Association, 2013). This most recently revised manual's task force made the decision to remove previous subtypes of Asperger's and Pervasive Developmental Disorder used in DSM-IV and instead combine these under the umbrella diagnosis of ASD. The argument was not that Autism is a homogenous mixture; heterogeneity and scales of severity are one of the hallmarks of ASD. After reviewing multiple studies, the task force argued that there was no clear evidence of where one subtype ended and another began (Volkmar & McPartland, 2014). Instead, they proposed using the overarching term of ASD and including three levels of severity in two core areas: (a) social and communication issues and (b) restricted and repetitive behaviors and interests. The levels ranged from 1: "requiring support" to 3: "requiring very significant support." Critics argued that the new classifications would exclude some children previously included for   23 diagnosis; potentially jeopardizing their access to support services (McPartland, Reichow, & Volkmar, 2012). As with any change in classification systems, these new definitions make it challenging to compare current results to studies using older manuals. For the sake of consistency, ASD will be used in place of earlier diagnostic categories when discussing research prior to 2013.   The Autism Diagnostic Observation Schedule (ADOS) is currently utilized by clinical psychologists to assist with diagnosis (Lord, Risi, Lambrecht, Cook, Leventhal, DiLavore, Pickles, et al., 2000). It is a semi-structured assessment of communication, social interaction, and play (or imagination while using given materials) that allows the clinician to observe behavior responses to a set of standardized activities. There are four modules to allow assessment of individuals with verbal abilities ranging from nonverbal to highly verbal (Western Psychological Services, 2001). It is considered the gold standard for researchers as the roughly 45-minute assessment allows for confirmation of diagnosis without the time-intensive process of going through the formal diagnosis with interviews of caretakers and extensive observations.   1.3.2 Measures of intelligence, autism severity, and social competence  The Wechsler Abbreviated Scale of Intelligence (WASI-II) is a general intelligence quotient (IQ) test standardized for ages 6-90. This brief measure of cognitive ability allows Full Scale IQ (FSIQ), verbal IQ, and nonverbal IQ scores (Wechsler & Hsiao-pin, 2011). There are four subtests included in the WASI-II: Vocabulary, Block Design, Similarities, and Matrix Reasoning. Matrix reasoning and block design tasks comprise the nonverbal IQ score while the vocabulary and similarities subtests make up the verbal IQ score. The matrix reasoning and   24 vocabulary tasks can be combined to make a shorter FSIQ-2 score (Wechsler & Hsiao-pin, 2011) when experimental needs do not require more information.    The Autism Spectrum Quotient (AQ) is a brief, self-administered scale that measures traits associated with ASD in the typical population (Baron-Cohen, Wheelwright, Skinner, Martin, & Clubley, 2001). It is a screening tool for individuals who may warrant further evaluation by clinicians. It consists of 50 questions, with 10 questions for each of five categories associated with ASD: social skills, attention switching, attention to detail, communication, and imagination. Questions earn a score of 1 point if the participant "agrees" or "disagrees" with the statement. Average scores for individuals with ASD are 35.8, while the average score for controls is 16.4 (Baron-Cohen et al., 2001). Male controls in the mathematics and sciences average a score of 19.3, making a score of 20 an appropriate cut-off point when designing experiments. It is the point of greatest separation between controls and individuals with ASD, while still allowing for those higher scoring controls.   The Multidimensional Social Competence Scale (MSCS) is a 77-item parent rating scale used to assess individual differences in social competence in adolescents with ASD (Yager & Iarocci, 2013). The expression and severity of social competence in ASD varies. This 77-item questionnaire assesses seven distinct domains of social competence: social motivation, social inferencing, demonstrating empathic concern, social knowledge, verbal conversation skills, nonverbal sending skills, and emotion regulation (Yager & Iarocci, 2013). Social Motivation refers to how much interest and enjoyment one has when interacting with others while Social Inferencing is the ability to detect social cues. Individuals who are able to recognize when someone is hurt or upset and respond in a kind manner are said to demonstrate Empathic Concern. Social Knowledge assesses knowledge of the social rules and norms in specific social   25 situations. Verbal Conversation Skills require an ability to start and maintain a reciprocal conversation and end it appropriately. Nonverbal Sending Skills refer to the ability to understand the "sending" of social communication cues like pointing, gestures, eye contact, and facial expressions. The Emotion Regulation subdomain assesses the ability to avoid extreme reactions to frustrating circumstances. The MSCS is not a diagnostic tool and does not have a specific cut-off score qualifying someone as having ASD. However, individuals with ASD will tend to score lower than their typically-developing peers (Yager & Iarocci, 2013).    1.3.3 Sensory issues in ASD  While not always considered a diagnostic criterion, sensory-perceptual abnormalities have long been associated with ASD (for a review of the early literature, see O'Neill & Jones, 1997). Sensory overload and sensory processing difficulties have been described (Iarocci & McDonald, 2006; Manning, Tibber, Charman, Dakin, & Pellicano, 2015; Simmons et al., 2009). Unusual responses to sensory stimuli are present early in development (Wiggins, Robins, Bakeman, & Adamson, 2009). Self-reports have described hyper- or hypo-sensitivity across all sensory modalities (Gowen & Hamilton, 2013; Leekam, Nieto, Libby, Wing, & Gould, 2007; Simmons et al., 2009). Examples of hyper-sensitive responses to stimuli include focusing on tiny dust particles or discomfort around loud sounds. Some individuals with ASD show hypo-sensitivity to visual stimuli such as a preoccupation with reflections and/or moving objects directly in front of the eyes (Leekam et al., 2007). Atypical processing of visual stimuli has been indicated as a potential source for some of the social impairments in this population.    26 1.4 Vision in ASD Studies using a variety of perceptual tasks have demonstrated atypical visual functioning in individuals with ASD when compared to controls. Results have indicated that multiple aspects of visual functioning are altered in ASD, including perception of texture (Bertone, Mottron, Jelenic, & Faubert, 2005; Rivest, Jemel, Bertone, McKerral, & Mottron, 2013), motion perception (Milne et al., 2002; Schauder, Park, Tadin, & Bennetto, 2017), colour discrimination (Franklin et al., 2010; Zachi et al., 2017), susceptibility to visual illusions (Happe, 1996; Ishida, Kamio, & Nakamizo, 2009), biological motion (Blake, Turner, Smoski, Pozdol, & Stone, 2003; Puglia & Morris, 2017), and face perception (Adolphs, Sears, & Piven, 2001; Dawson, Webb, & McPartland, 2005; Pallett, Cohen, & Dobkins, 2013; Weigelt, Koldewyn, & Kanwisher, 2012). Despite the range of visual processing differences in ASD, it has been challenging to identify the source of these abnormalities in visual perception.   The differences found in visual processing in ASD have led to the creation of a number of experimental paradigms to explore these patterns of altered performance (Scherf, Luna, Kimchi, Minshew, & Behrmann, 2008; Shah & Frith, 1983). The results from these tasks have been used as evidence to develop various visuo-cognitive theories and models to provide possible explanations for the findings (Happe, 1996; Happe & Frith, 2006; Mottron, Dawson, Soulières, Hubert, & Burack, 2006). Results from optometric studies of children with ASD have indicated increased rates of refractive error and strabismus compared to the general population (Black, McCarus, Collins, & Jensen, 2013; Ikeda, Davitt, Ultmann, Maxim, & Cruz, 2013; Scharre & Creedon, 1992). Processing of low-level visual stimuli, including aspects of spatial frequency (Guy, Mottron, Berthiaume, & Bertone, 2016; Kéïta, Guy, Berthiaume, Mottron, & Bertone, 2014), orientation (Bertone et al., 2005; Dickinson, Bruyns-Haylett, Smith, Jones, &   27 Milne, 2016; Dickinson, Jones, & Milne, 2014), and colour perception (Franklin, Sowden, Burley, Notman, & Alder, 2008; Franklin et al., 2010; Heaton, Ludlow, & Roberson, 2008), have also been indicated as a potential source for the differences in performance in various experimental paradigms. Performance on a variety of face processing tasks have indicated altered perception of face identity and expression for many individuals with ASD (Bailey, Braeutigam, Jousmäki, & Swithenby, 2005; Barton, Cherkasova, Hefter, et al., 2004; Behrmann, Avidan, et al., 2006). The face perception differences have led to the formulation of hypotheses that relate face processing skills to social competence difficulties in ASD (Adolphs et al., 2001; Baron-Cohen et al., 2000; Dawson, Webb, & McPartland, 2005; Schultz, 2005). Each of these aspects of visual processing in ASD will be discussed in more detail in the following sections.  1.4.1 Visuo-cognitive theories of visual processing in ASD Multiple studies have indicated altered processing of details via a bias towards local features compared to global ones (Falter, Elliott, & Bailey, 2012; Falter, Grant, & Davis, 2010) (for a meta-analysis please see Van der Hallen, Evers, Brewaeys, Van den Noortgate, & Wagemans, 2015). One of the earliest studies to demonstrate a bias towards details used the Children's Embedded Figures Test (CEFT) with children with ASD (Shah & Frith, 1983). This task asks subjects to find a particular simple shape (e.g. a triangle) within a larger, more complex image. Children with ASD have increased accuracy and speed in the CEFT compared to controls (Happe, 1999; Jolliffe & Baron-Cohen, 1997), indicating enhanced processing of local features.   Another experimental paradigm used to study the bias towards local processing utilizes Navon letters. This set of stimuli consists of large letters comprised of many smaller letters (e.g. a large letter "s" is made of smaller images of the letter "h"). They were developed to examine   28 the concept of "gist", where one extracts the essence of a visual scene on a global level (Navon, 1977). The overall meaning of a scene (e.g. a forest) is identified rather than focusing on the details (e.g. individual trees). When Navon letters were tested in adults with ASD, subjects were faster than neurotypical controls when identifying local level components compared to global level (Scherf et al., 2008). In a task similar to the Navon letters, individuals with ASD were asked to identify either the larger or smaller components of stimuli consisting of a large shape (e.g. circle) made of smaller shapes (e.g. diamonds). In a study of subjects with ASD ranging from children to adults, global shape processing was found to be impaired across development (Scherf et al., 2008). Children with ASD showed a precedence for attending to local features when presented with a divided attention task, but are capable of voluntary selective attention if asked to do a visual task that requires global processing (Plaisted, Swettenham, & Rees, 1999). A later study offers a potential explanation, suggesting that individuals with ASD may be less sensitive to structural global bias, while having an implicit task bias towards local level processing compared to controls (Iarocci, Burack, Shore, Mottron, & Enns, 2006).  Some experimental paradigms have found that individuals with ASD have superior processing of visual stimuli compared to controls. Several studies have found faster reaction times during visual search tasks (Gliga et al., 2015; O'Riordan, 2004; O'Riordan, Plaisted, Driver, & Baron-Cohen, 2001; Plaisted, O'Riordan, & Baron-Cohen, 1998a; Plaisted et al., 1999), but not all (Iarocci & Armstrong, 2014). Additionally, discrimination between highly similar stimuli is improved for individuals with ASD when compared to their neurotypical counterparts (Plaisted, O'Riordan, & Baron‐Cohen, 1998b; Plaisted, Saksida, Alcantara, & Weisblatt, 2003). Superior performance in such tasks has been taken as evidence that individuals   29 with ASD have enhanced processing for details and has informed the development of a number of cognitive models of ASD. The weak coherence (WC) model was formulated based on a number of findings, including detail-based bias towards local features, difficulty identifying the gist of a stimuli, and being less susceptible to visual illusions (Happe, 1996, 1999). This account was originally called "weak central coherence" and suggested that a core deficit in global processing was driving social impairments. In studies that followed the proposal of the WC model, evidence suggested that WC could be the source of why children with ASD had abnormal global processing in a global dot motion task (Pellicano, Gibson, Maybery, Durkin, & Badcock, 2005). Additional studies countered the assertion that global deficits were obligatory. Results from one study suggested that perceptual mechanisms increase the focus on individual features, but not at the expense of the ability to integrate features for tasks requiring global processing (Plaisted et al., 2003). These criticisms led to the development of an updated WC model that no longer included global deficits as a primary feature of ASD, but instead suggested global processing deficits are a secondary outcome of a strong bias towards localized features in visual processing tasks (Happe & Frith, 2006; Simmons et al., 2009).   Enhanced Perceptual Functioning (EPF) is another influential account proposed to explain the visual processing patterns in ASD. While there are some overlaps between the WC and EPF, they approach the bias for local details from different directions. When looking at evidence on both sides, the question becomes whether differences in low-level visual processes or behavioural biases towards certain information are driving atypicalities in visual function in ASD (Dakin & Frith, 2005). Whereas the WC model proposes a global processing deficit for individuals with ASD, the EPF account approaches it as a consequence of superior processing of   30 “simple visual material” (Mottron et al., 2006; Simmons et al., 2009). Studies have found evidence of superior local perception and an enhanced role of early cortical processing for visual functioning in adolescents and adults with ASD (Caron, Mottron, Berthiaume, & Dawson, 2006; Latham, Chung, Allen, Tavassoli, & Baron-Cohen, 2013). While there is a bias in the attentional system for local features (Iarocci et al., 2006) or sensitivity for high spatial frequency information (Kéïta et al., 2014), these biases are not mandatory. As global processing is not impaired (Van der Hallen et al., 2015), the EPF model suggests superior processing, not a deficit, is responsible for the response patterns of individuals with ASD in a variety of visual tasks.  As with many early models of disorders, the EPF has been the subject of criticism in the literature. The most common criticism of the EPF model is that it draws support for some of its principles from savants (Mottron & Belleville, 1993; Mottron et al., 2006). The term autistic savant refers to a person with a developmental disability who demonstrates superior performance, or a “splinter” ability in one specific context (e.g. piano) or knowledge in a limited field (e.g. astrophysics). Generally, these relatively rare occurrences of savant-abilities only occur in a subset of people with ASD. Another criticism of EPF comes from studies that did not find superior processing of local details, a hallmark of the EPF, for individuals with ASD in a variety of tasks (Muth, Honekopp, & Falter, 2014; Scherf et al., 2008). These results suggest that enhanced perception for local details may only occur for a subset of individuals on the spectrum, instead of being ubiquitous as suggested by EPF.   1.4.2 Optometric concerns in ASD As previously discussed, refractive errors result when light is unable to focus directly on the retina. The first major study of refractive errors in children with ASD found that the incidence of   31 significant refractive errors was 44% (Scharre & Creedon, 1992), much higher than the roughly twelve percent of neurotypical children who require prescription lenses (Kawuma & Mayeku, 2002). An important caveat to the findings of this study is that the children tested were diagnosed under the narrower DSM-III guidelines. The results, therefore, may not generalize to the rest of the ASD population given current diagnostic criteria (Simmons et al., 2009). More recently, retrospective chart reviews found significant refractive errors for 29% (Ikeda et al., 2013), 27% (Black et al., 2013), and 22% (Kabatas et al., 2015) of children and adolescents with ASD. A prospective study in Nigeria screened children with ASD for optometric issues and reported that 33% had significant refractive errors (Ezegwui et al., 2014). A major limitation of these findings are the large number of subjects with ASD who are unable to cooperate with optometric exams (Breidenstein, 2015), meaning significant refractive errors may be under-diagnosed in the ASD population as a whole.   In addition to refractive errors, higher incidence rates of strabismus, or improper alignment of the eyes, have also been reported in ASD. This condition often requires surgical correction, and if left untreated can lead to amblyopia, or long-term impairments in binocular vision (Gunton, Wasserman, & DeBenedictis, 2015). From the earliest studies of optometric issues in children with ASD, reported incidence rates of strabismus have been varied, with 20.6% for children ranging from 2-11 years (Scharre & Creedon, 1992), or up to 50% for 7-9 year olds with ASD (Kaplan, Rimland, & Edelson, 1999). Both rates are considerably higher than the 3.7% estimates for typically-developing children in the same age range (Roberts & Rowland, 1978). Retrospective chart reviews also reveal a great deal of variability in the incidence of strabismus for the groups tested, with 21% of children ranging from 6 months to 15 years old (Ikeda et al., 2013), 41% of children ranging from 2-20 years (Black et al., 2013), and   32 8.6% of Turkish children ranging from 18 months–17 years of age (Kabatas et al., 2015) for children with ASD meeting the criteria for the condition. There are multiple factors that may be influencing the different results between reports, including disparities in the age of assessment, a lack of reporting race, and different data collection methods. Despite the widespread variability across reports, even the most conservative estimates are still considerably higher for children with ASD compared to the general population. Recent studies of strabismus incidence in the non-clinical population suggest that rates range from 3.6% of Asian and 3.3% of Caucasian children aged 6-72 months (McKean-Cowdin et al., 2013), to 5.7% of children aged 36-72 months in Eastern China (Chen et al., 2016), to 3.3% in Caucasian and 2.1% in African American children aged 6 through 71 months (Friedman et al., 2009). Thus, while the rates for strabismus in ASD are varied between reports, the evidence supports a general result of higher incidence for individuals with ASD compared to the general population. Despite these findings of increased refractive errors and strabismus, it was suggested that individuals with ASD may have significantly enhanced visual acuity. While refractive errors can impact visual acuity by creating errors in how the light is focused on the retina, the health of the retina and the processing of visual stimuli in the cortex also influence the clarity of vision. If one corrects refractive errors using prescription lenses, it is possible that differences in the retina or cortical pathways may be influencing visual acuity in ASD.  One study found enhanced visual acuity in adults with ASD averaged 20/07, and suggested superior visual processing began at the retina (E. Ashwin, Ashwin, Rhydderch, Howells, & Baron-Cohen, 2009). Given the neural resolution provided via cone density in the retina, the best possible visual acuity for humans was believed to be roughly 20/08 (Applegate, 2000). Superior visual acuity would suggest individuals with ASD had increased photoreceptor   33 density or altered neuronal pathways. Follow up analyses contested the initial report of superior visual acuity (Falkmer et al., 2011; Kéïta, Mottron, & Bertone, 2010). A later study found that the original report had a methodological issue and that individuals with ASD did in fact have normal visual acuity (Tavassoli, Latham, Bach, Dakin, & Baron-Cohen, 2011). While there are higher incidences of refractive errors and strabismus for individuals with ASD, these issues alone do not explain the atypicalities in visual perception associated with this disorder. Subsequently, differences in visual processing must be sought upstream from the retina.  1.4.3 Low-level visual perception in ASD Early visual areas such as V1 and V2 carry out spatial frequency and orientation analyses, similar to a Fourier decomposition. These regions are the next possible source for altered visual perception in ASD and have been examined in numerous studies. One proposal of the EPF model is that the superior processing of "simple visual material" associated with this population is a result of overfunctioning of the earliest visual areas (Caron et al., 2006). If individuals with ASD have atypical processing for the most basic components of a visual stimulus, this suggests an early source along the visual pathway. Conversely, if low-level vision is typical in ASD, this suggests that the source for altered perception occurs later in the visual pathway. The general trend of the literature suggests that low-level perception is intact and unremarkable in ASD, although contradictory results have also been reported.  Contrast thresholds were measured in adults with ASD across five different spatial frequencies, and the ASD group demonstrated they were able to detect low and high frequency gratings comparably to control participants (Behrmann, Avidan, et al., 2006). These null results are supported by a study of adolescents with ASD, which found that contrast sensitivity was   34 unremarkable across the seven spatial frequencies tested (Koh, Milne, & Dobkins, 2010). One fMRI study found visual field representation of central to periphery were normally distributed, indicating early sensory visual areas are typical in ASD (Hadjikhani et al., 2004). The results from each of these studies suggest that altered processing of spatial frequency is not the source of the visual atypicalities associated with ASD.  In direct contradiction of the null results for spatial frequency processing in ASD, some studies have found that contrast sensitivity processing for certain spatial frequencies is altered in ASD. Contrast sensitivity thresholds for sinusoidal gratings were measured across a range of spatial frequencies in adolescents and adults with ASD. Subjects with ASD had lower thresholds for luminance-defined, high spatial frequency gratings (8 cpd) compared to controls (Kéïta et al., 2014). Children and adolescents ranging from 6-16 years of age were assessed for contrast sensitivity of gratings at five distinct spatial frequencies (Guy et al., 2016). Across the developmental time course, the ASD group demonstrated a lower level of sensitivity for mid spatial frequencies (2 and 4 cpd), an effect that was most pronounced in the younger participants. These results failed to reach significance for the older participants tested, suggesting that processing abilities for mid spatial frequency ranges continue to develop, albeit more slowly, for individuals with ASD. Additionally, an electroencephalography (EEG) study looking at visually evoked potentials in children with ASD found abnormal processing of high-spatial frequency gratings (Boeschoten, Kenemans, Engeland, & Kemner, 2007). These findings are contrasted by an earlier EEG study which found VEPs were unremarkable for low and high spatial frequencies, but in adults with ASD, mid-range spatial frequencies elicited a neural processing signature similar to the characteristic one associated with high-range spatial frequencies (Jemel, Mimeault,   35 Saint-Amour, Hosein, & Mottron, 2010). Together, these studies suggest atypical development in processing spatial frequency information for individuals with ASD.  The second fundamental aspect of visual function performed in areas V1 and V2, orientation processing of visual stimuli, has also been examined in individuals with ASD. Multiple studies have also reported null results for altered visual perception of orientation for low-level stimuli. A recent study that presented a variety of motion and orientation tasks to children ages 6-14 with ASD did not find differences in processing for the orientation tasks (Manning, Tibber, & Dakin, 2017). Similarly, no signs of enhanced perception were found in an orientation discrimination task for adults with ASD (Freyberg, Robertson, & Baron-Cohen, 2016). There were no differences found in V1 during a fMRI study of adults with ASD, indicating that orientation processing in that region is likely typical (Schwarzkopf, Anderson, de Haas, White, & Rees, 2014).   As with the spatial frequency results, other studies of orientation processing in ASD have presented conflicting results. Discrimination thresholds around vertical and one oblique orientation were measured for a non-clinical sample of adults and compared to AQ scores (Dickinson et al., 2014). Higher AQ scores were related to lower thresholds (better performance) on the oblique orientations. Additionally, orientation discrimination was found to be superior in adults with ASD, directly contradicting earlier findings (Dickinson et al., 2016). In one orientation-identification task, subjects were presented with a vertical or horizontal sinusoidal grating in noise and asked to indicate the orientation of the stimuli. Contrast thresholds for orientation identification were measured in adolescents and adults with ASD and it was found that subjects with ASD had enhanced processing for the least complex stimuli (Bertone et al., 2005).   36   When the results for and against altered perception of low-level spatial frequency and orientation components of visual processing, are analyzed together, it becomes difficult to identify the source of the discrepancies. It is possible that methodological choices may be driving at least some of the contradictory results. Some experiments have used sinusoidal gratings as stimuli, which can introduce high spatial frequencies when displayed on a computer screen. These high-spatial-frequency components may have enabled the enhanced performance found by Dickinson et al. (2014). Additionally, some experiments that use simple stimuli may have relatively complex tasks, which would recruit upstream cortical regions. Thus, the claims of superior processing may actually be a reflection of the complexity of the task itself. For instance, when using Gabor patches as the stimuli, higher AQ scores were correlated with performance on a visual search task, but not on an orientation discrimination task (Brock, Xu, & Brooks, 2011). Additionally, given the heterogeneous characteristics of the ASD population as a whole, it is possible that “sampling” differences in participant selection for different studies are driving some of the disparate results. This suggests careful attention must be paid to experiment design, participant selection, and instructions to the participant.   With such contradictory results, a gap remains in the literature as to whether low-level visual processes for spatial frequency and orientation discrimination in ASD are atypical or not. This debate can be addressed with a carefully characterized sample of individuals with ASD and rigorous psychophysical studies utilizing carefully selected stimuli and the most basic of visual tasks. By eliminating the confounds that plague these controversial studies, it will be possible to gain a better understanding of the state of low-level visual processing in ASD. Chapter 3 of this thesis is devoted to this particular topic, the results of which were published (Shafai, Armstrong, Iarocci, & Oruc, 2015).   37  Colour perception also begins as a low-level visual process initiated in the earliest visual areas (Roe et al., 2012). In the domain of colour vision, studies of individuals with ASD have revealed altered perception in a variety of tasks. Rates of colour vision loss may be elevated in ASD (Zachi et al., 2017), which may occur either at the retinal level and/or reflect reduced cortical integration. The latter possibility is supported by findings suggesting impaired colour pathway activity in ASD (T. Fujita, Yamasaki, Kamio, Hirose, & Tobimatsu, 2011). It has been demonstrated that colour discrimination, detection, and extraction of similarity of colours are impaired for children with ASD (Franklin et al., 2008; Franklin et al., 2010; Heaton et al., 2008; Zachi et al., 2017), although this may be influenced by age and nonverbal IQ (Cranwell, Pearce, Loveridge, & Hurlbert, 2015). When labeling demands are not included in tasks, colour memory was enhanced for children with ASD in one study (Heaton et al., 2008), but not for another (Franklin et al., 2008). A review by Bakroon and Lakshminarayanan (2016) argues that fMRI studies of chromatic discrimination in ASD would enable individuals to assess the possibility of a neural basis to chromatic sensitivity. Indeed, an fMRI study by (Vogan et al., 2014) utilized a colour matching task and found that increasing difficulty of the task caused increased reliance on posterior brain regions while frontal and parietal lobes were more heavily activated for controls.  1.4.4 Face processing in ASD As social deficits are a hallmark of ASD, visual perception of the human face has been a focus of research with this population. Despite a multitude of face processing studies in ASD, a consensus has not been reached as to whether there is an impairment, and if so, to which aspect of face processing. Multiple studies have suggested that some individuals with ASD struggle with identifying faces, face expressions, gaze, and eye contact (Bailey et al., 2005; Barton, Hefter,   38 Cherkasova, & Manoach, 2007; Churches, Baron-Cohen, & Ring, 2012; Falck-Ytter, 2008; Pallett et al., 2013; Simmons et al., 2009; S. Wallace, Coleman, & Bailey, 2008b; Wolf et al., 2008).  It has been suggested that holistic processing is impaired in ASD and that these impairments drive at least part of the face processing differences found in this population. In a task comparing face and object processing in adults with ASD, stimuli were shown quickly (40 msec) to test holistic processing and found that discrimination of faces was impaired while processing of cars was typical (S. Wallace, Coleman, & Bailey, 2008a). The second task in this study altered the second-order configural properties of face and house stimuli and found that participants with ASD were sensitive to changes in house stimuli but not faces. The authors argue that these results point to a deficit in holistic processing that is specific to faces (S. Wallace et al., 2008a).  An important question in the literature centers on whether or not there is a causal relationship between difficulties with faces and social competence (Chevallier, Kohls, Troiani, Brodkin, & Schultz, 2012; Dawson, Webb, Wijsman, et al., 2005; Schultz, 2005). The social motivation hypothesis suggests that reduced social interest may be related to problems with face perception in ASD. According to this model, if a person has reduced interest in social engagement, then the typical development of face processing skills can be negatively impacted. Alternatively, impairments with face perception may lead to reduced success in social relations.   An illuminating perspective for this argument comes not from studies with ASD, but from studies of individuals with (congenital) developmental prosopagnosia (DP). Unlike those with acquired prosopagnosia, individuals with DP never had intact face identification skills. A face can be classified as a face, but the identity of the face is not readily distinguishable to   39 individuals with DP. Throughout childhood and adolescence, individuals report that they compensate by using similar strategies as individuals with acquired prosopagnosia such as a person's voice or body type (Kracke, 1994). Additionally, individuals with DP can struggle with social isolation and peer relationships (Barton, Cherkasova, Press, Intriligator, & O’Connor, 2003). The similarities in consequences for poor face recognition abilities or poor social competence have led researchers to examine the relationship between DP and ASD.  It is possible that an initial difficulty with processing faces is compounded throughout development. Children and adolescents with ASD are slightly impaired with face perception compared to their neurotypical counterparts, while adults with ASD are far worse at faces, potentially as a result of a lack of improvement associated with development of normal face processing (O’Hearn, Schroer, Minshew, & Luna, 2010). Thus, DP is not directly influencing symptoms related to ASD or vice versa, but dysfunction of face processing for adults in both groups suggests the two conditions share at least some common mechanisms for the altered perception of faces.  A later study looked at social cognition in twelve individuals with DP and found that for eleven of them, social cognition was normal (Duchaine, Murray, Turner, White, & Garrido, 2009). This suggests that it is possible to develop typical social skills despite lacking access to face identity information throughout development.   Another study chose to approach the issue from the ASD side and instead assessed whether individuals with ASD showed signs of prosopagnosia. In this study, individuals with social developmental disorders were assessed for face recognition abilities and compared to prosopagnosic patients (Barton, Cherkasova, Hefter, et al., 2004). Cluster analysis revealed that eight of the twenty-four patients with ASD had normal, intact face processing while the rest were   40 impaired in face recognition, but less so than individuals with prosopagnosia. A later study with the same group of participants found that the ability to detect changes in feature position or colours for upright and inverted faces was again split into two subgroups of participants with ASD (Barton et al., 2007). One group performed as well as controls in a famous face task and had slightly reduced accuracy in the configuration and feature tasks, while another group was severely impaired in all tasks. Together, these studies are taken as evidence that while face processing is impaired for some individuals with ASD, it is not an innate characteristic of the disorder.  The social motivation hypothesis suggests that reduced attention to faces may drive, at least in part, the impaired performance in face processing tasks associated with this population. Some claim this impairment may be due to attention disparities between participants with ASD and neurotypical controls when viewing images of a social and non-social nature (Sasson, Elison, Turner-Brown, Dichter, & Bodfish, 2010). Eye tracking studies have allowed researchers to assess attention when viewing faces and objects in a variety of paradigms. These studies are especially useful when assessing attention during early development. Eighteen high-risk infants between the ages of 4-7 months were assessed as part of an eye-tracking study that measured the preference for social stimuli via looking time. When presented with the mother's face and a stranger's face, all infants showed a preference looking at the mother. However, when showed two images of strangers in which one image had a person with averted eyes and the other with direct gaze, the high-risk infants did not show a preference for the direct gaze images in a similar manner to low-risk infants (Nele, Ellen, Petra, & Herbert, 2015). This indicates that differences in social attention begin within the first months of life for high-risk infants.    41 A longitudinal study assessed eye fixations for low and high-risk infants with ASD and found that for those later diagnosed, attention to the eyes began normally but declined between the 2-6 months of age (W. Jones & Klin, 2013). This marked decrease in percent of visual fixation time to eyes continued through the first 24 months of life, where it stabilizes at roughly half the amount of the low-risk infants. Another longitudinal study also used eye-tracking to measure eye gaze location and looking time at 3, 6, 9, and 12 months of age for infants with low and high risk of ASD. The children were later assessed for ASD around their third birthday, and those who had been considered high risk demonstrated a reduced preference for the eye region with age. Children who were later diagnosed with ASD demonstrated a clear preference for mouth fixations throughout their first year of development (Rutherford, Walsh, & Lee, 2015). Together, these longitudinal eye-tracking studies provide evidence that attention to faces, especially in the eye region, in the earliest months of development is altered in ASD.  It is possible that a developing infant will focus less towards faces if his or her parents struggle with social settings, independently of any ASD diagnosis. In a study of 223 typically developing infants and the biological parent that spent the largest amount of time with the infant, those parents who reported higher levels of social discomfort and avoidance had infants who spent less time attending to faces than objects. These infants also showed smaller N400 amplitudes when viewing faces compared to the response amplitudes of infants whose parents were more comfortable in social situations (E. J. H. Jones, Venema, Earl, Lowy, & Webb, 2017). It is possible that the level of interest in faces is influenced by both genetic mechanisms and environmental influences, which would indicate that each of these factors play a role in the development of face processing skills.    42  Indeed, in a study comparing face processing abilities in parent or adult siblings of individuals with ASD with individuals with ASD, performance in face discrimination, face expression recognition, and eye-gaze direction tasks were all impaired for the relatives of individuals with ASD, but not to the same degree as the participants with ASD (S. Wallace, Sebastian, Pellicano, Parr, & Bailey, 2010). The authors argue that these results suggest a genetic component to face processing abilities.  Eye tracking studies continue to provide useful information later in development when assessing children, adolescents, and adults with confirmed diagnoses. Children with ASD were impaired in face recognition, but showed typical performance when discriminating between buildings (Boucher & Lewis, 1992). Importantly, this study included analysis of looking behaviour and found that the impaired face recognition in ASD was not correlated with diminished attention to faces as a class of stimuli.  Thirteen children with ASD were asked to passively view various scenes while wearing an eye tracker and then asked to complete a matching task for faces and objects (Wilson, Brock, & Palermo, 2010). The eye-tracking results revealed that children with ASD spend equal amounts of time attending to objects and people, whereas controls spent a larger proportion of their time looking at people. There was a significant correlation between looking time and performance in the matching task. Children on the spectrum who performed best on a recognition task were more likely to attend to people first in the visual scenes.  Another eye tracking study looked at visual attention when passively viewing images of upright and inverted human faces, monkey faces, and shapes (McPartland, Webb, Keehn, & Dawson, 2011). While adolescents with ASD demonstrated similar patterns of visual attention to control participants in that visual fixations tended to focus on the upper regions of visual stimuli,   43 they tended to look longer at the upper regions than neurotypical participants. Despite spending more time attending to the upper part of face stimuli, individuals with ASD were less accurate on a computerized face recognition task. Worse performance on the face recognition task was associated with lower levels of social adaptive functioning measures obtained via parent reports.   Attentional bias for faces compared to cars and houses was measured in adults with and without ASD using a visual cuing paradigm (D. J. Moore, Heavey, & Reidy, 2012). A face-bias, or faster response times when cued with a face compared to an object, was found in controls but was not present for the ASD participants. This study provides evidence of differences early on in cued attentional response to faces as a class of visual stimuli for individuals with ASD.   Given the wide interest in the relationship between attention and face processing in ASD, reviews and meta-analyses have made it possible to assess some general trends in the literature. It has been suggested that individuals with ASD do not attend to faces as much as objects, especially in the eye region (Simmons et al., 2009). However, another review challenged the aversion to the eye region, and instead suggested that individuals with ASD attend to everything in the face except for direction of eye gaze (Guillon, Hadjikhani, Baduel, & Rogé, 2014). Individuals with ASD tend to perform worse in visual experiments with more social content, e.g. the number of people in an image is greater than one (Chita-Tegmark, 2016b). Additionally, a meta-analyses of eye-tracking studies found atypical allocation for attention in individuals with ASD; with reduced attention to face stimuli and increased attention to the body and non-social elements of visual scenes (Chita-Tegmark, 2016a). Taken together, these reviews and meta-analyses suggest that individuals with ASD have altered attention to socially-relevant visual information.    44 1.4.4.1 Brain networks of face processing in ASD  As previously described in section 1.2.2, the face core network consists of the OFA, FFA, and pSTS. All of these regions, and the patterns of activation between them, have been the focus of a variety of activation studies of individuals with ASD.   The results of an early magnetoencephalography (MEG) study of adults with ASD suggest that the processing of human faces is partially located in regions outside the fusiform gyrus (Bailey et al., 2005). In line with these early findings, a recent study of children with ASD suggests that they may be compensating for difficulties with face recognition by recruiting the prefrontal cortex to assist the fusiform gyrus during face processing tasks (Herrington, Riley, Grupe, & Schultz, 2015). Another study found that face processing elicits less of a response in the right FFA for children with ASD compared to typically developing children, with changes through development that suggest altered connectivity between the fusiform system, the amygdala, and the frontal cortex (Joseph et al., 2015).   Recent trends in the literature indicate that patterns of functional connectivity, or the temporal activation patterns between distinct brain regions, have reduced activation across different regions of interest (ROIs) related to face processing for individuals with ASD. Adolescents and adults with ASD showed activation patterns similar to those employed during feature-based processing strategies, most commonly associated with perception of non-face objects (Schultz et al., 2000). Reduced activation of category-selective visual cortex was the most pronounced in the face-selective cortex, including the FFA, OFA, and pSTS for a group of high-functioning adults with ASD (Humphreys, Hasson, Avidan, Minshew, & Behrmann, 2008). A recent quantitative meta-analysis found that reduced patterns of activation have been consistently measured in one cluster located in the left FFA and occipital region according to   45 (Nickl-Jockschat et al., 2015). Activity in the fusiform was increased when children and adolescents with ASD were presented with images of their own personal restricted interests, suggesting that it may be that brain regions used for social functioning are recruited by different visual stimuli than controls (Foss-Feig et al., 2016). A fMRI analysis of heterogeneous group of individuals with ASD suggests that there are a wide range face discrimination abilities, and that individual performance is related to neuronal selectivity in the FFA in response to face stimuli (Jiang et al., 2013). Thus, the seemingly conflicting accounts of FFA activation in ASD could be due to the heterogeneity of abilities characteristic of this disorder.  It has been argued that face perception should be viewed as a process that is supported by cortical and subcortical brain networks distributed across the whole brain (Haxby et al., 2000). From this perspective, instead of looking at the activity within restricted ROIs, Nomi and Uddin (2015) suggest focusing on atypical connectivity within certain brain networks for the source of the social difficulties in ASD. In order to determine whether this atypical activation is associated with symptom severity in ASD, it is necessary to compare behavioural measures with the fMRI findings. fMRI activity was recorded for adults with ASD while being presented with audio, visual, and audiovisual stimuli of faces and voices. Functional connectivity between the temporal voice area and the superior and medial frontal gyrus was reduced in adults with ASD, all regions thought to play a role in interpreting social signals. This reduced connectivity was correlated with increasing symptom severity (Hoffmann, Brück, Kreifelts, Ethofer, & Wildgruber, 2016). Functional connectivity was decreased in regions associated with face identification and expression processing, with symptom severity being associated with decreased activation patterns (Cheng et al., 2017).   46  The fusiform gyrus is not the only face processing area indicated in the pathophysiology of ASD. It has been suggested that an early developmental breakdown of the amygdala-fusiform system is instrumental in causing the difficulties in face perception and social cognition skills associated with ASD (Schultz, 2005). If social motivation requires input from the amygdala, then a natural consequence of altered connectivity between these regions would be reduced processing of social information. Early functional neuroimaging results found that unlike neurotypical controls, individuals with ASD do not activate the amygdala when attempting to judge how a person might be feeling by attending to the eyes (Baron-Cohen et al., 2000). Activation magnitudes for nine adults with ASD also differed from controls in both explicit and implicit tasks using emotional face expressions, with the FFA and left amygdala both showing reduced activity (Critchley et al., 2000). These early findings led researchers to form the amygdala theory of autism, also referred to as the amygdala dysfunction hypothesis, which suggests that abnormal amygdala function is behind at least part of the symptoms of ASD.   Subsequent behavioural evidence supported this hypothesis when eight individuals with ASD were assessed for various aspects of face and emotion recognition. Researchers sought to determine if their behavioural results were similar to those found in an earlier case study of an individual with impaired face expression recognition following damage to the amygdala (Adolphs et al., 2001; Adolphs et al., 1994). Consistent with amygdala dysfunction, the findings suggest that face expression difficulties are not the result of early perceptual processes, but instead reflect impaired abilities when combining social knowledge with human faces.   Hyperarousal was found in the amygdala for nine adults with ASD when viewing the isolated lower half of a face. However, there was hypoarousal in the amygdala when viewing whole face, which may explain aversion to the eyes in other studies (Ishitobi et al., 2011).   47 Hypoactivation of subcortical face processing systems, including the amygdala, have been described in functional neuroimaging studies analyzing responses to face stimuli (Kim et al., 2015; Kleinhans et al., 2011). Decreased synchrony, or underconnectivity, between the amygdala and temporal lobe was found for twelve adults with ASD, a region considered to be important for expression processing (Monk, 2010). Multiple neuroimaging studies looking at resting-state functional connectivity in individuals with ASD have found hypoconnectivity between the amygdala and subcortical regions (Assaf et al., 2010; X. Guo et al., 2016; Rausch et al., 2016).   1.4.4.2.      Face Identification in ASD  One of the earliest studies to assess face recognition in ASD found that unfamiliar face recognition is impaired, while also demonstrating that individuals with ASD have an enhanced ability to discriminate between buildings rather than faces (Boucher & Lewis, 1992). Since this early study, the nature of face identification, and whether or not impairment exists, has been the subject of a heavily debated literature. It is possible that face impairments are only present in some, not all, individuals with ASD, perhaps explaining the some of the controversy (Barton, Cherkasova, Hefter, et al., 2004).  An influential review found no evidence of qualitative differences in face identification, although quantitative differences were found in a variety of measures (Weigelt et al., 2012). While there are face-specific deficits in recognition tasks for people with ASD, especially in tasks with face memory demands, it was argued that these were not the result of a fundamentally different face processing strategy. A more recent review, however, contends that both qualitative and quantitative differences are found in individuals with ASD (Tang et al., 2015).   48  A pupillometric and looking-time study of preschool children with ASD found increased dilation for inverted faces, which the authors interpreted as an indication of larger processing load, along with longer fixation on face features, which suggest a local processing bias for individuals with ASD (Falck-Ytter, 2008). A later eye-tracking study found that individuals with ASD spend equal amounts of time viewing people and objects, while controls spend significantly more time attending to people (Wilson et al., 2010). For the ASD participants, impaired performance on a face-matching task was correlated with less time focusing on people in the eye-tracking study.  More recently it was found that memory for face recognition was impaired, while recognition memory for houses was unimpaired for adolescents with ASD (Arkush, Smith-Collins, Fiorentini, & Skuse, 2013). Interestingly, performance for both tasks was correlated only for the ASD group, suggesting that this population does not utilize specialized processing for face recognition memory.  1.4.4.3.      Recognition of Facial Expressions in ASD  It is possible that the difficulties with face expressions play an important role in the social dysfunction in ASD. For example, adults with ASD were impaired with tasks assessing emotional expression processing, but not with a socially-complex task that didn't involve analysis of facial expressions (Walsh, Creighton, & Rutherford, 2016). A number of studies have sought to examine this potential explanation, many with conflicting results. One influential review of face expression recognition in ASD suggests that the inconsistent findings in the literature may be due to demographic and experiment paradigm differences (Harms, Martin, & Wallace, 2010). More recent trends in the literature suggest that particular expressions may be   49 more impaired than others. An important meta-analysis of emotion recognition in ASD literature found that there is emotion recognition difficulty in ASD, with recognition of happiness being less impaired than other expressions (Uljarevic & Hamilton, 2013). However, pupillary responses to happy faces were reduced in the direct-gaze condition for children with ASD, indicating less intrinsic reward value to the most salient form of the social stimuli (Sepeta et al., 2012).  Two groups have developed specialized interventions in an effort to assess and train children with ASD to improve face expression processing skills (Golan et al., 2010; Tanaka et al., 2012). The Transporters is a 3D children’s animation series that features eight characters that are vehicles with real-life faces of actors displaying emotions grafted on the “face” of each vehicle (Golan et al., 2010). Each of the fifteen 5-minute episodes focus on a distinct emotion or mental state. Children with ASD who received the intervention by watching the animations significantly improved in emotion recognition compared to children with ASD who did not receive any intervention. The Let's Face It! (LFI!) Emotion Skills Battery is an assessment and intervention tool designed to evaluate verbal and perceptual skills important for emotional facial expression recognition and uses training protocols to improve face processing skills (Tanaka et al., 2012). Individuals with ASD were impaired both when labeling angry expressions and when asked to generalize a particular expression across different identities, suggesting avenues for future training within this intervention tool.   A consequence of amygdala dysfunction could be altered performance with negative expressions as compared to positive ones. In one study, expression recognition was impaired for angry, fearful, and sad expressions, while recognition was typical for happy expressions (Farran, Branson, & King, 2011). Adults with ASD were more likely to misinterpret happy faces as   50 neutral, while neutral faces were significantly more likely to be perceived as negative (Eack, Mazefsky, & Minshew, 2015). The bias towards perceiving expressions as negative was related to severity of social impairments, suggesting the two are linked in this disorder.  Behavioural studies designed to probe expression-specific abilities in this group have largely supported the neuroimaging data of altered amygdala activity (English, Maybery, & Visser, 2017; Howard et al., 2000; Kleinhans et al., 2010). Individuals with ASD performed similarly to those with amygdala lesions on recognition tasks for basic emotions (Adolphs et al., 2001). Additionally, trustworthiness ratings were impaired for a subset of the ASD group in a manner consistent with bilateral amygdala damage, suggesting both groups have difficulty integrating the socially relevant information from a face. Adult males with ASD performed typically with a face identity task but were less accurate with emotion recognition (C. Ashwin, Chapman, Colle, & Baron-Cohen, 2006). These impairments were specific to negative emotions, which would be in line with results from studies of patients with lesions to the amygdala. A later study of adult males with ASD also found that processing of negative expressions was impaired for a visual “face in the crowd” search paradigm (Farran et al., 2011). Most recently, a behavioural study tested children and adolescents with the dynamic affect recognition evaluation and found that identification of negative emotions was impacted for both groups (Back, Francis, Skankland, Wasserburg, & Jacob, 2016). While rates of amygdala dysfunction in the general population are unknown, an experiment utilizing an emotion recognition task found neurotypical individuals with higher AQ scores showed significantly worse abilities to distinguish anger, disgust, and sadness compared to individuals with lower AQ scores (Poljac, Poljac, & Wagemans, 2013). Children with ASD had reduced N170 amplitude across all face stimuli, but most pronounced in fearful expressions,   51 indicating difficulty at the structural encoding stage of emotional expression processing (Tye et al., 2014). Adolescents with ASD were impaired in identifying disgust and sadness and fixated more to the mouth region during an eye tracking study of emotion recognition and expression tasks (Wieckowski & White, 2017). An odd-ball eye tracking looking-time study of paired faces displaying neutral, disgusted (negative), and happy (positive) expressions for children and adults with ASD (Crawford, Moss, Anderson, Oliver, & McCleery, 2015). Both groups had an increased number of fixations for faces displaying disgust as compared to happiness, which was taken as evidence as a preference, or increased processing load, for negative expressions.  1.5 Objectives Previous research from behavioural studies of visual processing in individuals with ASD has demonstrated that there may be enhanced perception for certain aspects of a visual image. Given the wide range of visual stimuli and paradigms tested, it has been difficult to locate the neural source of this altered perceptual processing. Does the bias begin at the earliest parts of the visual system or are the behavioural results a consequence of higher-level visual processes? Is altered visual perception present throughout different levels of visual complexity, or do only certain types of visual stimuli or visual tasks elicit altered response patterns?   1.5.1 Chapter 2 objectives Multiple studies have indicated a higher incidence of refractive errors for children with ASD (Black et al., 2013; Ikeda et al., 2013; Scharre & Creedon, 1992), but few have assessed this in the adult population (Simmons et al., 2009). As incidence of refractive error increase with age, a gap remains in the literature as to the developmental trajectory for incidence of refractive error in   52 this population. Do individuals with ASD remain at the childhood frequency for refractive error and the general population eventually catches up? Or, does the level for refractive error continue to climb at a higher rate for adults with ASD? This would be reflected in higher incidence of refractive error in the adult group as compared to the age-matched control participants. Given the findings of atypical optometric measurements in children with ASD, which include higher rates of myopia, hyperopia, and astigmatism (Anketell, Saunders, Gallagher, Bailey, & Little, 2016; Black et al., 2013; Ikeda et al., 2013; Scharre & Creedon, 1992), we hypothesize that there would be increased rates of refractive errors in our group of adults with ASD compared to Controls and the general population. As rates of refractive error increase with age (Koretz et al., 2001), we predicted that the adult ASD group would have higher rates of clinically significant refractive errors compared to those reported in the studies with children. We also predicted that levels of myopia, hyperopia, and astigmatism would all be classified as more severe in our ASD group. Finally, we predicted that these rates would be evident across development, such that younger adults with ASD in our study would have higher rates and severity of myopia compared to Controls within the same age range, and that this trend would continue for older adults in our study. In Chapter 2, we clinically assessed refractive errors in adults with and without ASD to determine if adults with ASD have higher incidence or severity of refractive error than Control participants or the general population.  1.5.2 Chapter 3 objectives The EPF suggests that superior perception for certain aspects of a given image are the source of altered visual processing found in ASD. One principle of this model suggests that individuals with ASD have enhanced low-level visual processing of images (Mottron et al., 2006). Many low-level processes occur in the early visual cortex, suggesting that enhanced perception could   53 begin early in the visual processing pathway. In Chapter 3, we examine one of the earliest levels of cortical processing by focusing on one of the most basic visual image properties, orientation. This property has long been associated with V1 organization (De Valois et al., 1982). We made multiple experiment design decisions in a deliberate effort to minimize contributions from higher level processing. We used simple, exclusively visual tasks that did not require verbal labeling or stimuli categorization. Additionally, we chose to use highly specified simple stimuli (i.e. Gabor patches) so that the information content of the image is limited to low-level visual properties.  In three experiments, we systematically assessed three aspects of basic processing of visual orientation. We asked: "Do adults with ASD differ from individuals without ASD in measures of visual processing for (a) discrimination, (b) veridical perception, and (c) detection of orientation?" Orientation discrimination refers to the ability to discern a difference between one orientation and another. Veridical perception of orientation means that the observer is accurate in the perception of orientation at a specific angle (e.g. vertical is accurately perceived at 90 degrees). Orientation detection is the ability to identify the interval that presents a stimulus in a particular orientation against the interval that presents a blank screen. We chose to focus on orientation processing for visual stimuli because this is one of the most well characterized aspects of the early visual areas. From the earliest studies of orientation processing, one of the most reliable findings that has been observed in multiple species is that performance is superior at cardinal angles (horizontal and vertical) compared to oblique angles (e.g. diagonal; Appelle, 1972). In the general population, better performance at cardinal orientations is a well-established phenomenon, commonly referred to as the oblique effect. We assessed visual orientation processing across a range of base orientations for adults with and without ASD, allowing us to make meaningful comparisons both qualitatively and quantitatively. First, we asked: "Does the   54 shape of the curve (e.g. superior perception at cardinal angles) indicate a qualitative difference in how visual orientation is processed for adults with ASD and neurotypical controls?" The general trend in the literature towards altered visual perceptual response patterns in ASD (Behrmann, Thomas, & Humphreys, 2006; Bertone et al., 2005; Caron et al., 2006; Dakin & Frith, 2005) led us to hypothesize that adults with ASD may demonstrate qualitative differences in orientation processing compared to controls. Specifically, if individuals with ASD have qualitative differences in orientation processing, then we would expect to observe deviations from the classic oblique effect (e.g. a shallower or flat discrimination curve indicating a lack of typical enhanced perception at cardinal angles). Next, we asked, "Do adults with ASD show quantitative differences such as enhanced or impaired visual orientation processing, compared to individuals without ASD?" Based on the EPF, we hypothesized that individuals with ASD would demonstrate superior performance in each of our three tasks. In our three experiments, we predicted more precise discrimination, more accurate perception, and greater sensitivity for orientation in individuals with ASD compared to Controls. We assessed qualitative and quantitative differences in visual orientation processing between each group.  1.5.3 Chapter 4 objectives In our effort to understand the causes of social dysfunction in ASD, we next focused on high-level processing of the images of the upmost social importance: the human face. By gaining a greater comprehension of the differences found in face processing for individuals with ASD, specific rehabilitation paradigms can be developed to address the most pressing concerns that are impacting quality of life for this population. In Chapter 4, we examine two different aspects of face processing. Specifically, we asked, "Do severity of symptoms associated with ASD and social competence correlate with measures of (a) face identity, and (b) face expression?" We   55 used face identification and expression discrimination experiments and two questionnaires to investigate the relationship between face processing ability and social difficulties associated with the disorder. Face memory was assessed using the previously-mentioned CFMT (Duchaine & Nakayama, 2006). Another aspect of face identification is face-specific perception, or an individual’s ability to distinguish an individual face from another, as compared to the ability to distinguish one non-face object from another. For our second face identification experiment, we created a task that allowed us to measure an individual’s perception for identities of faces compared to houses. We hypothesized there would be reduced performance for face memory and perception in ASD compared to Controls. This would be demonstrated by lower accuracy in the memory task and higher thresholds in the face-specific perception task. As performance with non-face tasks is often typical in ASD, we predicted that perception abilities in the house identification task would be on par with Controls.   Another important aspect of face processing, and arguably the more socially-relevant facet, is the ability to differentiate between emotional expressions. Our final face experiment sought to assess an individual's ability to discriminate between different expression morphs displayed by the same individual without the use of explicit labeling, thereby avoiding some of the confounds that plague this literature (Harms et al., 2010). We first ask, "Do individuals with ASD demonstrate significantly larger impairments when asked to discriminate one expression type from another?" The amygdala dysfunction hypothesis predicts that difficulties in face expression processing should be limited to negative expressions (Schultz, 2005). The use of expression morphs allows discrimination thresholds to be calculated for positive and negative expressions, thereby permitting comparisons to be made of relative performance for each   56 individual expression. We hypothesized that individuals with ASD would show impaired face expression discrimination abilities, specifically for negative expressions.  Finally, we examined the relationship between face processing abilities in ASD and social competence and ASD symptom severity. If face identification and expression processing are related to social competence, then individuals with lower social competence and/or higher scores of ASD symptom severity would demonstrate greater impairments in face processing performance. We hypothesized that face memory scores would be positively correlated with our social competence measures and negatively correlated with AQ scores. For our face-specific perception and expression discrimination tasks, we hypothesized that threshold scores would be negatively correlated with our social competence measures and positively correlated with AQ scores for our adults with ASD.  The primary goal of this dissertation is to investigate perception of simple and complex visual stimuli in adults with ASD and control subjects and to determine whether any qualitative and quantitative differences found are associated with measures of social competence and autism severity. By contributing to a greater understanding of this multifaceted neurodevelopmental disorder, this research seeks to provide pertinent information regarding sensory issues in ASD with the ultimate goal of informing treatments that will lessen that impact of these sensitivities on people with ASD and their families.    57 Chapter 2: Optometric assessment of adults with ASD  The enhanced perceptual functioning model suggests that individuals with ASD are biased towards local details of visual stimuli and this, in part, influences the superior processing of details commonly reported in the literature (Mottron et al., 2006). It has been suggested that this predisposition for minute features of an object or scene may be the result of atypical bottom-up processes beginning as early in the visual stream as the retina or Visual Areas 1 and 2. If an image cannot be focused directly on the retina due to the shape of the eye, cornea, or lens, then a refractive error will present as blurred vision, or reduced visual acuity (Crick & Khaw, 1997). As some of the most severely impaired individuals with ASD often lack the ability to communicate with optometrists, the absence of adequate correction via prescription lenses could be compounding the problems associated with this disorder. Studies looking at the rates of refractive errors in children with ASD have generally found increased incidence compared to controls, although there is wide variability between reports (Black et al., 2013; Ezegwui et al., 2014; Ikeda et al., 2013; Kabatas et al., 2015; Scharre & Creedon, 1992). In addition, despite the various studies reporting incidence of refractive error in children with ASD, there remains a wide gap in the literature as to whether these trends are also found in the adult population with ASD.   In order to determine if refractive errors are more prevalent for adults with ASD compared to their neurotypical peers, we collaborated with a practicing optometrist from the community. Based on the visual exam performed by the optometrist, we obtained a detailed account of each participant's visual acuity while ensuring that individuals who presented with refractive errors wore appropriate lens corrections. We compared the rates of refractive errors (myopia, hyperopia, and astigmatism) between participants with ASD and control participants, as well as known estimates within the general population.    58 2.1 Methods  2.1.1 Participants The presence of refractive error was assessed for 35 individuals with ASD (13 females, ages 16-47, mean age = 23.8, SD=7.9) and 35 control participants (13 females, ages 18-45, mean age = 25.5, SD=6.4) by a practicing optometrist. ASD participants were primarily recruited through our collaboration with the Autism and Developmental Disorders (ADDL) at Simon Fraser University. Control participants were primarily recruited from the University of British Columbia’s Department of Psychology’s paid study list serve. For the majority of our participants (N =22), diagnosis was confirmed using the ADOS. Clinical diagnosis was confirmed via the diagnosing clinician who referred the participant to our study. The institutional review boards of the University of British Columbia, Simon Fraser University, and Vancouver General Hospital approved the protocol. All subjects gave informed written consent in accordance with the Declaration of Helsinki.    2.1.2 Setup for the eye exam  An automated refractor, or autorefractor, is a machine utilized by optometrists during eye examinations to provide computer-based measurements of an individual's refractive error and lens prescription for each eye. This device determines when a patient is able to focus an image (e.g. a red hot air balloon) and calculates the refraction of the eye and reports three values, the spherical power (SPH), cylinder power (CYL), and axis of astigmatism (AXIS). The SPH of the eye is a report of the lens power (in diopters) required to correct for refractive error, with a minus sign indicating myopia (nearsightedness) and a plus sign indicating hyperopia (farsightedness). A CYL value is included for individuals who present with astigmatism and describes the lens power (in diopters) required to correct for the irregular shape of the cornea. Axis refers to the   59 direction, or lens meridian, of the astigmatism on the cornea (in degrees). Axis numbers range from 0 to 180, where 90 degrees represents the vertical meridian and 180 degrees represents horizontal meridian directions for the astigmatism.  Reading eye cards display short sentences at decreasing font sizes and are designed to assess visual acuity at closer distances. These cards are often used to measure visual accommodation, or the process of the eye changing optical power by focusing on objects at near or far distances. Accommodation is necessary for near-distance tasks like reading or using a computer. Using an accommodation convergence rule, the reading card is presented at 67 cm from the participant. Individuals are asked to read the short sentences beginning with the largest print and to continue reading until the font becomes too blurry to read. The smallest sentence correctly read is converted to report standard Snellen values.  2.1.3 Procedure  Participants were asked to remove any eyeglasses or contact lenses before sitting at an adjustable table that allowed the autorefractor to be moved to the appropriate height for each individual. The optometrist asked participants to lean forward and place their chin and forehead in the headrest so one eye was in front of the measuring window. She then instructed participants to avoid blinking or squinting so the machine could take a picture of the eye while a red hot air balloon comes in and out of focus. Once the autorefractor has completed its measurements of one eye, it would send the results to a computer before automatically moving the measuring window to the other eye and repeating the procedure. The SPH, CYL, and AXIS values were recorded in the participants’ file.   60  For participants wearing glasses, lenses were measured using a lensometer to confirm they were the appropriate prescription for that individual's type and degree of refractive error. If the prescription measured by the lensometer matched the autorefractor values, the glasses could be worn for the rest of the eye examination and the experiments that followed. If the lenses were not the correct prescription, the optometrist would use an instrument called a manual refractor, or phoropter, to test individual lenses on each eye in order to determine the correct prescription. Once the correct prescription had been determined, the optometrist would loan us trial lenses for the duration of the eye examination and subsequent experiments. Contact lenses are not able to be measured with a lensometer, so to confirm they were the correct prescription individuals were asked to wear their contacts during the rest of the examination. If these individuals were unable to reach normal visual acuity while wearing contacts, it was assumed that the prescription was incorrect. Just as in the case of inappropriate prescription glasses, any individuals whose contact lenses were deemed insufficient would be loaned a pair of corrective glasses, trial lenses, for the duration of the experiments that followed (2 ASD, 1 control).  As the psychophysical experiments that followed the eye exam take place at intermediate distances, it was most important that participants have adequate vision for these distances. While seated in the examination chair, the optometrist would use a measuring rod to display a reading eye chart 60 cm away from the top of the forehead. This distance was chosen to ensure that an individual with 20/20 vision could view the smallest image size (corresponding to 5 pixels) in our overall series of experiments. Participants were asked to read the smallest sentence they could clearly see without squinting. Participants who were unable to clearly see sentences corresponding to 20/20 with correction were loaned a pair of trial lenses for the duration of the experiments that followed the eye examination.     61  2.1.4 Data analysis  The optometrist recorded the results for each of the measurements; including SPH, CYL, and AXIS values. ASD and control groups were compared for overall incidence of a) clinically significant refractive error, b) degree of refractive error, c) type of refractive error, and d) whether an astigmatism is present. A two-sample t-test was used to compare significant differences in optometric measures. A chi-square test was used to compare distribution across severity categories. Myopia (near-sightedness) was classified as “mild” if the spherical equivalent value was between -0.25 and -2.0 diopters (D), “moderate” if the value was between -2.25 D and -7.9 D; and finally, “severe” if the spherical value was greater than -7.9 D (Vitale, Sperduto, Ferris, & Iii, 2009). Hyperopia (far-sightedness) was classified as “low” if the spherical value was between 0 D and +2.00 D, “moderate” if the refractive error ranged between +2.25 D and +5.00 D, and “high” for errors greater than +5.00 (B. D. Moore et al., 2008). “Plano” is a term describing the lack of SPH, or spherical refractive errors.  Astigmatism is also classified by severity and is based on the lens power prescription for the CYL values. Unfortunately, there is a great deal of variability in the literature as to the precise CYL values used to classify severity of astigmatism, making it difficult to compare findings between studies. A number of studies chose to use 1.0 diopters as the cut-off CYL for astigmatism, but few chose to classify levels of severity beyond that general cut-off point (Anketell et al., 2016; Ezegwui et al., 2014; Scharre & Creedon, 1992). For the present study, we chose to classify CYL values less than 1.0 diopters as “non-significant astigmatism” or (NSA), CYL values between 1.0-1.5 diopters as “low astigmatism”, CYL values were “moderate   62 astigmatism” if between 1.51 and 2.5 diopters, and “severe astigmatism” if the CYL value is above 2.5 diopters (Heidary, Ying, Maguire, & Young, 2005).  Astigmatisms can be further classified into types based on the orientation of the AXIS values. With-the-rule (WTR) axis values are between 1° to 15° or 165° to 180°, against-the-rule (ATR) for axis values between 75° to 105°, and oblique (OBL) for axis values between 16° and 74° or between 106° and 164° (Anketell et al., 2016).  2.2 Results  Values from the optometric measures for each ASD participant are presented in Table 2.1. Participants are ordered by age and autorefraction values from the left and right eye are included. Individuals who wear corrective lenses are indicated in the last column with “aided”, while those who do not require prescriptions to see intermediate distances are “unaided.”     63  ASD subject number Age Sex Eye Autorefraction SPH Autorefraction CYL Autorefraction AXIS Refractive error severity1,2 Astigmatism severity3 Type of astigmatism4 Aided or Unaided Acuity A01 16 F R 3.5 -1.25 8 Moderate hyperopia Low astigmatism WTR Aided    L 3.5 -0.5 163 Moderate hyperopia NSA OBL  A02 17 M R -0.75 -1.75 180 Mild myopia Moderate astigmatism WTR Aided    L -0.75 -2.25 3 Mild myopia Moderate astigmatism WTR  A03 17 F R 0.5 -0.25 159 Low hyperopia NSA OBL Unaided    L 0.75 0 0 Low hyperopia None None  A04 17 M R -2.25 0 0 Moderate myopia None None Aided    L -1.75 -1.5 172 Mild myopia Low astigmatism WTR  A05 17 M R -1.75 -0.5 100 Mild myopia NSA ATR Unaided    L -1.25 -0.25 134 Mild myopia NSA OBL  A06 17 F R Plano -0.75 170  None NSA WTR Unaided    L Plano -0.5 4  None NSA WTR  A07 18 F R -0.25 -0.5 151 Mild myopia NSA OBL Unaided    L -1.5 -0.25 17 Mild myopia NSA OBL    64 ASD subject number Age Sex Eye Autorefraction SPH Autorefraction CYL Autorefraction AXIS Refractive error severity1,2 Astigmatism severity3 Type of astigmatism4 Aided or Unaided Acuity A08 18 M R Plano -0.25 150  None NSA OBL Unaided    L -0.5 -0.5 68 Mild myopia NSA OBL  A09 18 M R -1.5 -0.5 102 Mild myopia NSA ATR Aided    L -1.75 -0.25 106 Mild myopia NSA OBL  A10 19 M R 1.75 -0.5 163 Low hyperopia NSA OBL Aided    L 1.5 -0.5 175 Low hyperopia NSA WTR  A11 19 M R Plano -0.25 100  None NSA ATR Unaided    L -1.25 -0.25 63 Mild myopia NSA OBL  A12 19 F R -5.5 -1.25 10 Moderate myopia Low astigmatism WTR Aided    L -3.75 -0.75 118 Moderate myopia NSA OBL  A13 20 M R -0.75 -0.25 15 Mild myopia NSA WTR Unaided    L Plano -0.25 134  None NSA OBL  A14 20 M R -1.25 0 0 Mild myopia None None Aided    L -2.25 0 0 Moderate myopia None None    65 ASD subject number Age Sex Eye Autorefraction SPH Autorefraction CYL Autorefraction AXIS Refractive error severity1,2 Astigmatism severity3 Type of astigmatism4 Aided or Unaided Acuity A15 20 F R -3.5 -0.75 16 Moderate myopia NSA OBL Aided    L -4.0 0 0 Moderate myopia None None   A16 21 M R -1.5 0 0 Mild myopia None None Unaided    L -1.5 0 0 Mild myopia None None  A17 21 M R -1.25 -0.25 111 Mild myopia NSA OBL Unaided    L -1.25 0 0 Mild myopia None None  A18 21 M R -2.75 -0.75 114 Moderate myopia NSA OBL Aided    L -2.25 -1.0 81 Moderate myopia Low astigmatism ATR  A19 22 M R -1.25 -0.5 170 Mild myopia NSA WTR Aided    L -1 0 0 Mild myopia None None  A20 22 M R -0.5 -0.75 120 Mild myopia NSA OBL Unaided    L -0.25 -0.5 172 Mild myopia NSA WTR  A21 22 M R -9.75 -1.25 164 Severe myopia Low astigmatism OBL Aided    L -9.0 -1.25 168 Severe myopia Low astigmatism WTR    66 ASD subject number Age Sex Eye Autorefraction SPH Autorefraction CYL Autorefraction AXIS Refractive error severity1,2 Astigmatism severity3 Type of astigmatism4 Aided or Unaided Acuity A22 22 F R -6.0 -1.0 158 Moderate myopia Low astigmatism OBL Aided    L -7.0 -0.75 37 Moderate myopia NSA OBL  A23 23 F R -4.0 0 0 Moderate myopia None None Aided    L -4.0 0 0 Moderate myopia None None  A24 23 M R -4.0 -0.25 170 Moderate myopia NSA WTR Aided    L -2.25 -1.25 9 Moderate myopia Low astigmatism WTR  A25 25 M R -0.25 -0.5 158 Mild myopia NSA OBL Unaided    L Plano -0.5 15  None NSA WTR  A26 26 M R -2.25 -1.0 177 Moderate myopia Low astigmatism WTR Aided    L -2.0 -1.5 169 Moderate myopia Low astigmatism WTR  A27 27 M R -0.25 -1.25 175 Mild myopia Low astigmatism WTR Aided    L 1.5 -2.25 172 Low hyperopia Moderate astigmatism WTR  A28 28 F R Plano 0 0  None None None Unaided    L Plano 0 0  None None None    67 ASD subject number Age Sex Eye Autorefraction SPH Autorefraction CYL Autorefraction AXIS Refractive error severity1,2 Astigmatism severity3 Type of astigmatism4 Aided or Unaided Acuity A29 28 M R -4.5 -0.5 6 Moderate myopia NSA WTR Aided    L -4.25 -0.5 79 Moderate myopia NSA ATR  A30 28 F R 0.5 -0.5 112 Low hyperopia NSA OBL Unaided    L 0.75 -0.75 75 Low hyperopia NSA ATR  A31 29 F R -0.25 0 0 Mild myopia None None Unaided    L -0.25 0 0 Mild myopia None None  A32 40 F R -0.5 -0.75 65 Mild myopia NSA OBL Aided    L -2.25 -0.75 104 Moderate myopia NSA ATR  A33 40 M R -0.25 -0.5 134 Mild myopia NSA OBL Unaided    L -0.25 -0.75 3 Mild myopia NSA WTR  A34 46 M R -0.5 -0.5 115 Mild myopia NSA OBL Aided    L Plano -1.75 65  None Moderate astigmatism OBL  A35 47 F R -0.75 -0.25 151 Mild myopia NSA OBL Unaided    L -0.5 0 0 Mild myopia None None     68 Table 2.1 Optometric values for participants with ASD. Individuals are listed in order by chronological age. Autorefraction values describing the refractive error and astigmatism are listed in SPH (spherical power), CYL (cylindrical power), and AXIS (direction of astigmatism in degrees). “Plano” is a term used when describing a lack of SPH refractive errors. The level of severity for the spherical values is reported as mild to severe myopia or low to high hyperopia. 1,2 Severity of astigmatism is reported as non-significant astigmatism (NSA), low, moderate, and severe based on the CYL values in diopters. 3 The type of astigmatism for each eye is classified as with-the-rule (WTR), against-the-rule (ATR), or oblique (OBL). 4 Whether the individual required corrective lenses during the experiment is indicated in the last column. Those who required prescription lenses were “aided” while those who are “unaided” did not require correction for intermediate viewing distances. 1(Vitale et al., 2009); 2(B. D. Moore et al., 2008); 3(Heidary et al., 2005); 4(Anketell et al., 2016).  Table 2.2 shows the values from optometric measures for our control participant group. As with our ASD participant group, participants are ordered by chronological age and autorefractor values are listed. All reported values and severity classifications are based on the same guidelines used for our ASD participant group.   Control subject number Age Sex Eye Autorefraction Sph Autorefraction Cyl Autorefraction Axis Refractive error severity1,2 Astigmatism severity3 Type of astigmatism4 Aided or Unaided Acuity C01 18 M R -2.5 -0.75 2 Moderate myopia NSA WTR Aided    L -2.75 -0.5 163 Moderate myopia NSA OBL  C02 19 M R -2.25 -0.5 96 Moderate myopia NSA ATR Aided    L -2.75 -0.25 105 Moderate myopia NSA ATR  C03 19 F R -7.5 -0.25 81 Moderate myopia NSA ATR Aided    L -7.5 -0.5 99 Moderate myopia NSA ATR    69 Control subject number Age Sex Eye Autorefraction Sph Autorefraction Cyl Autorefraction Axis Refractive error severity1,2 Astigmatism severity3 Type of astigmatism4 Aided or Unaided Acuity C04 19 F R Plano  0  0  None None None Unaided    L Plano -0.25 74  None NSA OBL  C05 21 F R -0.5 -0.25 94 Mild myopia NSA ATR Unaided    L -0.25 0 0 Mild myopia None None  C06 21 M R -1.0 -0.25 169 Mild myopia NSA WTR Unaided    L -1.0 -0.5 66 Mild myopia NSA OBL  C07 21 M R -2 -0.25 149 Moderate myopia NSA OBL Aided    L -2.5 -1.0 7 Moderate myopia Low astigmatism WTR  C08 21 M R -0.25 -0.5 74 Mild myopia NSA OBL Unaided    L -0.25 -0.5  110 Mild myopia NSA OBL  C09 22 M R -2.25 -2.5 19 Moderate myopia Moderate astigmatism OBL Aided    L -2.75 -2.0 19 Moderate myopia Moderate astigmatism OBL  C10 22 M R -3.0 -0.25 180 Moderate myopia NSA WTR Aided    L -3.25 -0.25 146 Moderate myopia NSA OBL    70 Control subject number Age Sex Eye Autorefraction Sph Autorefraction Cyl Autorefraction Axis Refractive error severity1,2 Astigmatism severity3 Type of astigmatism4 Aided or Unaided Acuity C11 22 F R Plano -0.5 150  None NSA OBL Unaided    L Plano 0 0  None None None  C12 22 M R -2.75 -1.25 166 Moderate myopia Low astigmatism WTR Aided    L -2.25 -1.0 7 Moderate myopia Low astigmatism WTR  C13 23 M R Plano -0.25 131  None NSA OBL Unaided    L -0.5 -0.25 25 Mild myopia NSA OBL  C14 23 M R 0.25 -3.5 5 Low hyperopia Severe astigmatism  WTR Aided    L -1.25 -2.5 179 Mild myopia Moderate astigmatism WTR  C15 23 M R Plano -0.25 157  None NSA OBL Unaided    L Plano 0 0  None None None  C16 23 F R -10.5 -0.5 110 Severe myopia NSA OBL Aided    L -9.25 0 0 Severe myopia None None  C17 23 M R -2.0 -0.25 120 Moderate myopia NSA OBL Aided    L -2.5 -0.5 95 Moderate myopia NSA ATR    71 Control subject number Age Sex Eye Autorefraction Sph Autorefraction Cyl Autorefraction Axis Refractive error severity1,2 Astigmatism severity3 Type of astigmatism4 Aided or Unaided Acuity C18 24 F R -6.0 -0.75 178 Moderate myopia NSA WTR Aided    L -6.0 0 0 Moderate myopia None None  C19 24 F R -0.75 -0.75 102 Mild myopia NSA ATR Unaided    L Plano -0.75 85  None NSA ATR  C20 24 M R -2.25 -0.75 169 Moderate myopia NSA WTR Aided    L -2.0 -1.0 175 Moderate myopia Low astigmatism WTR  C21 24 M R Plano 0 0  None None None Unaided    L Plano 0 0  None None None  C22 25 F R -4.0 -0.25 167 Moderate myopia NSA WTR Aided    L -3.5 -1.0 175 Moderate myopia Low astigmatism WTR  C23 25 M R -2.25 -1.0 173 Moderate myopia Low astigmatism WTR Aided    L -2.5 -1.0 7 Moderate myopia Low astigmatism WTR  C24 25 F R -5.5 0 0 Moderate myopia None None Aided    L -3.75 -1.0 2 Moderate myopia Low astigmatism WTR    72 Control subject number Age Sex Eye Autorefraction Sph Autorefraction Cyl Autorefraction Axis Refractive error severity1,2 Astigmatism severity3 Type of astigmatism4 Aided or Unaided Acuity C25 26 F R -6.5 0 0 Moderate myopia None None Aided    L -6.25 -0.25 11 Moderate myopia NSA WTR  C26 26 F R -2.25 -0.5 163 Moderate myopia NSA OBL Aided    L -2.25 -0.5 97 Moderate myopia NSA ATR  C27 26 M R 0.25 -0.75 100 Low hyperopia NSA ATR Unaided    L -0.5 0 0 Mild myopia None None  C28 27 M R -0.5 -0.75 69 Mild myopia NSA OBL Aided    L -1.25 -0.25 76 Mild myopia NSA ATR  C29 29 F R -3.75 -0.25 95 Moderate myopia NSA ATR Aided    L -6.0 -0.25 165 Moderate myopia NSA WTR  C30 30 M R 0.25 -0.25 137 Low hyperopia NSA OBL Unaided    L 0.5 0 0 Low hyperopia None None  C31 31 F R Plano -0.25 124  None NSA OBL Aided    L 0.5 -0.5 77 Low hyperopia NSA ATR    73 Control subject number Age Sex Eye Autorefraction Sph Autorefraction Cyl Autorefraction Axis Refractive error severity1,2 Astigmatism severity3 Type of astigmatism4 Aided or Unaided Acuity C32 38 M R -0.75 -1.5 180 Mild myopia Low astigmatism WTR Unaided    L -0.75 -0.75 5 Mild myopia NSA WTR  C33 39 M R Plano 0 0 None None None Unaided    L 0.25 0 0 Low hyperopia None None  C34 42 M R -2.75 -0.5 96 Moderate myopia NSA ATR Aided    L -2 -0.75 75 Moderate myopia NSA ATR  C35 45 M R -4.5 -0.25 65 Moderate myopia NSA OBL Aided    L -4.75 -0.25 117 Moderate myopia NSA OBL   Table 2.2 Optometric values for control participants. Individuals are listed in order by chronological age. Autorefraction values describing the refractive error and astigmatism are listed in SPH (spherical power), CYL (cylindrical power), and AXIS (direction of astigmatism in degrees). “Plano” is a term used when describing a lack of SPH refractive errors. The level of severity for the spherical values is reported as mild to severe myopia or low to high hyperopia. 1,2 Severity of astigmatism is reported as non-significant astigmatism (NSA), low, moderate, and severe based on the CYL values in diopters. 3 The type of astigmatism for each eye is classified as with-the-rule (WTR), against-the-rule (ATR), or oblique (OBL). 4 Whether the individual required corrective lenses during the experiment is indicated in the last column. Those who required prescription lenses were “aided” while those who are “unaided” did not require correction for intermediate viewing distances. 1(Vitale et al., 2009); 2(B. D. Moore et al., 2008); 3(Heidary et al., 2005); 4(Anketell et al., 2016).  We calculated the spherical equivalent (SphEq) value for each eye, defined as !"ℎ$% = !'( − *+,-  . We then analyzed data from the eye with the greater absolute spherical equivalent value as defined in Vitale et al. (2008). This particular study classified   74 myopia and hyperopia slightly differently than in the previously described studies. Clinically significant myopia was defined as a SphEq value less than -1.0 D, severe myopia less than -5.0 D, and hyperopia was defined as 3.0 D or greater. Nineteen (54.3%) of our ASD participants were classified as having clinically significant myopia, of those 2 (5.7%) were classified as having severe myopia. One (2.9%) individual with ASD had hyperopia. For our Control participants, 20 (57.1%) were myopic, six (17.1%) of which had severe myopia. No Controls had clinically significant hyperopia. A two-sample, two-tailed Student’s t-test of the eye with greater absolute SphEq value (in diopters) indicated there was no significant difference between the ASD and Control groups (t(68)=1.27, p=0.21). For comparison, rates of refractive error in the general population were 33.1% (95% CI: 31.5-34.7%) for myopia, 6.5% (95% CI: 5.8-7.2%) for severe myopia, and 3.6% (95% CI: 3.2-4.0%) for hyperopia (Vitale et al., 2008). When we look just at clinically significant myopia within our participant groups, our ASD group had 19 individuals who met criteria for myopia while our Controls had 20 individuals. A two-sample, two-tailed Student’s t-test of SphEq values indicated there was no significant difference between groups for severity of clinically significant myopia (t(37)=1.12, p=0.27). Our ASD and Control participants are outside the 95% confidence interval reported in Vitale et al. (2008) for myopia and hyperopia in the general population. In both of our participant groups, rates of myopia were higher than reported rates in the general population. Another study of rates of refractive errors in the general population used different criteria for myopia, specifying -0.75 D or lower as being clinically significant (K. M. Williams et al., 2015). This study also used SphEq, but averaged both eyes to use a single SphEq per individual. When we average the SphEq values from each participant and reanalyze our results with this new cut-off, we find that 17 (48.6%) of adults with ASD and 20 (57.1%) of Control participants   75 are classified as having significant myopia. A two-sample, two-tailed Student’s t-test of the average SphEq value for each participant indicated no significant differences between our two groups (t(68)=1.41, p=0.16). Our group rates of myopia are both higher than and outside the confidence interval range reported for the general population at 30.6% (95% CI: 30.4-30.9%) (K. M. Williams et al., 2015).  To allow for comparison between our results and other reported rates of refractive error in the general population, we next classified our findings based on the criteria described by two other studies. Severity of myopia and hyperopia were specified based on SphEq value classification definitions in Vitale et al. (2009) and B. D. Moore et al. (2008), respectively. We summarized the relevant information from the right eye to analyze the types of SphEq refractive errors (myopia and hyperopia) and relative rates of occurrence in Table 2.3. We excluded classifications that were not represented in our sample (i.e. none of our participants had severe hyperopia, so that option is not included in the table). The overall percentage of eyes with  each classification of SphEq refractive error severity is reported for each participant group.   Table 2.3 Types and severity of refractive error for the right eye of participants with ASD and controls. The number and percentage of participants with each kind of SphEq refractive error are listed, excluding those categories that were not represented in our sample. Myopia is considered “mild” if it is between -0.25D and -2.0 D, “moderate” if the SphEq is between -2.25 D and -7.9 D, and “severe” if the SphEq was lower than -7.9 D 1(Vitale et al., 2009). Hyperopia is “mild” if it is between +0.25 and +2.00 D, and “moderate” if between +2.25 D and +5.00 D 2(B. D. Moore et al., 2008).   Classification1,2 ASD- number of eyes (35 total) ASD- percentage of eyes Control- number of eyes (35 total) Control- percentage of eyes None 9 25.7 9 25.7 Myopia- mild 11 31.4 8 22.9 Myopia- moderate 8 22.9 13 37.1 Myopia- severe 1 2.9 1 2.9 Hyperopia- low 5 14.3 4 11.4 Hyperopia- moderate 1 2.9 0 0   76 Table 2.3 shows a peak frequency of mild myopia for adults with ASD, while Controls have a peak frequency of moderate myopia. Based on a chi-square test, there was no significant difference in the relative distribution across severity categories for our ASD and Control participants (p = 0.73). When all forms of myopia are combined according to the Vitale et al. (2009) classification criteria, we find 20 (57.1%) of ASD participants and 22 (62.9%) of Control participants have myopia.  Figure 2.1 displays the distribution of SphEq refractive error for both eyes in our ASD and Control groups, reported in diopters. Without consideration of clinical significance, refractive errors were generally classified as “myopia” if the SPH value is negative, “hyperopia” if the value is positive, and “Plano” if the SPH value is 0 (Scharre & Creedon, 1992) myopia.       77   Figure 2.1  Number of eyes with SphEq values. The ASD group is represented in red, controls are in blue. Negative values indicate myopia, positive values indicate hyperopia, and 0 is “Plano” or the absence of refractive error.   For our ASD group, SphEq refractive errors ranged from -9.125 D to 4.125 D, with a median of -0.8125 D. In our Control group, SphEq refractive errors ranged from -10.25 D to 2.0 D, with a median of -1.6875 D. Scharre and Creedon (1992) defined clinically significant myopia as less than or equal to -1 D and hyperopia as greater than or equal to 1 D. If we classify both eyes for our participants according to these criteria, we find that 35 (50%) of eyes in our ASD group and 39 (56%) of eyes in our Control group meet the criteria for clinically significant myopia. Six (8.6%) of eyes in our ASD group and 1 (1.4%) of eyes in our Control group meet the criteria for hyperopia. A two-sample two-tailed Student’s t-test showed no significant differences for clinically significant myopia between our ASD and Control groups (t(72)=1.49,   78 p=0.14). For children with ASD, Scharre and Creedon (1992) report 8.8% had clinically significant myopia and 17.6% had hyperopia.  We next obtained values from the eye with the highest absolute CYL value to allow for comparisons between groups, where clinically significant astigmatism was defined as a CYL value of 1.0 D or greater (Vitale et al., 2008). For our ASD group, 11 (31.4%) met the criteria for astigmatism, while 9 (25.7%) of our Controls had clinically significant astigmatism. A two-sample two-tailed Student’s t-test shows no significant difference between the ASD and Control groups for the eye with worse CYL refractive error (t(67)=-0.14, p=0.89). For comparison, clinically significant astigmatism affects 36.2% (95% CI: 34.9-37.5%) of the general population (Vitale et al., 2008). For both our ASD and Control groups, rates of astigmatism were lower than the 95% confidence interval reported by Vitale et al. (2008).  K. M. Williams et al. (2015) used the same cut-off for significant astigmatism, but instead calculates an average CYL value for both eyes. When we use the same calculation, we find that 6 (17.1%) of our ASD and 5 (14.3%) of our Control participants have significant Astigmatism. K. M. Williams et al. (2015) reports a prevalence of 23.9% (95% CI: 23.7 - 24.1%). When we compare both studies to our results, our findings indicate that our ASD and Control have of lower rates of astigmatism than the general population, although it is worth noting that these findings are based on population samples with greater numbers of participants in older age ranges.  A power analysis of these astigmatism findings was conducted using the Pwr package in R programming language (Champely, 2012). We compared sample proportions for this analysis. Power is a function of the sample sizes in each group and the difference between proportions in each group. The larger the sample sizes and the further apart the proportions, the more power we   79 have to detect significant differences between groups. Power is also a function of the type 1 error rate, which we have set to 0.05. Equations for power for this test are specified in (Cohen, 1988). For comparing the ASD and Control groups based on the Vitale et al. (2008) study inclusion criteria and reported rates of astigmatism, we determined the power we have with this sample size to detect a 20% difference between the rates of astigmatism. We calculated power for two proportions with equal sample sizes across the two groups and found our power to be 0.65. Using a power calculation for two proportions with unequal sample sizes and a proportion difference of 0.20, we found that the power to detect significant differences between our ASD group and the general population was 0.71. Finally, assuming a 0.20 proportion of difference between groups, we found our ability to detect a significant difference between our Control group and the general population was 0.74. These values indicate that our study was moderately underpowered to detect a difference of 20%. Next, we carried out the same power analyses for our astigmatism results based on the inclusion criteria and rates reported in K. M. Williams et al. (2015) and we determined the power we have with this sample size to detect a 20% difference between the rates of astigmatism. We found that for two proportions with equal sample sizes across two groups, we had a power value of 0.66. Using a power calculation for two proportions with unequal sample sizes and a proportion difference of 0.20, we found that the power to detect significant differences between our ASD group and the general population was 0.86. Finally, assuming a 0.20 proportion of difference between groups, we found our ability to detect a significant difference between our Control group and the general population was 0.85. These values indicate that we have sufficient power to detect significant differences of 20% for both the ASD and Control groups compared to the general population.   80 Next, in Table 2.4, we summarized the presence and severity of astigmatism reported in CYL values for both eyes in our ASD and Control participants. NSA represents non-significant astigmatism, where such minor protuberances in the shape of the cornea do not cause a significant CYL error or require corrective lenses. We converted the classification criteria of Heidary et al. (2005) from positive to negative to allow our results to be analyzed. With this conversion, CYL values between -1.0 and -1.5 D are classified as “low astigmatism”, values between -1.51 D and -2.5 D are “moderate astigmatism”, and values -2.51 D or lower are classified as “severe astigmatism” (Heidary et al., 2005).    Table 2.4 Severity of astigmatism based on the CYL values for participants with ASD and controls. The number and percentage of participants with each severity of astigmatism are listed, excluding those categories that were not represented in our sample. Classification criteria were converted from positive to negative to allow our results to be compared to Heidary et al. (2005). Non-significant astigmatism (NSA) describes CYL values less than 1.0 D. CYL values between -1.0 and -1.5 D are classified as “low astigmatism”, values between -1.51 D and -2.5 D are “moderate astigmatism”, and values below -2.5 are “severe astigmatism” 1(Heidary et al., 2005). There are no significant differences between the ASD and Control group in terms of rates of each type of astigmatism (p=0.796) seen in Table 2.4, based on a chi-square test. For each group, NSA (less than 1.0 D) were the most common classification type for astigmatism. Classification1 ASD- number of eyes (70 total) ASD- percentage of eyes Controls- number of eyes (70 total) Controls- percentage of eyes None 16 22.86 14 20.00 NSA 39 55.71 43 61.43 Low astigmatism  11 15.71 9 12.86 Moderate astigmatism 4 5.71 3 4.29 Severe astigmatism  0 0 1 1.43   81 Figure 2.2 shows the CYL astigmatism severity in diopters for both ASD (red) and Control (blue) groups. Astigmatism manifested in 54 eyes for the ASD group and 56 eyes for the control group.   Figure 2.2  The number of eyes with and without CYL astigmatism severity in diopters. ASD (red) and Control (blue) participants were not significantly different from one another.   Scharre and Creedon (1992) defined clinically significant astigmatism as CYL values greater or equal to 1.0 D. When we convert this cut-off to a negative value to allow for comparisons with our own results, we find that 15 (21.4%) of ASD participant eyes and 13 (18.6%) of Control participant eyes have clinically significant astigmatism. These rates are both lower than the 29.4% reported by (Scharre & Creedon, 1992) for children with ASD.    82  Finally, in Table 2.5, individual eyes are classified by type based on the AXIS values. WTR describes astigmatisms that are oriented around the horizontal axis, or 1° to 15° or 165° to 180°. ATR describes astigmatisms around the vertical axis, or between 75° to 105°. OBL astigmatisms are oriented between the horizontal and vertical axis, with values ranging between 16° and 74° or between 106° and 164° (Anketell et al., 2016).      Table 2.5 Classification for each type of astigmatism as defined by the AXIS value. Each eye was classified based on the orientation of the astigmatism as with-the-rule (WTR) if it was between 1° to 15° or 165° to 180°, against-the-rule (ATR) if the axis value was between 75° to 105°, or oblique (OBL) for axis values between 16° and 74° or between 106° and 164°. 1(Anketell et al., 2016).  There was no significant difference in classification of axis values between individuals with ASD and Control participants (p=0.308), based on a chi-square test.  Figure 2.3 visually depicts Table 2.5 and shows the number of eyes with each type of astigmatism based on the AXIS values for the ASD group (red) and the control group (blue).  Classification1 ASD- number of eyes (70 total) ASD- percentage of eyes Controls- number of eyes (70 total) Controls- percentage of eyes None 16 22.86 14 20.00 WTR 21 30.00 20 28.57 ATR  7 10.00 15 21.43 OBL 26 37.14 21 30.00   83  Figure 2.3  The number of eyes with types of AXIS astigmatism. ASD (red) and Control (blue) participants were divided based on AXIS values classifications, described as WTR, ATR, or OBL depending on the orientation of the astigmatism.   As myopia makes up the majority of refractive errors in our sample, we turn our focus to comparisons of the severity of myopia between our participant groups and the general population.  Unfortunately, there are few studies that look at rates of refractive errors in the general population. The most comprehensive analysis to date was done by Vitale et al. (2009) looking at the National Health and Nutrition Examination Survey results from 1971-1972 and comparing those findings to results from the 1999-2004 versions of the same survey. They found that incidence of myopia significantly increased in recent years and analyzed the severity of myopia in the right eye as a function of age. To allow for comparisons between the Vital et al. study (2009) and our own, we next looked at severity of myopia in the right eye for our   84 participants in specific age ranges. As with Table 2.3, SphEq values were classified as mild myopia (higher than -2.0D), moderate myopia (between -2.0D and -7.9D), or severe myopia (lower than -7.9D). The number and percentage of participants with a classification of myopia are listed in Table 2.6. The final column includes the values from Vidal et al.’s analysis of the prevalence of myopia in the US general population between the years of 1999 and 2004, along with the reported 95% confidence interval.    85  Classification Age range Number of participants with ASD Percentage of participants w/ ASD Number of control participants Percentage of control participants Prevalence of myopia in the general population, (95% CI)1 All levels of myopia        12-17  2 33.3 0 0 33.9 (30.8-37.0)  18-24  13 72.2 14 66.7 38.1 (34.6-41.6)  25-34  3 42.9 6 60.0 44.0 (40.3-47.7)  35-44  0 0 1 33.3 42.9 (39.8-46.1)  45-54  2 100.0 1 100 44.8 (41.0-48.6)  Total 20 57.1 22 62.9 41.6 (39.8-43.4)        Myopia-  mild       <-0.25 to >-2.0 12-17  1 16.7 0 0 16.9 (14.8-19.0)  18-24  6 33.3 7 33.3 16.8 (13.8-19.8)  25-34  2 28.6 1 10.0 17.0 (14.8-19.3)  35-44  0 0 0 0 15.9 (13.5-18.2)  45-54  2 100.0 0 0 20.6 (18.1-23.1)  Total 11 31.4 8 22.9 17.5 (16.2-18.9)        Myopia- moderate       <-2.0 to >-7.9 12-17  1 16.7 0 0 16.7 (14.4-19.0)  18-24  6 33.3 6 28.6 19.9 (16.3-23.4)  25-34  1 14.3 5 50.0 24.7 (21.6-27.8)  35-44  0 0 1 33.3 24.6 (21.8-27.4)  45-54  0 0 1 100.0 23.1 (19.3-26.8)  Total 8 22.9 13 37.1 22.4 (20.7-24.1)   86        Myopia-  severe       <-7.9 12-17  0 0 0 0 0.3 (0.1-0.6)  18-24  1 5.6 1 4.8 1.4 (0.6-2.3)  25-34  0 0 0 0 2.2 (1.2-3.2)  35-44  0 0 0 0 2.4 (1.5-3.3)  45-54  0 0 0 0 1.1 (0.6-1.7)  Total 1 2.9 1 2.9 1.6 (1.3-2.0)  Table 2.6. Classification of myopia in the right eye for specific age ranges. Myopia is reported with all levels combined, mild myopia (defined as greater than -2.0 D), moderate myopia (between -2.0 D -7.9) and severe myopia (defined as less than or equal to -7.9 D). Numbers and percentages of myopia severity within our ASD and control participant groups are included for each specific age range. 1The final column contains the percentages in the general population for 1999-2004 as well as confidence intervals (CI) reported in Vitale et al. (2009).   There were no significant differences between ASD and Control participants for relative rates of myopia severity (p=0.46). As indicated in Table 2.6, our ASD and Control groups had greater incidence of myopia (57.1% and 62.9%, respectively) compared to 41.6% (95% CI: 39.8-43.4%) prevalence reported in the Vitale et al. (2009) study and were outside the boundaries of the 95 confidence interval. Rates of mild myopia in our ASD participants were higher (31.4%) than Controls (22.9%), and were outside the range of the 95% confidence interval for the general population (17.5%, 95% CI: 16.2-18.9%). Rates of moderate myopia for our ASD group (22.9%) were within the expected range for the general population (22.4%, 95% CI: 20.7-24.1%) while our Control group (37.1%) was higher than the prevalence for the general population. While there was only 1 individual in each group who had severe myopia, corresponding to 2.9% of each group, this was higher than the reported range for the general population (1.6%, 95% CI: 1.3-2.0%). Our results indicate that rates of mild and severe myopia are higher in ASD than expected ranges for the general population.  87 Additionally, our Control participants had higher ranges of myopia compared to the general population for all categories of myopia. Taken together, these results indicate differences for relative rates of myopia severity classification for individuals with ASD compared to our control group and the rates in the general population as reported by Vitale et al. (2009).1 A power analysis of our myopia findings was conducted using the Pwr package in R programming language (Champely, 2012). For comparing the ASD and Control groups we determined the power we have with this sample size to detect a 20% difference between the rates of myopia. We calculated power for two proportions with equal sample sizes across the two groups and found our power to be 0.40. Using a power calculation for two proportions with unequal sample sizes and a proportion difference of 0.20, we found that the power to detect significant differences between our ASD group and the general population was 0.66. We found our ability to detect a significant difference between our Control group and the general population was 0.85. These results indicate that we were moderately underpowered to detect a 20% difference between our participant groups and the ASD group compared to the general population. We did have sufficient power to detect a 20% difference between our Control group and the general population.                                                     1 The discrepancy between our previously reported rates of myopia in ASD of 54.3% for Vitale et al. (2008), 48.6% for K. M. Williams et al. (2015), and the rate of 57.1% reported in our comparison to Vitale et al. (2009) are due to differences in inclusion criteria, myopia severity definitions, and SphEq calculations between the three studies. The discrepancies in reported myopia rates in our Control population of 57.1% in Vitale et al. (2008), 57.1% in K. M. Williams et al. (2015), and 62.9% in Vitale et al. (2009) are due to the same differences between studies.   88 2.3 Discussion We did not find significant differences between ASD and control participants for SPH, CYL, and AXIS values. Unfortunately, it is difficult to compare our findings to rates in the general population as there is wide variability in the literature regarding criteria for clinically significant myopia and reported prevalence. If we use the same -0.75 D cut-off for myopia as K. M. Williams et al. (2015), we find that the 48.6% of adults with ASD and 57.1% of Control participants are not within the confidence interval of the 30.6% (95% CI: 30.4–30.9%) of individuals with myopia in the general population. When we use a -1.0 D criterion for myopia as in Vitale et al. (2008), we also find higher rates of myopia for ASD (54.3%) and Control (57.1%) participants compared to those reported in the general population 33.1% (95% CI: 31.5-34.7%) for the eye with the most severe absolute SphEq. Rates of myopia for ASD (57.1%) and Control (62.9%) groups were higher compared to the 41.6% (95% CI: 39.8-43.4%) prevalence in the general population reported in Vitale et al. (2009). Regardless of the specific criteria used, we find higher rates of myopia than each of the three population prevalence reports in the literature. The prevalence of astigmatism in our ASD population (17.1%) and control group (14.3%) are both lower than rates for adults in the general population at 23.9% (95% CI: 23.7-24.1%) reported by K. M. Williams et al. (2015). We had sufficient power above a 0.80 level to detect a 20% difference between both our ASD and Control groups compared to the general population. Adults with astigmatism alone, meaning a Plano SPH value combined with a CYL value of -1.0 D or less, make up between 14–16% of the general population (Gomez-Salazar et al., 2017). Only 10% of our ASD group and 8.6% of our control participants have astigmatism alone. When we compare our results to another study that also used CYL values of 1.0 D or greater to classify clinically significant astigmatism, we found rates of astigmatism were lower   89 for both our ASD and Control groups compared to the prevalence of 36.2% for the general population reported in (Vitale et al., 2008). We did not have sufficient power to detect a significant difference between both groups and the general population as our values were all below the 0.80 level. Finally, when we compare our results to those reported for children with ASD in Scharre and Creedon (1992), we find lower rates of clinically significant astigmatism for our ASD (21.4%) and Control (18.6%) participants compared to the reported 29.4% for children on the spectrum. It is possible that the discrepancy can be explained by the differences in diagnostic criteria between the earlier study and our own (Simmons et al., 2009). Taken together, these results indicate that for both our adult participant groups rates of astigmatism are lower compared to the general population. It is possible that these differences reflect the age differences between our participants and data used in the population studies. The previously described population studies all utilized data from thousands of individuals across a range of ages (e.g. 60+ years old), while ours were limited to younger age groups and did not include individuals past their late 40s.  A wide range of refractive errors have been reported for children with ASD, making it hard to compare our adult participants with children and draw any firm conclusions as to how prevalent these optometric issues are within this population (Bakroon & Lakshminarayanan, 2016). While some studies suggest the magnitude and prevalence of astigmatic errors are significantly greater for a group of children with ASD (Anketell et al., 2016), others have found that these differences are not quite so pronounced (Black et al., 2013; Ikeda et al., 2013). As some of the earlier studies reporting significant differences in refractive error were based on the narrower diagnostic criteria of DSM-III (Denis, Burillon, Livet, & Burguière, 1997; Scharre & Creedon, 1992), it is possible that these increased rates of optometric issues only apply to a   90 subset of individuals diagnosed under the current DSM-5 criteria (Simmons et al., 2009). As rates of myopia increase with age, our findings of higher rates of clinically significant myopia in our ASD and Control groups compared to reported rates for children with ASD in Scharre and Creedon (1992) are likely due to age differences as participants in their study ranged from 2-11 years old and we tested adults from 16-47 years old. It is also possible that children with ASD have higher incidence of refractive errors caused by higher SPH values and/or the presence of astigmatism, but that rates within the general population increase throughout the lifespan until they become even with adults with ASD. This would suggest that individuals with ASD begin with a disadvantage in optometric measures that becomes less pronounced during development in adolescence and adulthood. These results are limited by our small sample size compared to the established norms from thousands of individuals reported in Vitale et al. (2009) across a five-year span. Our power values were lower than 0.80 for our ASD group, making it less likely that the test will correctly reject a null hypothesis. We did have sufficient power to detect significant differences between the Control group and general population. Additionally, as rates of myopia continue to increase (Bloom, Friedman, & Chuck, 2010; Pan, Ramamurthy, & Saw, 2012), it is possible that the results from 1999 to 2004 used in the Vitale et al. (2009) study are no longer representative of the general population, specifically for young adults (French, Morgan, Mitchell, & Rose, 2013; Ramamurthy, Lin Chua, & Saw, 2015; Sherwin et al., 2012). The higher incidence of mild myopia within our ASD and control sample compared to general population estimates could be the result of the cumulative effects of increased time spent engaging in near-focused activities, such as looking at computer screens, video game use, and handheld electronic devices (Foster & Jiang, 2014; Ramessur, Williams, & Hammond, 2015). These are all activities that are known to   91 be common circumscribed interests for adolescents and young adults with ASD (Engelhardt, Mazurek, & Hilgard, 2017; Gillespie-Lynch, Kapp, Shane-Simpson, Smith, & Hutman, 2014; Mazurek, Engelhardt, & Clark, 2015; Orsmond & Kuo, 2011). Additionally, there has been an increase of time spent indoors and engaged in near-focused activities for younger members of the general population, which could explain the increased rate of myopia in our Control group (Dolgin, 2015; Ramamurthy et al., 2015). Thus, it is possible that the incidence rates of myopia in both the ASD and Control groups are a reflection of environmental factors regarding lifestyle choices.  While it has been established that some ethnicities are more likely to have refractive error, including higher rates of refractive error for individuals of East Asian descent (Foster & Jiang, 2014; Kleinstein et al., 2003; Pan et al., 2012), it is unlikely that these results can be explained by the ethnic background of our participants. Despite Vancouver, British Columbia having a large East Asian population of roughly 31.3 percent (Statistics Canada, 2016) only five of our 35 ASD participants (14.3 percent) were of East Asian descent. Four of our 35 Control participants (11.4 percent) were of East Asian descent. Thus, as each participant group had fewer members of East Asian descent than the surrounding population, we do not find evidence that the ethnic makeup of our participant groups could explain the higher rates of refractive error.   Given the wide variation in reports of visual acuity in adults in the general population, and the paucity of optometric reports in adults with ASD, careful, methodical investigations into basic visual function within this population are paramount if we are to draw meaningful conclusions on the topic. We report that there are comparable rates of myopia, hyperopia, and astigmatism between our ASD and Control groups. Additionally, rates of myopia for both groups were higher than previously reported prevalence in the general population. These results, along   92 with earlier reports of higher incidence of refractive errors in children with ASD, suggests that further analysis with a larger number of both children and adults is warranted.      93 Chapter 3: Visual orientation processing in ASD  The EPF suggests that enhanced low-level perceptual processing explains the superior performance with detail-oriented processing tasks commonly reported in ASD. This altered processing of visual information may occur upstream of the retina, possibly in the earliest cortical areas. As previously discussed, V1 is associated with processing of basic image properties like orientation and spatial frequency (De Valois et al., 1982; Enroth-Cugell & Robson, 1966). In order to examine whether there is enhanced processing at this early visual area, we chose three tasks to examine basic processing of visual orientation: specifically, (a) discrimination, (b) veridical perception, and (c) detection of orientation. Our tasks required only visual judgments, thereby avoiding potential pitfalls associated with tasks involving verbal labeling or explicit categorization of stimuli.  In order to tap into responses from the earliest visual processing areas, many studies choose to use “simple” stimuli. Unfortunately, there is a great deal of confusion in the literature as to what constitutes a truly “simple” stimulus. For instance, while images of shapes, like a square or a triangle, may appear to be simple, these stimuli are broadband in frequency and contain edges at various orientations in the same image. To study the properties of low-level spatial vision, it is typical to use stimuli that resemble the assumed visual receptive fields for that level (De Valois et al., 1982; Hubel & Wiesel, 1968; Tootell et al., 1981). We chose to use a Gabor pattern as our stimulus as they are localized in both space and frequency domains and are a common choice in visual experiments, allowing us to make meaningful comparisons with results of other studies.   In three experiments, we methodically assessed features of visual orientation processing across a range of base orientations in a group of adults with ASD and their neurotypical   94 counterparts. For each of the three experiments, manipulation of the base orientation afforded the opportunity to assess the status of the oblique effect- the robust finding of superior orientation processing of cardinal (vertical and horizontal) angles as compared to oblique angles (Appelle, 1972). Across numerous studies of orientation processing, the consistent finding of superior performance at cardinal orientation produces a “M” shaped performance curve. The presence, or lack thereof, of this characteristic response pattern allows us to make qualitative and quantitative comparisons between individuals with and without ASD. The general shape of the performance curve for each participant allowed us to determine if qualitative differences existed for the oblique effect. In addition, quantitative deviations were measured in terms of diminished or enhanced performance on the task as compared to neurotypical controls. Two main hypotheses were tested throughout the three experiments: (a) Performance patterns for the oblique effect in adults with ASD will be qualitatively distinct from controls (e.g. do not demonstrate superior performance at cardinal angles, thus lacking the oblique effect), and (b) Individuals with ASD are quantitatively distinct from controls and will demonstrate superior orientation perception at some or all base orientations.  3.1  General methods  Two groups of 29 adults with and without ASD participated in this study. All participants from the ASD and control groups completed Experiment 1; 16 of those participants from each group took part in Experiment 2; and 13 from each group who completed Experiment 1 also completed Experiment 3. The ethics review boards of the University of British Columbia, Simon Fraser University, and Vancouver General Hospital all approved the protocol. Informed consent was acquired before the experiments in agreement with the guidelines of the Declaration of Helsinki.    95  Verbal and nonverbal intelligence was assessed for all participants using the Wechsler Abbreviated Scale for Intelligence (Wechsler & Hsiao-pin, 2011). Participants with Full Scale IQ scores greater than 75 were included in the study. As we were assessing visual performance using nonverbal experimental tasks, we matched the participant groups on nonverbal IQ and age for each individual experiment (Burack, Iarocci, Flanagan, & Bowler, 2004). These groups were not matched on verbal IQ. In Experiment 1, the ASD group had significantly lower verbal IQ scores than the control group (p=0.02), but the difference was not significant for Experiment 2 (p=0.18) or Experiment 3 (p=0.09).  As discussed in Chapter 2, an optometrist confirmed normal or corrected-to-normal vision for all participants. The optometric screening included auto-refraction and manual refraction if necessary. If participants were wearing prescription glasses, they were measured using a lensometer to confirm they were the appropriate prescription. If a participant did not have the correct prescription, trial lenses were loaned for the duration of the experiment.   All participants completed the Autism Spectrum Quotient (AQ) questionnaire (Baron-Cohen et al., 2001). As described in Chapter 1, this 50-item self-report questionnaire is a measure of ASD symptoms; and it is suggested that individuals that score higher than 32 be referred to a specialist for ASD screening. We chose to exclude any control participants with scores higher than 20, as this is the point of greatest separation between adults with and without ASD while still allowing for individuals in the mathematics and sciences that traditionally score higher on the AQ (Baron-Cohen et al., 2001).  As all ASD participants (N=29) were adults, the majority had previously been diagnosed by a clinician under the Diagnostic and Statistics Manual of Mental Disorders, Fourth Edition (American Psychiatric Association, 2000). The Autism Diagnostic Observation Schedule (Lord,   96 Risi, Lambrecht, Cook, Leventhal, DiLavore, Pickles, et al., 2000) was used to verify a diagnosis of ASD for the majority of our participants (N = 23). All other participants had formal diagnoses confirmed via their referring clinician. For participants in the ASD group (N=29), comorbid psychiatric disorders included epilepsy/seizures (2), depression (1), attention-deficit/hyperactivity disorder (1), posttraumatic stress disorder (1), and obsessive-compulsive disorder (1). Both participants with seizures were taking prescribed seizure medications.  3.2 Experiment 1: Precision  We measured orientation discrimination thresholds for a range of base orientations spanning 180 degrees from the horizontal position. Using discrimination threshold measurements around eight base orientations, we were able to assess each participant's precision of orientation perception. In line with previous literature on the oblique effect, we expected cardinal angles (vertical and horizontal) would have higher precision (i.e. lower thresholds) compared to oblique orientations.  3.2.1 Methods  3.2.1.1       Participants Twenty-nine adults with ASD (eight women, 21 men, age = 23.2 ± 7.1 years) and 29 without ASD (nine women, 20 men, age 26.3 ± 7.8 years) participated in this experiment. The mean WASI-II Full Scale IQ scores for both groups were in the "Average" range, with scores for the ASD participants ranging from 76 to 134 (M=100.21 ± 14.99), while scores for the control group ranged from 77 to 134 (M = 108.72 ± 12.81). Both groups were matched on nonverbal IQ (p = 0.16) and age (p = 0.11). All participants had their diagnosis confirmed by a clinician, with the majority (N = 23) receiving ADOS.   97                   Table 3.1 Participant demographics for Experiment 1. Information for both participant groups (N = 29) is presented with the mean, standard deviation (SD), maximum (Max), and minimum (Min) for the following: Age, Block Design (BD) subtest T-score, Vocabulary (V) subtest T-score, Matrix Reasoning (MR) subtest T-score, Similarities subtest T-score, Verbal IQ (VIQ), Nonverbal IQ (NVIQ) score, and Full-Scale IQ (FIQ) score.  3.2.1.2       Experimental setup The experiments were displayed on a Sony Trinitron 17-in monitor (model Multiscan17sell) using a computer equipped with a Cambridge Research Systems (CRS; Rochester, UK) VSG 2/3 graphics card. Luminance calibration was performed via gamma correction using a CRS OptiCAL photometer (model OP200-E) and software provided by CRS. Mean luminance of the display was set to 17.4 cd/m2. The experiment was programmed in MATLAB (MathWorks, Inc.) and PsychToolbox (Brainard, 1997; Pelli, 1997). Participants had a viewing distance of 70 cm from the screen.  3.2.1.3 Stimuli and procedure We used a 3-cpd Gabor patch with a fixed Michelson contrast of 0.5 displayed at base orientations ranging from 22.5° to 180°. Trials were broken into eight experimental blocks by base (reference) orientation (22.5°, 45°, 67.5°, 90°, 112.5°, 135°, 157.5°, 180°) and completed in  Age BD V MR S VIQ NVIQ FIQ ASD         Mean 23.7 51.4 50.3 55.2 49.8 98.7 104.8 101.7 SD 7.1 12.5 10.6 9.3 12.3 15.5 16.6 14.4 Max  46 75 69 74 80 126 147 134 Min 17 32 24 36 26 63 79 79          Control         Mean 26.7 53.3 54.8 55.2 52.8 106.1 107.1 107.4 SD 8.0 9.4 8.5 9.2 9.6 12.7 13.9 12.1 Max  50 75 76 71 69 134 137 125 Min 18 36 34 36 34 76 81 77   98 a random order by each participant. Before being tested on a given reference orientation, a brief warmup of 10 trials was given to acclimate the participant to the new reference orientation. Warmup trials were also used to ensure that all participants understood the task instructions. Orientation increments at each trial were controlled by two randomly interleaved psychophysical staircases of 40 trials each. The Quest procedure (Watson & Pelli, 1983) in Psychophysics Toolbox was used to implement the staircases.  Trials began with a fixation cross presented for 150 ms, then a 150-ms presentation of the reference-orientation stimulus, another 150-ms fixation cross, followed by the test stimulus for 150 ms, and finally a blank screen that remained until the participant responded on a computer keyboard. Participants were asked to press 1 if the test stimulus was rotated to the left (counterclockwise) or to press 2 if it was rotated to the right (clockwise) compared to the reference-orientation stimulus (see Figure 3.1 for an illustration of the procedure). Correct responses were indicated via auditory feedback in the form of a single click; incorrect responses were indicated with two clicks.   Figure 3.1  Experiment 1: Protocol for measuring precision of orientation perception using an orientation discrimination task. A reference orientation Gabor was first displayed followed by a test orientation Gabor. Participants were then asked to say whether the test orientation was clockwise or counterclockwise compared to the reference orientation. We measured discrimination thresholds for eight reference orientations (22.5°, 45°, 67.5°, 90°, 112.5°, 135°, 157.5°, and 180°). Experiment 1: Precision (Task = discrimination)+ +ReferenceScreenTestScreenAnswerScreenWas it clockwise or counter-clockwise?FixationScreen150 ms 150 ms 150 ms 150 msFixation Screen  99   If the threshold estimates for a block varied too greatly from one another (defined as one staircase estimate being more than twice the value of the other), the block was repeated until consistent threshold estimates could be obtained. The groups did not significantly vary regarding the need to repeat blocks.  3.2.1.4      Data Analysis Two randomly interleaved staircases were used in each block to estimate orientation discrimination thresholds at a criterion accuracy of 82 percent. For each of the eight reference orientations tested, two staircase threshold estimates from a block were averaged to produce an overall threshold estimate for each reference orientation.  3.2.2 Results and discussion  Orientation discrimination thresholds are plotted as a function of reference orientation in Figure 3.2 for the group with ASD (red curve) and the control group (black curve). A repeated-measures ANOVA was used to analyze the thresholds, with reference orientation as the within-subject factor and group (ASD, control) as the between-subject factor. A significant main effect of reference orientation, F(7, 392) = 23.22, p << 0.001 was found. Post hoc Tukey-Kramer comparisons found that thresholds for all oblique orientations differed significantly from those at the cardinal angles (all ps < 0.05). In line with the oblique effect, thresholds at oblique orientations were significantly higher, while vertical and horizontal thresholds were superior and not significantly different from one another (all ps > 0.05). Our findings are consistent with previous studies of the oblique effect in orientation discrimination (De Valois et al., 1982; Essock, 1980; Westheimer & Beard, 1998).   100   Figure 3.2 Results of Experiment 1. Orientation discrimination thresholds are plotted as a function of base (reference) orientation for the group with ASD (red curve) and the neurotypical control group (black curve). Consistent with previous findings, precision was higher at cardinal orientations (indicated by lower discrimination thresholds) compared to oblique orientations (with significantly higher thresholds). No group differences were found in precision for orientation perception.  There was no main effect for group, F(1, 56) = 0.00, p = 0.97, and no significant interaction between group and reference orientation (F(7, 392) = 0.41, p = 0.90.). An absence of group effects indicates there are no quantitative differences as the group with ASD did not perform better or worse than the control group. Additionally, as both groups showed a characteristic oblique effect, the absence of interaction effects indicate a lack of qualitative differences.    ASDControlBase orientation (deg)0 22.5 45 67.5 90 112.5135157.5180Discrimination threshold (deg)0 1 2 3 4 5 6 7 8 9 10  101 3.3 Experiment 2: Veridical perception  Examining the ability to discriminate around various reference orientations in Experiment 1 enabled us to assess a participant's sensitivity to fine-grain deviations around those reference orientations. It did not, however, establish whether the perceived orientation at those base angles was accurate. In Experiment 2, we asked participants to manually adjust the orientation of a Gabor stimulus to a given target orientation. By finding the central tendency of repeated settings to a specified target orientation across trials, we could infer the accuracy of orientation perception at those orientations while also examining any systematic biases away from veridical perception. We could also infer the precision of perception by measuring the variability of a participant's settings across trials for a given orientation.  3.3.1 Methods  3.3.1.1      Participants All adults with ASD who participated in Experiment 1 were invited to participate in Experiment 2; sixteen of the original 29 agreed to return to the lab for this additional experiment. Sixteen adult participants with ASD (four women, 12 men, age = 24.9 ± 8.7 years) and 16 adult controls (five women, 11 men, age = 24.9 ± 4.5 years) participated in both Experiments 1 and 2. WASI-II Full Scale IQ scores for participants with ASD ranged from 79 to 129 (M = 100.88 ± 14.20) and the control group ranged from 85 to 123 (M = 108.81 ± 10.99). The groups were matched on nonverbal IQ (p = 0.2) and age (p = 0.98). All participants had clinical diagnoses and confirmation via ADOS was obtained for the majority (N = 10) of our participants.      102                   Table 3.2 Participant demographics for Experiment 2. Information for both participant groups (N = 16) is presented with the mean, standard deviation (SD), maximum (Max), and minimum (Min) for the following: Age, Block Design (BD) subtest T-score, Vocabulary (V) subtest T-score, Matrix Reasoning (MR) subtest T-score, Similarities subtest T-score, Verbal IQ (VIQ), Nonverbal IQ (NVIQ) score, and Full-Scale IQ (FIQ) score.  3.3.1.2.     Experimental setup This experiment was presented on a Dell laptop computer (model 3750) with a 17-in. antiglare LED screen. The experiment was programmed in SuperLab version 5.0 (www.superlab.com).   3.3.1.3      Stimuli and procedure Sine-phase Gabor patches at a fixed Michelson contrast of 0.8 were displayed on a constant grey background of 33 cd/m2 luminance. Depending on participant comfort, the viewing distance ranged from 50 to 70 cm, which corresponded to a spatial frequency range of 3 to 4 cpd for the Gabor patch.  Using a method-of-adjustment paradigm, we asked participants to manually adjust (via key press) the orientation of a Gabor patch to a specified target angle (vertical, 90°; horizontal, 180°; oblique, 45°). Each block consisted of 40 trials of a specific target orientation. Trials began with a fixation cross, followed by a Gabor at a random starting orientation. Participants were  Age BD V MR S VIQ NVIQ FIQ ASD         Mean 24.9 48.0 51.3 54.2 49.2 100.2 102.3 100.9 SD 8.7 13.1 9.8 9.8 10.3 14.2 17.4 14.2 Max  46 75 69 68 65 126 139 129 Min 17 32 35 36 29 74 79 79          Control         Mean 24.9 53.9 55.0 56.8 53.3 106.6 108.8 108.8 SD 4.5 7.8 8.0 6.3 10.4 12.5 9.0 11.0 Max  38 71 76 68 69 128 118 123 Min 19 36 45 46 34 87 87 85   103 instructed to press 1 on the number pad to rotate the stimulus counterclockwise or 2 to rotate it clockwise. Once they had adjusted the Gabor to the specified target orientation, they were asked to indicate they had finished the trial via key press. No feedback was given. Prior to being tested at each orientation, participants were given instructions and completed three warmup trials with the examiner present to answer any questions. Task comprehension was demonstrated during performance on the warmup trials. Figure 3.3 illustrates the experimental procedure.    Figure 3.3  Experiment 2: protocol for measuring accuracy of perceived orientation using an adjustment task. Participants were asked to manually adjust (via key press) a randomly oriented Gabor to a specified target orientation (vertical, 90°; horizontal, 180°; or oblique, 45°).  3.3.1.4      Data analysis We computed error for each trial as the difference between the target orientation and the participant's final orientation setting for the Gabor. For each target orientation, absolute error, or absolute difference from the target orientation, was calculated as the absolute value of the average error across all 40 trials in the specific block. We also computed the sample standard deviation across the 40 trials in each block. Together, the measures of absolute mean error and standard deviation of error allowed us to assess veridicality of perception (former) and precision (latter) for each target orientation. Experiment 2: Accuracy (Task = adjustment)+Press enter to accept your final settingRandom starting orientationSubjectivesetting ofverticalInstruction:make it verticalManual Adjustmentvia keypress  104  3.3.2 Results and Discussion  Average absolute error across participants is presented as a function of target orientation in Figure 3.4A. Veridical perception is determined by the central tendency of settings for a given target orientation. The overall net bias in settings signifies deviations from veridical perception. This is calculated as absolute mean error for each participant as a function of target orientation.  As shown in Figure 3.4A, both participant groups were near veridical in their perception of vertical and horizontal orientations. However, substantial deviations from veridicality, or biases, were shown at the diagonal orientation, which is consistent with the oblique effect. Absolute mean error was submitted to a repeated-measures ANOVA, with target orientation (45°, 90°, 180°) as a within-subject factor and group (ASD, control) as a between-subjects factor. A significant main effect was found for target orientation, F(2,60) = 40.49, p << 0.0001, but there was no main effect for group, F(1,30) = 0.14, p = 0.71, and no interaction between group and target orientation, F(2,60) = 0.11, p = 0.90. We used post-hoc comparisons and found that absolute error at the 45° orientation (M = 4.75) was significantly greater than the absolute error at 0° (M = 0.27) and 90° (M = 0.38) orientations (Tukey-Kramer, both ps < 0.05), while the cardinal-orientation conditions did not significantly differ from one another (Tukey-Kramer, p > 0.05).    105  Figure 3.4 Results of Experiment 2. (A) Absolute mean error are averaged for each group and plotted as a function of target orientation for the group with ASD (red curve) and the control group (black curve). Orientation perception was near veridical for vertical and horizontal orientations, however, significant biases were found for the oblique orientation. No differences were observed between groups. (B) Group averages of standard deviation of orientation settings are plotted as a function of target orientation for the group with ASD (red curve) and the control group (black curve). Inversely related to the standard deviation, the precision of orientation perception was high for vertical and horizontal orientations, but was significantly decreased for the oblique orientation. No group differences were observed.   While the central tendency of a participant’s settings is related to veridicality, the variability of those settings around the central tendency is a measure of precision. It is possible, therefore, that an individual may have veridical but imprecise perception and vice versa. As previously mentioned in Experiment 1, discrimination thresholds are the gold standard for assessing precision. The current experiment allows for an additional estimate for precision using the participant’s variability of orientation settings, and provides an opportunity to verify the results of Experiment 1. Inversely related to precision, the standard deviation of each individual’s settings was averaged across participants and is presented as a function of target orientation in Figure 3.4B. As predicted by the results from Experiment 1, and consistent with the literature regarding the oblique effect, precision was highest for cardinal orientations and lowest for the oblique target orientation. Figure 3.4B shows that both the group with ASD and BIAS PRECISIONA.) B.)Target orientation (deg)0 45 90Absolute error (deg)12345678ASDControlTarget orientation (deg)0 45 90    Standard deviation of perceived orientation (deg)1 2 3 4 5 6 7 8 9 10ASDControl  106 the control group demonstrate the oblique effect, and lack any obvious qualitative or quantitative differences. In order to statistically confirm these observations, we submitted the standard deviations of orientation settings to a repeated-measures ANOVA, with target orientation (45°, 90°, 180°) as a within-subject factor and group (ASD, control) as a between-subjects factor. There was a significant main effect of orientation, F(2,60) = 68.27, p << 0.001, but no main effect of group, F(1,30) = 0.01, p = 0.93, and no interaction between the two, F(2,60) = 0.21, p = 0.81. Post hoc comparisons showed that the standard deviation of settings at the oblique 45° orientation (M = 6.72) was significantly greater than those at 0° (M = 0.74) and 90° (M = 0.81) orientations (Tukey-Kramer, both ps < 0.05), while the vertical and horizontal orientations did not significantly differ from each other (Tukey-Kramer, p > 0.05).  The lack of a group main effect and a lack of interaction for the measures of bias and precision suggest that there are neither quantitative nor qualitative differences between the group with ASD and the control group.  3.4 Experiment 3: Sensitivity  We measured contrast thresholds for detecting a Gabor patch as a function of stimulus orientation. We predicted that we would replicate the oblique effect, with higher sensitivity (i.e., lower thresholds) around vertical and horizontal orientations compared to oblique orientations. We chose to use a high-spatial-frequency Gabor stimulus because the oblique effect is more pronounced at higher spatial frequencies (Berkley, Kitterle, & Watkins, 1975; Campbell et al., 1966; Essock & Lehmkuhle, 1982; Lennie, 1974; Mitchell, Freeman, & Westheimer, 1967).      107 3.4.1 Methods  3.4.1.1     Participants All adults participating in this study had taken part in Experiment 1, thirteen with ASD (four women, age = 24.1 ± 7.2 years) and 13 without (five women, age = 23.6 years ± 3.1 years). WASI-II Full Scale IQ scores for the group with ASD were between 79 to 119 (M = 99.00 ± 13.25), while the IQ scores for the control group were between 98 to 123 (M = 109.92 ± 8.46). The groups were matched on nonverbal IQ (p = 0.11), and age (p = 0.83).                    Table 3.3. Participant demographics for Experiment 3. Information for both participant groups (N = 13) is presented with the mean, standard deviation (SD), maximum (Max), and minimum (Min) for the following: Age, Block Design (BD) subtest T-score, Vocabulary (V) subtest T-score, Matrix Reasoning (MR) subtest T-score, Similarities subtest T-score, Verbal IQ (VIQ), Nonverbal IQ (NVIQ) score, and Full-Scale IQ (FIQ) score.  3.4.1.2       Experimental setup We implemented this experiment on a computer equipped with a CRS VSG 2/3 graphics card and a Sony Trinitron 17-in. monitor (model Multiscan17-seII). A CRS OptiCAL photometer (model OP200-E) and software provided by CRS were used for gamma correction. The display’s mean luminance was 17.4 cd/m2. We programmed this experiment in MATLAB using tools from  Age BD V MR S VIQ NVIQ FIQ ASD         Mean 24.1 45.7 51.2 53.9 48.0 99.2 99.9 99.0 SD 7.2 11.8 10.6 8.4 10.9 15.2 14.5 13.3 Max  40 68 69 68 65 126 132 119 Min 17 32 35 40 29 74 79 79          Control         Mean 23.6 55.1 56.1 55.0 55.3 109.1 108.3 109.9 SD 3.1 8.3 8.4 7.0 10.3 12.7 10.8 8.5 Max  29 71 75 68 69 134 118 123 Min 19 37 45 40 39 91 81 98   108 CRS VSG Toolbox for MATLAB and Psychophysics Toolbox. Participants were seated at a distance of 77 cm from the screen.  3.4.1.3      Stimuli and procedure A 22 cpd sine-phase Gabor was used for this experiment at vertical (90°), horizontal (180°), and oblique (45° and 135°) orientations.   Participants were asked to detect a Gabor in a two-interval forced-choice task. Trials began with a 150-ms fixation cross, then Interval 1 for 150 ms, another 150-ms fixation cross, followed by Interval 2 for 150 ms, and finally a blank screen that remained until the participant entered a response. Participants were instructed to press 1 or 2 on the keyboard if the Gabor was presented during the corresponding interval. We used an auditory beep to mark each interval in order to prevent confusion between the two. A single auditory click was presented as feedback for correct answers, while two clicks were used as feedback for incorrect responses. A new trial was presented immediately following the participant’s response. We asked participants to remain fixated on the center of the screen throughout each trial. For each orientation tested, a single warmup block of 10 trials preceded the experimental block. A schematic illustration of the task is presented in Figure 3.5.     109  Figure 3.5  Experiment 3: Protocol for measuring orientation sensitivity using a detection task. A Gabor stimulus (in one of four orientations: vertical, 90°; horizontal, 180°; or oblique, 45° and 135°) was displayed in one interval and a blank screen in the other. Participants were asked to indicate which interval contained the stimulus. A psychophysical staircase controlled the contrast of the Gabor stimulus, allowing for estimation of detection contrast thresholds at each orientation.  We blocked all trials by orientation, and participants completed the four orientation blocks (vertical, 90°; horizontal, 180°; oblique, 45° and 135°) in a random order following a brief warmup. Each block utilized two randomly interleaved staircases in order to estimate two independent thresholds for detection of the Gabor patch at that block’s specific orientation. The QUEST procedure in Psychophysics Toolbox was used to implement each staircase. Each staircase contained 40 trials, for a total of 80 trials in each block.  If the threshold estimates of a particular block were deemed to be too far apart from one another (defined as one value being more than double the other), the block was discarded and repeated until threshold criterion were met. The group with ASD and control group did not differ in their need to repeat blocks.  3.4.1.4       Data analysis Each reference orientation (45°, 90°, 135°, and 180°) was tested as a separate block. Each block used two randomly interleaved staircases to estimate contrast detection thresholds, with an Experiment 3: Sensitivity (Task = detection)+ +AnswerScreenInterval 1 Interval 2Was it on the 1st or 2nd interval?FixationScreen150 ms 150 msFixationScreen150 ms 150 msBlankScreen  110 accuracy criterion of 82 percent. The two staircase estimates from a block were averaged to produce a single threshold estimate for that block’s orientation.  3.4.2 Results and Discussion  Contrast threshold values for detecting a Gabor stimulus as a function of orientation are plotted on a log axis and are presented in Figure 3.6, with the group with ASD represented in red and the control group shown in black. We submitted the log contrast thresholds to a repeated-measures ANOVA, with orientation (vertical, 90°; horizontal, 180°; oblique, 45° and 135°) as a within-subject factor and group (ASD, control) as a between-subjects factor. There was a highly significant main effect of orientation F(3, 72) = 23.11, p << 0.001. No main effect was found for group F(1, 24) = 1.70, p = 0.21 and no interaction between group and orientation, F(3, 72) = 0.15, p = 0.93.    111  Figure 3.6 Results of Experiment 3. Detection contrast thresholds are plotted as a function of orientation, with 90° being vertical and 180° is horizontal (data at 0° is also horizontal and is identical to data plotted at 180°). The group with ASD (red curve) demonstrates slightly less sensitive performance compared to the control group (black curve), but this difference was not significant.  The oblique effect, or a characteristic “M” pattern observed here, was qualitatively demonstrated by both participant groups-i.e., detection contrast thresholds were lower for cardinal angles compared to oblique orientations. This observation of qualitatively-similar results demonstrated via the oblique effect is supported by the lack of an interaction between participant group and orientation. Quantitatively, no evidence was found that individuals with ASD show superior performance in orientation detection at any orientation. The group with ASD demonstrated a trend in the opposite direction, with slightly lower sensitivity across all four orientations tested.  Thresholds at the four orientations were compared using post hoc pair-wise comparisons, the results of which showed that thresholds were significantly lower at the vertical orientation Orientation (deg)0 45 90 135 180Detection contrast threshold0.10.20.40.8ASDControl  112 than all other orientations (Tukey-Kramer multiple-comparison test, all ps < 0.05). While horizontal detection thresholds were also lower than thresholds at the oblique orientations, this difference was not significant.   The group with ASD did not demonstrate qualitative or quantitative differences in contrast detection thresholds for the four tested orientations, as suggested by the lack of a group effect and the lack of an interaction.  3.5 General Discussion  Individuals with ASD have demonstrated superior performance in a variety of visual tasks and paradigms (Gliga et al., 2015; Happe, 1999; Jolliffe & Baron-Cohen, 1997; Kéïta et al., 2014; Manning et al., 2015; O'Riordan et al., 2001; Plaisted et al., 1998a; Plaisted et al., 1998b; Plaisted et al., 2003; Plaisted et al., 1999; Shah & Frith, 1983). Despite the multitude of studies, the precise source of altered perception remains unclear. A clear picture of the factors driving the enhanced visual performance could provide invaluable insight into the brain processes that contribute to the complex symptoms associated with ASD.  This study is one part of a line of research seeking to systematically assess whether individuals with ASD show enhanced visual cortical processing at the earliest stages of vision. For this study, we chose to focus on basic orientation processing, an aspect of visual processing long associated with V1. Our approach included several important methodological choices; beginning with recruitment of a large number of participants to allow the findings to be more generalizable to the ASD population as a whole. Additionally, we ensured all participants were carefully assessed for an accurate diagnosis and matched the subjects on age and non-verbal IQ. We also confirmed normal or corrected-to-normal visual acuity via consultation with an   113 optometrist. Next, we selected a truly simple stimulus by using a Gabor patch and chose three basic visual tasks to isolate responses from the earliest levels of visual processing as much as possible. All three tasks were conducted with the same subsets of participants, allowing for qualitative comparisons of the results across experiments, (of the 29 participants who completed Experiment 1, 16 also completed Experiment 2, and 13 also completed Experiment 3). Finally, these three experiments were designed to measure performance across various base orientations, allowing us to compare the two groups at multiple data points. We utilized the well-established finding of the oblique effect, where superior performance is found for cardinal base orientations compared to oblique ones. The rich literature describing the oblique effect allowed us to make predictions and compare the performance of each participant group both qualitatively and quantitatively. The absence of statistically significant differences between the groups supports the interpretation that the oblique effect is both present and typical for individuals with ASD compared to controls. Additionally, obtaining a rich collection of data across three experiments allowed us to systematically assess whether any differences were present between groups. These data allowed us to test our two main hypotheses: (a) Individuals with ASD are qualitatively distinct from controls and show an altered pattern of visual perception that differs from the oblique effect, and (b) individuals with ASD are quantitatively distinct from individuals without ASD and show superior orientation perception. After thorough analysis, we conclude that neither hypothesis is supported.  Our findings are consistent with the Brock et al. (2011) study of neurotypical individuals that assessed orientation discrimination around a horizontal reference and found no correlation between discrimination thresholds and AQ scores. These results are particularly relevant when viewed in context of the larger experiment, as individuals with higher AQ scores from that same   114 group of participants showed significantly faster reaction time in a search task using the same stimuli as in the discrimination experiment. In contrast, a later study by Dickinson et al. (2014) found that within a large group of neurotypical individuals, higher AQ scores were correlated with lower discrimination thresholds around one oblique orientation tested, but there was no association at the vertical orientation. The authors suggested the reason for lack of a relationship stemmed from ceiling effects, as a hallmark of the oblique effect is the dramatic improvement in perception around cardinal axes. These results are inconsistent with the results of Experiment 1 in our study, as we also measured orientation discrimination thresholds using relatively similar methods. Instead of a single oblique angle and the vertical axis, we looked at both cardinal and four separate oblique orientations. No evidence was found for enhanced perception for any of the base orientations tested in our study. There are several potential explanations for the differences between the findings of Dickinson et al. and our own, including methodological differences and participant group selection. A possible methodological reason behind the different findings was that sinusoidal gratings were used as the stimuli for the Dickenson et al. study, while both the Brock et al. (2011) and the present study used Gabor patches. While the sine-wave grating used in the Dickenson et al. study could theoretically contain power at single spatial frequency (specifically 3 cpd for the Dickenson et al. study), a stimulus would have to cover an infinite spatial extent. By presenting this stimulus on a limited aperture with sharp edges, like the computer screen used in the Dickenson et al. study, high spatial frequencies were introduced to the stimulus image. Recent findings in the literature suggest that high-spatial-frequency components could enable enhanced performance (e.g., Kéïta et al., 2014; Latham et al., 2013). In contrast, the Gabor stimulus chosen for the present study is localized in both spatial and spatial-frequency domains.   115 This ensures the spatial frequency tested is strictly focused at the specified spatial frequencies. Regarding participant group selection differences, the Dickenson et al. study tested non-clinical neurotypical individuals while using AQ as a measure of traits related to ASD. Our study, however, tested individuals who had been clinically diagnosed with ASD and compared them to a control group without ASD and with AQ scores of 20 or less. One important qualification here is that while the AQ is associated with autistic traits, it is not a diagnostic tool and cannot determine whether an individual does or does not have ASD. Another important subject selection difference between our study and Dickenson et al. was that general intelligence was not assessed in their participants.  It is possible that general intelligence may be fundamental in determining whether enhanced processing is observed for an individual with ASD (Caron et al., 2006). Bertone et al. (2005) asked participants to identify whether a given stimulus was vertical or horizontal. They reported that contrast thresholds for participants with ASD were superior to their neurotypical counterparts, which is in contrast to our own study. Important methodological differences, however, may explain the discrepancy. This orientation-identification task used a sine-wave grating in noise, in contrast to our own study, which used Gabors. Another methodological difference is that identification tasks are potentially higher-order visual tasks compared to all three of the more basic orientation tasks we used in our study. Possibly the most salient reason behind our contradictory findings, however, could be the striking difference between the clinical groups who participated in each study. In Bertone et al., 83% of the participants with ASD had a relative Block Design test peak (as indicated in Caron et al. 2006, pg. 1800). As described in Chapter 1, the Block Design task is one subtest of the WASI-II that assesses aspects of nonverbal intelligence. Individuals are asked to manually rearrange blocks to replicate an image. The term   116 “Block Design test peak” (or BDT-peak), describes a profile in which the Block Design test was the highest score of WASI-II subtests. While early indications suggested a strong association of BDT-peak and ASD, the actual rate of BDT-peak for adults with ASD with average intelligence was shown to be just 22% (Siegel, Minshew, & Goldstein, 1996). These results are consistent with the findings of our own study, as seven of our 29 participants, or 24% of our sample with ASD, had a BDT-peak (defined as having the highest score in the Block Design subtest relative to the other subtests).  To examine whether enhanced perception of low-level visual stimuli is associated with superior performance in the Block Design test, we reanalyzed our data after separating the ASD group into two new subgroups: “BDT-peak” and “No BDT-peak.” Figure 3.7 shows results from each of the three experiments with the BDT-peak participants in red and no BDT-peak individuals in blue; Figure 3.7A shows the results of Experiment 1 and that BDT-peak individuals demonstrate superior orientation discrimination thresholds compared to both controls and the no BDT-peak ASD group; Figure 3.7B shows the results from Experiment 2, where BDT-peak individuals show a trend for superior bias and precision at oblique angels compared to the other two groups; and Figure 3.7C shows that there is no discernible trend for differences in detection thresholds for any of the groups. These analyses all failed to reach statistical significance due to the low number of participants who demonstrate a BDT-peak. This provides a possible explanation for the discrepancy between our findings and those of enhanced orientation processing in Bertone et al. (2005) study. As the present study was not designed to specifically assess BDT-peak in ASD in relation to orientation processing, we cannot provide strong evidence to support or refute the BDT-peak relationship to enhanced visual perceptual abilities. The lack of large participant numbers in Experiment 2 and 3 (three and two BDT-peak   117 individuals, respectively), make it difficult to extrapolate these findings to the ASD population as a whole. It is worth noting that a later study by Meilleur, Berthiaume, Bertone, and Mottron (2014) with a different group of participants used the same task and was unable to replicate the findings of the Bertone et al. (2005) study. Once again, the specific makeup of a participant group with ASD continues to be of upmost importance when comparing findings from different studies. Given the 22% incidence rate of a BDT-peak in the adult population with ASD, future studies of enhanced perceptual processing within this group must utilize measures of IQ to ensure use of truly representative samples of each group. Future research of perceptual processing in individuals with ASD should assess the BDT-peak as a factor.   118  Figure 3.7. Data from participants with ASD who have a Block Design test peak. After separating the group with ASD into two groups based on whether they had a BDT-peak (red curve) or not (blue curve), data from each of the three experiments are replotted. This post hoc observation lacked statistical power due to the low participant number. The results of Experiments 1 (A) and 2 (B) suggest that the presence of a BDT-peak could be an important factor in whether superior perceptual processing is observed in individuals with ASD, though this was not observed in Experiment 3 (C).  Base orientation (deg)0 22.54567.590112.5135157.5180Discrimination threshold (deg)0 1 2 3 4 5 6 7 8 9 101112A) Experiment 1: PrecisionB) Experiment 2: AccuracyBias PrecisionC) Experiment 3: SensitivityASD - BDT peak (N=7)ASD - No BDT peak (N=22)ControlASD - BDT peak (N=2)ASD - No BDT peak (N=11)ControlTarget orientation (deg)Absolute error (deg)123456780 45 90Target orientation (deg)     Standard deviation of perceived orientation (deg)1 2 3 4 5 6 7 8 9 100 45 90ASD - BDT peak (N=3)ASD - No BDT peak (N=13)ControlOrientation (deg)0 45 90 135 180Detection contrast threshold0.10.20.40.8  119  Ours is not the first study to suggest that enhanced perceptual functioning is confined to certain subgroups of individuals with ASD. Bonnel et al. (2010) used simple tones in a pitch discrimination task and found that individuals with autism had superior performance while those with Asperger’s syndrome did not. Whether a BDT-peak is present may be associated with speech delay, a major determinant when diagnosing individuals on the spectrum while DMS-IV was still in effect (Ehlers et al., 1997). The current study does not provide any insight into this issue, which suggests future studies are needed to examine the relationship between BDT-peak, perceptual functioning, and speech delay in ASD.  In the present study, the findings do not support the hypothesis that within the general population of individuals with ASD, enhanced perceptual processing involves the lowest levels of visual orientation processing. This conclusion is supported by the recent fMRI findings by Schwarzkopf et al. (2014), which characterized response selectivity for the human visual cortex and found larger receptive fields in the extrastriate regions (V2, V4, and MT), but not V1, for the participant group with ASD. Additionally, a recent meta-analysis found that individuals with ASD tend to have atypical fMRI response patterns that cluster around occipital and temporal regions associated with expertise processing of visual stimuli, whereas there were no significant differences between groups in the lower-level visual areas (Samson, Mottron, Soulières, & Zeffiro, 2012). Taken together, these results indicate that the altered perceptual abilities associated with ASD may be related to other aspects of visual processing in V1, such as spatial frequency, or alternatively atypical perception may originate further along in the visual pathway. Given the lack of scientific consensus for many perceptual tasks of visual processing in ASD (see Behrmann, Thomas, et al., 2006; Coulter, 2009; Simmons et al., 2009), carefully   120 designed studies seeking to provide insight to the most intricate aspects of visual performance via meticulously designed psychophysical experiments are necessary to conclusively determine which levels of visual processing are enhanced or unremarkable in ASD. The present study, while specifically focused on visual orientation processing, provides insight into this discussion, and offers evidence against the hypothesis that orientation processing at the earliest levels of the visual pathway are the source of superior perceptual processing in ASD. Our conclusion is consistent with recent interpretations of the EPF model (Mottron et al., 2006). Meilleur et al. (2014) found atypical perception was associated with ASD-specific perceptual performance across various levels of visual processing. Further research is necessary to determine whether enhanced perceptual processing is associated with subgroups of individuals with ASD, including studies comparing results between those who do and do not show a BDT-peak.           121 Chapter 4: Association of social competence with face processing abilities in adults with ASD  According to the social motivation hypothesis of ASD, individuals on the spectrum struggle with processing face identity and expressions as a result of having less visual experience with faces (Dawson, Webb, & McPartland, 2005; Schultz, 2005). This model of face processing in ASD suggests that a lack of motivation to attend to faces results in less visual exposure and impedes the typical development of face processing skills (Chevallier et al., 2012). Face processing skills develop throughout childhood and into adulthood (Germine, Duchaine, & Nakayama, 2011; Meinhardt-Injac, Persike, Imhof, & Meinhardt, 2015). It is thought that the quality and frequency of exposure to faces contribute to the development of expertise (Mondloch, Le Grand, & Maurer, 2003). If individuals with ASD are not looking at faces and therefore have less experience, it is plausible that their lack of exposure might be driving their difficulty with developing face processing skills. Face recognition skills are important for successful social interactions (Dalrymple et al., 2014; Yardley, McDermott, Pisarski, Duchaine, & Nakayama, 2008). For instance, individuals with prosopagnosia indicate that an inability to recognize an employer negatively impacts professional relationships. It is also problematic for social interactions if an individual is unable to tell how another person is feeling based on their facial expressions, as is the case with functional outcomes for individuals with schizophrenia and social affect recognition impairments (Poole, Tobias, & Vinogradov, 2000). Personal relationships could be negatively impacted if an individual does not correctly interpret when another person is feeling sad or angry. For individuals with ASD, it is possible that social impairments are driving face processing deficits, or conversely, that face processing issues are causing or contributing to social impairments. Finally, these two possibilities are not mutually exclusive, and may co-occur   122 simultaneously. Regardless of the exact direction of causality, a common feature is that face abilities and some aspect of social skills should be associated. In the present study, we assessed both identity and expression performance in adults with ASD. We chose to use simple visual tasks to reduce the likelihood of confounds due to verbal labeling or explicit labeling of stimuli. Additionally, we assessed seven domains of social competency using the Multidimensional Social Competence Scale (MSCS), thereby allowing us to examine potential relationships between face processing ability and aspects of social competency.  Face abilities were assessed in the identity domain and expression domain. In the identity domain, we evaluated face-specific perception and face-specific memory. Face-specific perception was assessed via identification contrast thresholds for faces vs. houses; and face-specific memory was assessed using the Cambridge Face Memory Test (CFMT) (Duchaine & Nakayama, 2006). In the expression domain, we assessed discrimination ability for angry, happy, and sad expressions. Severity of traits associated with ASD was assessed using the Autism Spectrum Quotient (AQ) (Baron-Cohen et al., 2001). Social competence was evaluated using the MSCS for each individual (Yager & Iarocci, 2013). These measures provided the opportunity to compare social competence with face identification and expression discrimination abilities within the same group of participants. This allowed us to look for differences in the measures between groups and examine the association between face processing abilities and social competence within each group. We tested two main hypotheses: (a) face identification and (b) face expression discrimination impairments are associated with social competence issues in adults with ASD.   As social competence issues are a core feature of ASD, it is possible that face processing abilities are also associated with symptom severity. On the other hand, ASD symptoms go   123 beyond the social aspects and include other features such as stereotyped movements and restricted and repetitive behaviours, which one would not expect to be connected with face abilities. In addition, social competence is not a unitary skill, but is comprised of several distinct aspects of social competence, such as “social knowledge” and “nonverbal sending skills”. In this study, we sought to examine the relationship between face processing abilities, social competence, and ASD symptom severity to determine whether meaningful associations could be found between these distinct features of ASD.  4.1  General methods  Two groups of 27 adults, with and without ASD, completed this study. The ethics review boards of the University of British Columbia, Simon Fraser University, and Vancouver General Hospital all approved the protocol. Informed consent was obtained in agreement with the Declaration of Helsinki. All subjects were naïve to the purposes of the study.  All participants were assessed for verbal and non-verbal intelligence using the WASI-II (Wechsler & Zhou, 2011). We included participants with a FIQ greater than 75 to prevent confounds due to intellectual disabilities. As our face tasks were nonverbal measures of visual performance, we chose to match the two groups on nonverbal IQ (Burack et al., 2004). The groups were also matched on age (p = 0.32), but were not matched on verbal IQ (p < 0.05) or gender (ASD: 10 females, Control: 11 females).   As described in Chapter 2, an optometrist performed a visual examination of all participants in order to confirm normal or corrected-to-normal vision. The eye exam began with an auto-refraction, followed by a manual refraction if necessary. Participants with prescription glasses had the prescription confirmed via a lensometer. Any participants without the correct   124 prescription were loaned a set of trial lenses for the duration of the experiment (1 ASD, 1 Control).  All participants completed the AQ questionnaire (Baron-Cohen et al., 2001). We chose to include control participants with an AQ score below 20, as this is the point of greatest separation between individuals with and without ASD while still allowing for neurotypical individuals who may score higher than average, such as those in science and mathematics fields (Baron-Cohen et al., 2001).   As previously explained in Chapter 1, the MSCS provides an assessment of seven distinct domains of social competency (Yager & Iarocci, 2013). This self-report questionnaire provides competency scores in social motivation, social inferencing, demonstrating empathic concern, social knowledge, verbal conversation skills, nonverbal sending skills, and emotion regulation. The social motivation domain assesses how much interest and general enjoyment an individual has when interacting with others. Social inferencing refers to the ability to detect and interpret social cues. Demonstrating empathic concern requires an individual to recognize when another person is upset and respond appropriately. Individuals with social knowledge are aware of social rules and the appropriate response in specific social situations. Verbal conversation skills involve the ability to maintain an equal exchange and end it appropriately. Nonverbal sending skills require the ability to send nonverbal signals, such as using pointing, gestures, eye contact, or facial expressions to communicate with another individual. Finally, the emotion regulation domain assesses the ability to control emotional, extreme reactions or outbursts in frustrating situations. The MSCS measure does not have a cutoff recommendation for ASD, but individuals with ASD tend to score lower than their neurotypical counterparts.    125   Diagnosis was confirmed for the majority of participants (N = 23) using the Autism Diagnostic Observation Schedule (ADOS) (Lord, Risi, Lambrecht, Cook, Leventhal, DiLavore, & Rutter, 2000). The participants with ASD self-reported the presence of any psychiatric disorders, including epilepsy/seizure (2), depression (1), attention-deficit/hyperactivity disorder (1), and obsessive-compulsive disorder (1). Both participants with seizures reported taking seizure medication.  4.1.1 Participants  27 adult participants with ASD (10 female, age = 23.9, SD=7.8 years old) and 27 without ASD (11 female, age = 25.9, SD=6.2 years old) took part in all the experiments. WASI-II Full Scale IQ scores for both groups were in the “Average” range, with mean nonverbal IQ scores of 103.0, SD=17.6 for the group with ASD and 108.6, SD=8.3 for the controls. All participants had normal or corrected-to-normal vision. All participants had clinical diagnoses of ASD and the majority (N = 21) had their diagnosis confirmed via ADOS.            126                   Table 4.1. Participant demographics for all face experiments. Information for both participant groups (N = 27) is presented with the mean, standard deviation (SD), maximum (Max), and minimum (Min) for the following: Age, Block Design (BD) subtest T-score, Vocabulary (V) subtest T-score, Matrix Reasoning (MR) subtest T-score, Similarities subtest T-score, Verbal IQ (VIQ), Nonverbal IQ (NVIQ) score, and Full-Scale IQ (FIQ) score.  4.1.2 Experimental setup  We used a computer equipped with a Cambridge Research Systems VSG 2/3 graphics card and SONY Trinitron 17-in. monitor (model Multiscan17seII). Gamma correction was carried out using a CRS OptiCAL photometer (model OP200-E) and software provided by CRS. Mean luminance of the display was 17.4 cd/m2. The experiment was programmed using Matlab (www.mathworks.com) and tools from the CRS VSG Toolbox for Matlab and Psychophysics Toolbox (Brainard, 1997; Pelli, 1997). Participant viewing distance was 70 cm from the computer screen.  4.2  Experiment 1: perception for facial identity  We measured identification contrast thresholds for faces, i.e. the minimum amount of contrast necessary to correctly identify one face among a set of alternatives (X. M. Guo et al., 2009; Oruc & Barton, 2010a, 2010b). Participants were also asked to complete a non-face control task using  Age BD V MR S VIQ NVIQ FIQ ASD         Mean 23.9 50.4 51.3 54.0 51.5 100.7 103.0 102.1 SD 7.8 12.7 10.5 10.1 11.3 14.3 17.6 15.2 Max  47 75 72 74 80 134 147 134 Min 17 32 35 36 29 74 77 79          Control         Mean 25.9 55.0 57.7 55.3 59.8 114.0 108.6 113.1 SD 6.2 5.9 9.2 6.0 9.2 12.7 8.3 11.2 Max  45 66 80 66 73 136 122 127 Min 19 44 34 36 36 76 83 77   127 house stimuli, chosen for their comparable complexity to faces. Participant group performance was compared using identification contrast thresholds for each class of stimuli. We assessed face-specific perception based on the ratio of face vs. house identification thresholds. These two independent tasks are not calibrated for difficulty, and thus the specific value of the ratio is arbitrary. However, a larger face advantage is indicated by a lower threshold ratio, i.e. a lower face-specific perception score (as this is a threshold based measure, lower thresholds indicate better performance). This allowed us to determine the relative advantage of processing faces as compared to house stimuli.  4.2.1 Stimuli  Two stimulus categories were used for this experiment: faces and houses. Exemplar images were converted to greyscale and resized using Adobe Photoshop (www.adobe.com). From a 70-cm viewing distance the face- or house-width was 6.75 degrees.   4.2.2.1      Face stimuli Figure 4.1 shows five female faces with neutral expression chosen to be our face stimuli that were selected from the Karolinska Database of Emotional faces (Lundqvist & Litton, 1998). Each individual face chosen lacked distinguishing features such as moles or birthmarks, allowing us to ensure that these atypical markings were not used by participants to assist in discriminating one face from another. After faces were converted to greyscale, external features were covered using an oval aperture. Faces were aligned using the tip of the nose as the horizontal origin; the height of the pupils was used to vertically align the faces. From a viewing distance of 70 cm, the face-width was 6.75 degrees in size. This face size was chosen based on previous work showing   128 that expert face processes may be engaged for faces larger than 6 degrees in width (Yang, Shafai, & Oruc, 2014).   Figure 4.1. The face stimuli. Images were converted to greyscale and external features were covered with an oval aperture.  4.2.2.1      House stimuli The control stimuli were created by taking photographs of the “portico” of a house; a term for a covered porch leading into the front entrance of a home. These are a common architectural feature for homes in the Vancouver, British Columbia region. Porticos were chosen as control stimuli as they are complex images with features arranged in a common first-order configuration (roof above door, door above stairs, and a pillar on each side). The individual house exemplars selected have subtle variations in the shape of the features themselves and second-order configuration (e.g. spacing between features). Five porticos were selected from more than 100 exemplars as having the most similar complex elements (e.g. stairs centered below the front door, house siding, etc.). Identifying features such as addresses, plants, and decorations were removed using Adobe Photoshop and Illustrator. To ensure the portico was the focal point of the stimuli, an oval aperture was used to cover the external information. Photoshop was used to scale the photographs so all components of the portico were contained within the height and width of the oval aperture (see Figure 4.2). As with the face stimuli, house stimuli were 6.75 degrees wide when viewed at a 70-cm distance.  Author's personal copyby a sequence of a white noise mask (50 ms), a fixation cross(150 ms), a blank screen (150 ms), the test face (150 ms), a blankscreen again (150 ms), and finally the answer display, which re-mained until the subject entered their response (Fig. 1A). Withsix different adapting stimuli and five different test faces, therewere 30 possible adaptor/test pairings. Five belong to the baselinecondition (adaptor stimulus is a blank screen), five to the congru-ent condition (adaptor stimulus is the same face as the test face)and twenty to the incongruent condition (adaptor stimulus is aface that is different from the test face). Thirty randomly inter-leaved staircases measured the individual contrast thresholds foreach of these 30 pairings. The adaptor/test pairs were then classi-fied into three types according to their congruency in facial iden-tity: congruent, where the adapting and test faces were the sameidentity, incongruent, where the adapting and test face were differ-ent identities, and baseline, where the adapting stimulus was ablank.There were four conditions in the current experiment: (1) up-right-adaptor/upright-test (UU), (2) upright-adaptor/inverted-test(UI), (3) inverted-adaptor/upright-test (IU), and (4) inverted-adap-tor/inverted-test (II) (Fig. 1B). All four conditions were measuredfor two adaptation durations, 100 ms and 1600 ms. Our prior workestablished th t recognition in the congruent condition is facili-tated at brief adapting durations of 20–200 ms, but thresholds atFig. 1. Illustration of the experimental design. (A) Adapting stimuli were displayed for 100 ms (short duration) or 1600 ms (long duration) at high contrast. Blank adaptorswere used to estimate baseline thresholds. This was followed by a 50 ms noise mask, 150 ms fixation cross, and 150 ms blank. A low-contrast test face was presentedafterwards for 150 ms before subjects indicated which test face they saw, in a 5-AFC paradigm. Trials are further broken up into congruent (same adapting and test face, notshown) and incongruent (different adapting and test face, as shown) conditions. A total of five different Caucasian female face stimuli were used from the Karolinska facedatabase. (B) Four conditions result from pairing adapting and test stimuli of two possible orientations (upright or inverted). The upright-adaptor, upright-test (UU) conditionwas the design used in our previous study (Oruc & Barton, 2008, submitted for publication). In the present study, we explored the remaining three conditions: upright/inverted (UI), inverted/upright (IU), and inverted/inverted (II).2256 X.M. Guo et al. / Vision Research 49 (2009) 2254–2260  129   Figure 4.2 The house, or portico, stimuli. A “portico” is an architectural feature describing an entrance to a house with a small roof, stairs, and two pillars. These were chosen as control stimuli for faces as they are complex images that have features arranged in the same first-order configuration (roof above door, door above stairs, and a pillar on each side). As with the face stimuli, the individual house exemplars chosen have subtle variations in the shape of the features themselves and second-order configuration (e.g. spacing between features, etc.).   For each class of stimuli, the root-mean-squared (RMS) contrast was specified as the standard deviation of luminance divided by the mean luminance. Each individual image was adjusted so that the mean luminance was set to half maximum luminance and the RMS contrast was set to 1 inside the oval aperture. This allowed us to ensure standard contrast across all stimuli images before contrast was manipulated during the experiment.  4.2.2 Procedure  Identification contrast thresholds were measured in a five-alternative forced-choice (5-AFC) paradigm. Trial sequences began with a 150-ms fixation cross, a 150-ms blank, a 150-ms stimulus, another 150-ms blank, and finally a choice screen displaying all five possible exemplars (see Figure 4.3). The choice screen remained on the screen until the participant responded via key press. Responses were entered using the number pad of a computer keyboard, where each key spatially corresponded to the position of the exemplar on the choice screen. Auditory feedback for correct answers was a single click; incorrect answers were indicated via   130 two clicks. We randomly assigned participants to begin with either the face or the house stimuli blocks, followed by the alternate task.   Figure 4.3 Protocol for the 5-AFC identification task. Each trial begins with a 150-ms fixation cross, a 150-ms blank screen, the test stimulus for 150 ms, followed by another 150-ms blank screen, and finally the answer choice screen was displayed until the participant indicated a response via key press.    Each stimulus condition (faces, houses) had three phases: (1) familiarization, (2) training, and (3) experimental.   4.2.3.1      Familiarization phase  An introductory animation was shown to participants prior to beginning each stimulus condition. The animation allowed all participants to receive the same instructions while minimizing Fixation150 msHouseStimuli150 msAnswer displayuntil key pressBlankScreen150 msBlankScreen150 ms+  131 potential difficulties arising from impaired social interactions with the experimenter. The animation began with a short description of the stimuli, followed by a 10-second familiarization period for each of the five exemplars used in the experiment. The familiarization period began with a “pop” animation in which an individual face (or house) was centered on the screen and faded in from grey. Once the supra-threshold image was visible, the animation took 0.5 seconds to show the face (or house) grow from the center of the screen until it covered the entire screen while still maintaining the aspect ratio of the image. Participants continued to view the exemplar at the large size for 9.5 seconds. After the 10 seconds of familiarization, the animation would proceed to the next exemplar face (or house). The task and procedure were described with examples of the stimulus condition. It concluded with four “mock” trials in which participants were asked to verbally indicate the correct answer. The animation was created in Keynote (Apple, 2011) and converted to an AVI video.   All participants were given a 20-item qualification quiz after viewing the animation to allow for familiarization with the stimuli and to ensure comprehension of the task. The quiz procedure was identical to the experimental procedure, except stimuli were all displayed at high (supra-threshold) contrast and a longer duration to allow participants an opportunity to practice the task with the least amount of difficulty prior to experimental manipulations. A 10-trial warmup allowed participants to practice the task and ask questions before they began the 20-trial qualification quiz. Criterion performance was defined as achieving 90% accuracy in the quiz, and participants were asked to repeat the quiz until they could reach this threshold criterion. Both the introductory animation and quiz were presented on a Dell laptop (model 3750) equipped with a 17-in. antiglare LED screen. The quiz was created using SuperLab version 5.0 (www.superlab.com).    132  4.2.3.2      Practice phase Once participants completed the quiz with supra-threshold images, the next phase utilized the same images but manipulated the contrast to measure contrast thresholds. To acclimate participants to the experimental conditions, two practice blocks were completed prior to the experimental phase. Individual practice blocks used two interleaved staircases of 40 trials each, for a total of 80 trials in each block. At each trial, participants were presented with test stimuli for 1.0 seconds in the first practice block and 500 milliseconds for the second. Each staircase produced a single contrast threshold estimate, resulting in two contrast threshold estimates per block. Staircases were implemented using the QUEST procedure (Watson & Pelli, 1983) and Psychophysics Toolbox (Brainard, 1997). Contrast threshold estimates for each block were compared by the researcher, and those that were deemed to be too far apart (defined as one estimate’s value being no more than twice the value of the other threshold estimate) were repeated until the participant performed consistently.   4.2.3.3      Experimental phase The experimental phase began with a short 10-trial warmup with 150-ms stimulus presentation. Trials began with a 150-ms fixation cross, a 150-ms blank screen, the 150-ms stimulus, another 150-ms blank, and finally the choice screen with all five exemplar images was displayed until the participant selected a response via key press (see Figure 4.3). Subjects used the keys of a computer keyboard to indicate which face (or house) they saw. The keys on the number pad were used to spatially represent the relative position of the face (or house) stimuli on the choice   133 screen. Feedback was provided via auditory clicks, with a single click indicating a correct response and two clicks indicating incorrect responses.  4.2.3 Data analysis  We measured contrast threshold estimates for identification at 82% accuracy. Blocks consisted of two interleaved staircases of 40 trials that each produced a contrast threshold estimate. Estimates from each block were compared following the experiment. If one value was more than twice the other estimate, the experimental block was repeated until the participant produced reliable contrast thresholds estimates.  4.2.4 Results and discussion  Identification contrast thresholds for faces and houses are shown for the ASD (red) and control group (blue) are shown in Figure 4.4A. We submitted raw thresholds to a repeated-measures ANOVA, with stimuli class (face, house) as the within-subjects factor and group (ASD, control) as the between-subjects factor. This revealed that there was no main effect of group (F(1,52)=0.63, p=0.43) and no main effect of stimulus category (F(1,52)=1.17, p=0.28), and a highly significant interaction between group and stimulus category (F(1,52)=8.82, p<0.01). The ASD group had significantly higher thresholds in the face task (ASD M= 0.0915, SD= 0.1471) compared to the Control group (Control M= 0.0387, SD=0.0716) based on a post-hoc comparison (p<0.05, Tukey-Kramer Multiple-Comparison test). Thresholds in the house condition did not differ significantly between the two groups (ASD mean =0.045, SD=0.0538; Control mean =0.06, SD=0.0863).    134  Figure 4.4. Results from Experiment 1. A) Face vs. House identification contrast thresholds for ASD (red) and Controls (blue). Lower thresholds indicate better performance. There was a significant interaction between group and stimulus category. B) Face-specific perception scores defined as log threshold-ratios between the two conditions. Face-specific perception scores differed significantly between ASD and Control groups.   To examine this significant interaction further, we calculated “face-specific perception scores” defined as the log threshold-ratio between face and house conditions. In Figure 4.4B, we show the log threshold ratio for ASD (red) group compared to the Control group (blue). When thresholds for the face condition are higher (i.e. worse) than in the house condition, as with our ASD group, the log-ratio is positive. Alternatively, when the face thresholds are lower (i.e. better) than thresholds for the house condition, as they are for the Control group, the values are negative. The face-specific perception score for the ASD group differed significantly from that of the Control group based on a two-sample two-tailed Student’s t-test (t(52)=3.53, p<0.001).    Face HouseIdentification contrast threshold0.020.040.080.16Identity (perception)ASD ControlFace-specific perception-0.8-0.6-0.4-0.200.20.4***p<0.001ASDControlA) B)  135 4.3 Experiment 2: memory for facial identity  The Cambridge Face Memory Test (CFMT) was completed in both the upright and inverted formats by each participant (Duchaine & Nakayama, 2006). As described in Chapter 1, this measure was designed to screen for prosopagnosia, or ‘face-blindness’. The CFMT has normative data from thousands of participants, enabling us to check our participant results against the established norms. Although designed as a standardized measure of face memory, the upright CFMT scores do not solely reflect face-specific processes. Indeed, upright and inverted CFMT scores are often found to be highly correlated (Russell, Duchaine, & Nakayama, 2009; Wilmer et al., 2010), as general visual abilities contribute to the processing of both upright and inverted conditions. The upright advantage, however, results from contributions of “face-specific” processes to the upright condition. We calculate “face-specific memory” by subtracting the inverted accuracy from the upright accuracy score. These scores also allowed for additional performance comparisons between participant groups.   4.3.1 Experimental setup  The CFMT was administered on a Dell laptop (model 3750) equipped with a 17-in. antiglare LED screen. Participants were asked to be seated at a comfortable distance from the screen, corresponding to a viewing distance ranging from 50 to 70 cm.   4.3.2 Stimuli  The CFMT (Duchaine & Nakayama, 2006) consists of male faces in two viewpoints: facing forward and side-facing 1/3 profile images where the head is turned either left or right. The males are wearing hats that cover the hairline and ears so only the internal features of the face are   136 exposed. There are 6 target male faces, each presented 12 times, and a total of 46 distractors. The same faces are presented upside down in the inverted condition.   4.3.3 Procedure  The CFMT has four parts: practice, introduction, novel images, and novel images with noise (Duchaine & Nakayama, 2006). During the introduction, subjects are asked to memorize a face shown in a sequence at three different angles. They are then presented with three faces and asked to select the face they memorized amongst two target distractors by pressing the number that corresponds to the target face. This is repeated twice more, so that each orientation previously presented is tested. Next, in the novel images phase a screen displaying a frontal view of each of the six target faces is shown for 20 seconds to allow the subject to “review” the identities. The phase continues with each of the 6 target faces being presented amongst two distractors at novel orientations. This is done five times for a total of 30 forced-choice items. Finally, in the novel images with noise condition, subjects are given another 20-second review of the frontal view of the 6 target faces before being tested with novel images presented with noise. The images have Gaussian-noise added and are presented at novel orientations, with one target and two distractors for each trial. Each of the 6 target faces are presented 4 times, for a total of 24 forced-choice items. The inverted condition is the exact same procedure and target faces, just with all the faces presented in the inverted position.       137 4.3.4 Data analysis  Both the upright and inverted CFMT scores are calculated as percent correct out of the 72 total items (Duchaine & Nakayama, 2006). Face-specific memory score is calculated by subtracting the inverted face score from the upright face score.  4.3.5 Results and discussion  Figure 4.5 shows the upright CFMT scores for the ASD group (red bar) and the control group (blue bar). Individuals with ASD scored significantly lower than Controls (p<0.001).   Figure 4.5. Results from Experiment 2: CFMT results. Participants with ASD (red bar) scored significantly lower than Controls (blue bar). Higher scores indicate better face memory performance.  The face-specific memory scores are presented in Figure 4.8. This score is calculated as the difference between the upright and inverted face memory results. While individuals with   ASD  ControlCFMT score (% correct)0.40.50.60.70.80.91CFMTp < 0.001**  138 ASD do have an upright face advantage, it is significantly lower than the control group (p < 0.036).    Figure 4.6. Results from Experiment 2: Face-specific memory is based on difference between upright and inverted CFMT scores. Participants with ASD (red) were scored significantly lower compared to Controls (blue).   Taken together, the CFMT results indicate that adults with ASD scored significantly lower in face memory tasks compared to Controls. The average score for the general population is 80.4% (SD=11.0) (Duchaine & Nakayama, 2006), and our controls are close to this average with a score of 81.2% (SD=10.8). Our participant group with ASD however, scored significantly lower with an average score of 66.7% (SD=17.4). It is worth noting that while this is significantly lower performance compared to the typical population and our age- and IQ-matched Controls, our individuals with ASD still outperform individuals with prosopagnosia with an   ASD  ControlCFMT difference score (% correct)00.050.10.150.20.25Face-specific memory*p < 0.036  139 average score of 50.7% (SD=13.4) (Duchaine & Nakayama, 2006). Prosopagnosia cut-off was defined to be scores more than two standard deviations below the control mean, meaning scores below 58.4% would be considered prosopagnosia. Using this criterion, we find that 12 of our 27 participants would qualify as being prosopagnosic. Conversely, none of our Control participants scored within the prosopagnosic range established in Duchaine and Nakayama (2006). Next, in Figure 4.7 we plot individual scores for all ASD (left panel) and Control (right panel) participants for upright and inverted CFMT. General visual abilities contribute to both the upright and inverted conditions, but expert face processing contributes predominantly to performance in the upright condition. Contributions of expert face processes exclusively in the upright condition have the effect of attenuating the strength of the correlation between upright and inverted scores. Thus, lower correlations suggest greater involvement of face-specific processes. The magnitude of the correlation in the ASD group (r=0.75, p << 0.001) is greater compared to Controls (r=0.39, p = 0.05) suggesting that individuals with ASD may not utilize expert face processes to the same extent as Control participants. Therefore, altogether, these results suggest that face-specific memory may be reduced in the ASD population.   140  Figure 4.7 Correlation between performance on the upright and inverted versions of the CFMT. The correlation for ASD (red) is higher than the correlation for Control (blue) participant groups. The significantly higher correlation for the ASD group indicates that individuals with ASD may not employ expert face processes to the same extent as Controls.   4.4  Experiment 3: face expression discrimination  Discrimination performance for facial expressions was assessed for ASD and Control participants using stimuli with graded expression strengths. We wanted to assess the observers’ ability to pick up on subtle expressions—for this reason we created stimuli with graded expression strength using a face morphing technique. Discrimination thresholds were measured between pairs of three face expressions (two out of angry, happy, and sad). By design, task difficulty was adaptively adjusted based on each individual participant’s level of performance for each task.   Correlation between upright and inverted CFMT scoresASD ControlCFMT upright (proportion correct)CFMT upright (proportion correct)CFMT inverted (proportion correct) CFMT inverted (proportion correct)0.4 0.6 0.8 10.30.40.50.60.70.80.9r=0.39, p=0.050.4 0.6 0.8 10.30.40.50.60.70.80.9r=0.75, p<<0.001  141 4.4.1 Experimental setup  As with Experiment 1, a computer equipped with a Cambridge Research Systems VSG 2/3 graphics card and SONY Trinitron 17-in. monitor (model Multiscan17seII) was utilized for this experiment. A CRS OptiCAL photometer (model OP200-E) and software provided by CRS allowed for gamma correction. The display had a mean luminance of 17.4 cd/m2. We programmed the experiment in Matlab (www.mathworks.com) using tools from the CRS VSG Toolbox for Matlab and Psychophysics Toolbox (Brainard, 1997; Pelli, 1997). Participants sat 70 cm away from the computer screen.  4.4.2 Stimuli  Three male faces were selected from the Karolinska Database of Emotional Faces (Lundqvist & Litton, 1998) as base identities for this experiment (see Figure 4.8A). For each base identity, we selected one neutral and two endpoint expression images from the database (two out of angry, happy, and sad). Images were converted to greyscale and external features were covered with an oval aperture. We aligned the faces horizontally by centering the tip of the nose, and vertically by aligning the height of the pupils through all the faces. From the experimental viewing distance of 70 cm, the faces were presented at a face-width of 6.75 degrees. Using Fantamorph software (Abrosoft, 2005), we made gradual, step-wise morphs between neutral and each of the two endpoint expressions (see Figure 4.8B and 4.8D).    142  Figure 4.8. Illustration of the stimuli and procedure for Experiment 3. (A) Faces of three males were selected as the base identities (A, B, and C). These were used as the neutral exemplar faces to generate three different sets of test stimuli. (B) Examples of the A-sad expression morphs displaying subtle changes in expression (D)(A)(C)base identityA CBbase identity- Aneutralsad angryend-point expressionend-pointexpressionmorph seriesA-sad40% sadmorph seriesA-angry40% angry.. .. ..........fixation150 msfixation150 msblankuntil key presstest screen 11 secondtest screen 21 second(B) morph series A-sad100% sad 80% sad 60% sad 40% sad 20% sad neutral  143 strength from 100% sad to neutral. (C) Illustration of stimulus generation. Generation of a test set began with a base identity with neutral expression (A in this example) being morphed towards two different endpoint expressions. The endpoint faces are the same individual (A) displaying an expression (40% sad and 40% angry in this example). Test stimuli consisted of the same base identity displaying subtle expressions to either endpoint expression. The expression strength was determined by the % morph-distance. (D) Experimental trial protocol. Discrimination thresholds were measured using a two-interval forced-choice paradigm. A 150-ms fixation cross, followed by test screen 1 for 1 second, another 150-ms fixation cross, test screen 2 for 1 second, and finally a blank screen that was displayed until a key was pressed. The test screens each contained one of the two-morph test stimuli (e.g. A-sad, A-angry) in a random sequential order. Participants were asked to indicate whether the first or second test screen contained the stimulus that was e.g., “angrier” (or “sadder”), in this example. The two test stimuli were constrained to be equidistant in % morph-distance from either side of neutral. The % morph-distance between neutral-angry and neutral-sad for each trial was determined by the Quest procedure.    Each ‘direction’ in the morph series consisted of morphing a neutral face towards the expression endpoint (e.g. sad), thereby producing neutral-sad morphs in which a face is gradually modified to display more sadness as it moves towards the extreme expression. Similarly, the same neutral face is also morphed towards the other expression endpoint (e.g. angry) to produce neutral-angry morphs. A series contained 41 images that were each a step along the morph series (i.e. 2.5% increments).   We created three sets of stimuli (A, B, C) that featured morph series between the following pairs of expressions: angrier/sadder, happier/sadder, angrier/happier. By utilizing the same physical stimuli (e.g. A series) to test the subject’s ability to discriminate between either “sadder” or “angrier”, we were able to assess performance across different expressions while maintaining the inherent difficulty of the task. We assessed discriminative abilities for three expressions tested for the ASD and Control groups.  4.4.3 Procedure  Each participant’s expression discrimination threshold for a target expression was measured in a two-interval forced-choice paradigm (2-IFC). As illustrated in Figure 4.8C, trials began with a 150-ms fixation cross, a 1 second test screen containing one of the two expression morphs (e.g.   144 A-sad morph), a second fixation cross for 150-ms, another 1 second test screen containing the other expression (e.g. A-angry morph), and a blank screen that remained until the participant indicated whether the first (1) or second (2) test screen contained the target expression. Having equidistant steps along the morph series between neutral and the endpoint expressions allowed us to determine how accurately a participant can discriminate one morph (e.g. sad-neutral) from another morph (e.g. angry-neutral) with the same expression strength (e.g. both 15% away from neutral). Auditory feedback was given as a single click for correct answers, two clicks for incorrect answers.   For each of the three stimulus conditions (base identities A, B, and C), subjects completed two phases: (1) familiarization and (2) experimental. Each base-identity and target expression was tested in separate blocks. The sequence of base identity presentations was randomized for each participant. Additionally, each participant was randomly assigned one of the two target expressions at each block.  4.4.3.1      Familiarization phase  An animation introducing the task was shown to participants before beginning each experiment. It was created to ensure that the same instructions were given to all participants while minimizing the impact of possible social impairment when interacting with the experimenter. The animation began with a short description of the stimuli, then a description of the task protocol with examples. Finally, four mock trials concluded the animation. It was created in Keynote (Apple, 2011) and converted to an AVI video.   Immediately following the animation, participants were asked to take a quiz. Participants were randomly assigned an order for the three base identities (i.e. A, B, C) and a target   145 expression (e.g. angrier). The short, 10-item quiz consisted of the same task and procedure as in the experiment, only with stimuli that were the maximum expression strength of a particular expression. The quiz had 10 trials and required at least 90% accuracy before participants could begin the experiment. We used Superlab version 5.0 (www.superlab.com) to create the quiz and presented it on a Dell laptop computer (model 3750) with a 17-in. antiglare LED screen.  4.4.3.2      Experimental phase  Following the familiarization phase, subjects were asked to complete 10 trials as a warmup before the experimental block. The experiment block was a single staircase of 40 trials to estimate the percent morph distance (as a measure of expression strength) discrimination thresholds. Participants were randomly assigned to either of the two possible target expressions in each series, and the target remained constant throughout the block. Participants were also randomly assigned a presentation order for the base identities (e.g. B, C, A).   4.4.4 Data analysis  Discrimination threshold estimates were based on 82% accuracy over a total of 40 trials. All subjects reached threshold for each block, many in their first try. For the small number of participants who did not reach threshold initially, the participant was retrained on the base -identity and target expression before attempting the experimental block again. Retraining consisted of repeating the animation and quiz until 90% accuracy was achieved. If the participant was unable to reach threshold after three repeat attempts, a longer break would be given before reattempting the animation and quiz.    146 4.4.5 Results and discussion  Figure 4.9 shows the expression discrimination thresholds (proportion morph distance) of each participant group in each of the three face series. Base identity A contained angry and sad expression morphs. Base identity B had participants discriminate between happy and sad images. Base identity C contained angry and happy images. Higher thresholds indicate poorer performance for these measures. Participants with ASD scored significantly higher in all three face series compared to neurotypical controls, indicating that more expression strength is necessary for an individual with ASD compared to a Control to do the same task. For base identity A, the ASD group averaged significantly poorer thresholds compared to the control group (ASD M=0.3409, Control M=0.2692, p<0.05). For base identity B, individuals with ASD scored significantly higher compared to the threshold values of the control group (ASD M=0.2645, Control M=0.1637, p<0.01). For base identity C, the ASD group scored significantly higher compared to the control group (ASD M=0.2292, Control M=0.1408, p<0.01).     147    Figure 4.9 Expression discrimination results. Participants with and without ASD completed three expression discrimination sets, resulting in expression discrimination thresholds of proportion morph distance. In block A, participants were asked to discriminate between angry and sad expressions. In block B, participants were asked to discriminate between happy and sad expressions. In block C, participants were asked to discriminate between angry and happy expressions. Lower thresholds indicate superior performance on the task.  In order to compare performance across different expressions, we normalized the expression discrimination thresholds by calculating z-scores by scaling the thresholds to be zero mean unit variance. For each base identity, we pooled the normalized thresholds based on the type of expression: angry, happy, or sad. It was necessary to normalize the results because raw thresholds across the series are not comparable, as they are partially based on the specific stimulus-series (e.g. Face identity A may be a better actor than Face identity B or C). Normalized discrimination scores pooled for each expression separately are shown in Figure 4.10 for the ASD and Control groups. These results show significant differences between the two groups for angry (ASD M=0.49, Control M=-0.20, p<0.05), sad (ASD M=0.33, Control M=-0.32, p<0.01), and happy (ASD M=0.34, Control M=-0.32, p<0.01) expressions. Based on these results, across the three expressions tested, individuals with ASD show significantly poorer performance.  A B CExpression discrimination threshold (proportion morph distance)00.10.20.30.4ASDControl****** p<0.05 ** p<0.01 A = Angry/SadB = Happy/SadC = Angry/Happy  148    Figure 4.10 Normalized expression discrimination thresholds from Experiment 3. The three stimulus series’ thresholds were normalized before being collapsed by expression type into three categories: angry, sad, and happy. Adults with ASD (red bars) scores were significantly different, indicating poorer performance, compared to controls (blue bars) on all three expressions tested.  Taken together, these results indicate that adults with ASD had significantly poorer performance in expression discrimination for all three tested expressions.  4.4.6      AQ and MSCS results Figure 4.11 shows mean MSCS scores for the group with ASD (red bar) and controls (blue bar). The ASD group scored significantly lower across all seven domains of the MSCS: including social motivation (ASD M=31.6, Controls M=40.7, p<0.01); social inferencing (ASD M=33.3, Controls M=43.5, p<0.01); empathic concern (ASD M=38.6, Controls M=45.3, p<0.01); social Angry  Sad HappyNormalized expression discrimination threshold-0.8-0.6-0.4-0.200.20.40.60.8ASDControl*p<0.05, **p<0.001* ** **  149 knowledge (ASD M=40.6, Controls M=46.0, p<0.01); verbal conversation skills (ASD M=33.7, Controls M=40.4, p<0.01); nonverbal sending skills (ASD M=35.7, Controls M=45.0, p<0.01); and emotion regulation (ASD M=34.4, Controls M=41.2, p<0.01). Consequently, there was a significant difference overall with a total MSCS scores for the ASD group and Controls (ASD M=247.9, Controls M=302.2, p<0.01).   Figure 4.11 Mean MSCS scores for ASD (red) and controls (blue) in each of seven distinct domains of social competence: social motivation (SM), social inferencing (SI), empathic concern (EC), social knowledge (SK), verbal conversation skills (VS), nonverbal sending skills (NS), and emotion regulation (ER). Lower scores indicate poorer social competency.   SM SI EC SK VS NS ER MSCS score05101520253035404550All p's<0.008ASDControl**** **********MSCS  150 Figure 4.12 shows the mean AQ scores for the group with ASD (red bar) and Controls (blue bar). The mean AQ scores for the ASD group (M= 28.19, SD=11.17) were significantly higher than that of the Control group (M=13.33, SD=4.17) (p<0.001).   Figure 4.12 Mean AQ scores for ASD (red) and controls (blue). The error bars represent +/- 1 standard error of the mean. Higher scores indicate more traits associated with ASD. Individuals with ASD had significantly higher scores than controls, indicating greater severity of ASD symptoms.  4.5 Correlation analysis  Next, we assess whether a relationship exists between social competence and face identity and expression performance for our participants with and without ASD. Throughout these correlation analyses, we used a hierarchical hypothesis testing framework to assess a small number of primary hypotheses. If these initial tests yielded significant results, more detailed analyses were conducted to determine the primary contributing factors to the association. Additionally, we   ASD  ControlAQ05101520253035AQp <0.001ASDControl***  151 computed all correlations separately for our ASD and Control group, allowing us to assess the presence of associations unique to one group or the other. Finally, all p values were False Discovery Rate (FDR) adjusted to allow testing of multiple hypotheses.  4.5.1 Results and discussion  We first assessed the relationship between face-specific perception and social competence. Face-specific perception was calculated based on log threshold-ratios between the face and house conditions, while social competence was based on the total MSCS score. In Figure 4.13, we show the scatter plot for the ASD group in red, and separately in blue for the Control group. No association was found between social competence and face-specific perception for the ASD (r=0.17, p=0.69) or Control (r=0.02, p=0.91) groups.  Figure 4.13 Correlation between social competence and face-specific perception. The total MSCS scores for ASD (red) and controls (blue) are plotted against face-specific perception scores. Please note that the y-axes are not in the same range due to group differences in the MSCS measures. Pearson correlation coefficient (r) values were low and non-significant between social competence and face-specific perception for both participant groups.   Face-specific perception and social competenceFace-specific perception-2 -1 0 1 2MSCS score100150200250300350r= 0.17 p=0.69Face-specific perception-2 -1 0 1 2MSCS score220240260280300320340360r= 0.02 p=0.91ASD Control  152 Next, we examined the relationship between face-specific perception and autism symptom severity. Figure 4.14 shows the correlation between total AQ scores and face-specific perception score separately for the ASD (red) group and Control group (blue). No association was found between ASD symptom severity and face-specific perception for the ASD group (r=-0.22, p = 0.53) or Control group (r=-0.18, p = 0.56).   Figure 4.14 Correlation between ASD symptom severity and face-specific perception. The total AQ scores for ASD (red) and controls (blue) are plotted against with face-specific perception scores. R-squared values were low and non-significant between ASD symptom severity and face-specific perception for both participant groups.          We next assessed our other measure of face identity processing: face-specific memory. We calculated face-specific memory based on the residuals between upright and inverted CFMT scores. In Figure 4.15, we show the association between face-specific memory and social competence as measured by the total MSCS score for the ASD group (red) and Control group Face-specific perception and ASD symptom severityFace-specific perception-2 -1 0 1 2AQ score01020304050r=-0.22 p=0.53Face-specific perception-2 -1 0 1 2AQ score01020304050r= -0.18 p=0.56ASD Control  153 (blue). There was no relationship between social competence and face-specific memory for the ASD group (r=-0.08, p = 0.69) or Control group (r=0.07, p = 0.91).    Figure 4.15 Correlation between social competence and face-specific memory. The total MSCS scores for ASD (red) and controls (blue) are plotted against face-specific memory scores. Please note that the range for the y-axes are different for each group as there is a wider range of scores for the ASD group compared to controls. There was no correlation between social competence and face-specific memory for either participant group.        Figure 4.16 shows the relationship between face-specific memory and ASD symptom severity as measured by the AQ for the ASD (red) group and Control group (blue). There were no significant associations between the two measures for the ASD group (r=0.10, p = 0.63) or Control group (r=0.12, p = 0.56).  Face-specific memory and social competenceASD ControlFace-specific memory-0.4 -0.2 0 0.2 0.4MSCS score100150200250300350r=-0.08 p=0.69Face-specific memory-0.3 -0.2 -0.1 0 0.1 0.2MSCS score220240260280300320340360r= 0.07 p=0.91  154  Figure 4.16 Correlation between ASD symptom severity and face-specific memory. The total AQ scores for ASD (red) and controls (blue) are plotted against face-specific memory scores. There was no relationship between social competence and face-specific memory for either participant group.        Taken together, the results indicate that face identification skills, including both perception and memory, are not correlated with social competence or ASD symptom severity. As there were no significant correlations, we do not proceed with additional analyses of face identification and instead turn our attention to face expression processing.  In Figure 4.17, we plot grand average of expression thresholds (across the three stimulus series) against total MSCS scores. Individuals with ASD (red) showed no significant correlation between face expression discrimination abilities in general and total MSCS score (r=-0.006, p=0.78). Control participants, however, did show an association between face expression discrimination abilities and total MSCS score (r=-0.56, p=0.007). As MSCS scores increase, expression discrimination thresholds decreased, indicating that greater social competence is related to improved expression discrimination abilities for Controls.  Face-specific memory and ASD symptom severityASD ControlFace-specific memory-0.4 -0.2 0 0.2 0.4AQ score01020304050r=0.10 p=0.63Face-specific memory-0.3 -0.2 -0.1 0 0.1 0.2AQ score01020304050r= 0.12 p=0.56  155  Figure 4.17 Correlation between social competence and expression discrimination. The log mean of the expression discrimination threshold across the three series tested is plotted against total MSCS scores for the ASD group (red) and Controls (blue). Note that y-axis limits differed slightly between the ASD and Control groups as the ASD group had wider variability in total MSCS scores. When all the expression discrimination thresholds tested are combined, there is not a significant association for the ASD group, but there is for the Control group.    We next examined the relationship between processing of individual expressions and social competence. In Figure 4.18, we plot z-scores from the ASD group for each expression tested (angry, happy, and sad) against total MSCS scores. The results indicate that there is no association with total MSCS and discrimination thresholds for angry (r=-0.13, p=0.82) or happy (r=0.04, p=0.83) expressions, but there is a significant interaction with social competence and discrimination thresholds for sad expressions (r=-0.46, p = 0.04) for individuals with ASD. Face expression discrimination and social competenceASD Control-2.5 -2 -1.5 -1 -0.5 0MSCS score150200250300350log mean expression discrimination thresholdr=-0.06 p=0.78-2.5 -2 -1.5 -1 -0.5 0MSCS score200250300350log mean expression discrimination thresholdr=-0.56 p=0.007  156   Figure 4.18 Correlation between social competence and expression discrimination thresholds for angry, happy, and sad expressions for the ASD participant group. The z-scores of the expression discrimination thresholds are plotted against total MSCS scores. The only significant association is between sad expression discrimination thresholds and overall social competence for ASD participants.    In Figure 4.19, we show the same analysis for our Control participant group, comparing total MSCS scores and normalized expression discrimination thresholds for each of the three expressions tested (angry, happy, and sad). There was no significant association between total MSCS scores and discrimination thresholds for happy expressions (r=-0.26, p=0.21) for the Sad z-scores-2 -1 0 1 2 3MSCS score100150200250300350Angry z-scores-2 0 2 4MSCS score150200250300350angry happysadHappy z-scores-2 -1 0 1 2 3MSCS score100150200250300350r=0.04 p=0.83r=-0.13 p=0.82r=-0.46 p=0.04ASD expression discrimination and social competence  157 Control group, but there were significant interactions between total MSCS scores and normalized discrimination thresholds for angry (r=-0.65, p=0.001) and sad (r=-0.41, p=0.04) expressions.    Figure 4.19 Correlation between social competence and expression discrimination thresholds for angry, happy, and sad expressions for the Control participant group. The z-scores of the expression discrimination thresholds are plotted against total MSCS scores. Significant correlations were found between angry and sad expression discrimination abilities and overall social competence.    Happy z-scores-1.5 -1 -0.5 0 0.5 1MSCS score260280300320340360Sad z-scores-3 -2 -1 0 1 2MSCS score220240260280300320340360Angry z-scores-2 -1 0 1 2MSCS score220240260280300320340360Control expression discrimination and social competenceangry happysadr=-0.41 p=0.04r=-0.26 p=0.21r=-0.65 p=0.001  158 Following significant correlations for the ASD group in discrimination of sad expressions and overall social competence, we next assessed the individual subdomains of the MSCS to see which aspects of social competence were driving this association. In Figure 4.20, we show the relationship between sad expression z-scores and each subdomain of the MSCS for our participants with ASD. Of the seven subdomains, only empathic concern had a significant correlation with the ability to discriminate sad expressions (p=0.01).    159  Figure 4.20 Correlation between social competence subdomains and expression discrimination thresholds for sad expression for the ASD participant group. The z-scores of the expression discrimination threshold are plotted against each of the seven MSCS subdomain scores. For our ASD group, a significant correlation was found between sad expression discrimination abilities and the empathic concern subdomain of the MSCS.    -5 0 5050Social motivation-5 0 5050Social inferencing-5 0 5050Empathic concern-5 0 5050Social knowledge-5 0 5050Verbal conversation-5 0 5050Nonverbal sending-5 0 5050Emotion regulationSad z-scoresMSCS subscorer=-0.29 p=0.16 r=-0.20 p=0.31r=-0.59 p=0.01 r=-0.41 p=0.10r=-0.37 p=0.12 r=-0.35 p=0.12r=-0.29 p=0.16All p values are FDR adjustedASD association between sad expression discrimination and MSCS subdomains  160 Next, we assess the correlation between sad expression processing and the individual subdomains of the MSCS for our Control group. In Figure 4.21, we show the relationship between sad z-scores and the seven subdomains of the MSCS for Control participants. We find significant correlations between sad z-scores and the social motivation (p=0.01) and empathic concern (p=0.01) domains of the MSCS.   161  Figure 4.21 Correlation between social competence subdomains and expression discrimination thresholds for sad expressions for the Control participant group. The z-scores of the expression discrimination thresholds are plotted against each of the seven MSCS subdomain scores. A significant correlation was found between sad expression discrimination abilities and the social motivation and empathic concern subdomains of the MSCS.    -5 0 5050Social inferencing-5 0 5050-5 0 5050Social knowledge-5 0 5050Verbal conversation-5 0 5050Nonverbal sending-5 0 5050Emotion regulationSad z-scoresMSCS subscorer=-0.39 p=0.08r=-0.57 p=0.01 r=-0.20 p=0.43r=-0.17 p=0.43 r=-0.17 p=0.43r=0.12 p=0.54All p values are FDR adjusted-5 0 5050Social motivationEmpathic concernr=-0.54 p=0.01Control group association between sad expression discrimination and MSCS subdomains  162 Finally, we look at the association between angry expression discrimination abilities and the subdomains of social competence for the Control group. In Figure 4.22, we show the relationship between each subdomain and performance with angry expressions. There was a significant correlation with the social inferencing (p<0.001), empathic concern (p<0.001), social knowledge (p<0.001), and nonverbal sending skills (p=0.002) subdomains of the MSCS.   163   Figure 4.22 Correlation between social competence subdomains and expression discrimination threshold for angry expressions for the Control participant group. The z-scores of the expression discrimination thresholds are plotted against each of the seven MSCS subdomain scores. A significant correlation was found between angry expression discrimination abilities and social inferencing, empathic concern, social knowledge, and nonverbal sending skills subdomains of the MSCS.    -2 0 2050Social motivation-2 0 2050Verbal conversation-2 0 2050Nonverbal sending-2 0 2050Emotion regulationAngry z-scoresMSCS subscorer=-0.28 p=0.17r=-0.60 p=0.002r=-0.40 p=0.06-2 0 2050Social knowledger=-0.64 p<0.001-2 0 2050Empathic concernr=-0.65 p<0.001-2 0 2050Social inferencingr=-0.69 p<0.001r=-0.38 p<0.06Control group association between angry expression discrimination and MSCS subdomains  164  Next, we assessed the relationship between expression discrimination thresholds and ASD symptom severity. In Figure 4.23 we plot the log of the mean scores across all three base identities for each participant against total AQ scores. There was no significant correlation between face expression discrimination abilities in general and overall ASD symptom severity for either the ASD group (r=-0.12, p=0.55) or Control group (r=-0.28, p=0.16).    Figure 4.23 Correlation between ASD symptom severity and expression discrimination threshold. The log mean of the expression discrimination thresholds across the three faces tested is plotted against total AQ scores for the ASD group (red) and Controls (blue). When all the expression discrimination thresholds tested are combined, there is not a significant association for the ASD group or Control group.    It is possible that significant associations with one expression are being masked by non-significant results between ASD symptom severity and the other expressions. In the next step of our analysis, we separate the three expressions tested (angry, happy, and sad) to check for associations between expression discrimination abilities in a particular expression and autism log mean expression discrimination thresholdlog mean expression discrimination thresholdFace expression discrimination and ASD symptom severityASD Control-2.5 -2 -1.5 -1 -0.5 0AQ score01020304050r=-0.28 p=0.16-2.5 -2 -1.5 -1 -0.5 0AQ score01020304050r=-0.12 p=0.55  165 symptom severity as measured by the total AQ score. In Figure 4.24, the normalized discrimination thresholds for each expression are plotted against total AQ scores for the ASD group. There were no significant interactions with ASD symptom severity for angry (p=0.13), happy (p=0.72), or sad (p=0.72) expressions.    Figure 4.24 Correlation between ASD symptom severity and expression discrimination thresholds for angry, happy, and sad expressions for our ASD participant group. Normalized discrimination thresholds for each one of the three expressions are plotted separately against total AQ scores. Across all three expressions, there are no significant associations with ASD symptom severity.    Angry z-scores-2 0 2 4AQ score01020304050ASD expression discrimination abilityand ASD symptom severityangry happysadr=0.41 p=0.13Sad z-scores-2 -1 0 1 2 3AQ score1020304050r=0.08 p=0.72Happy z-scores-2 -1 0 1 2 3AQ score1020304050r=0.07 p=0.72  166 Next, we performed the same analysis for our Control group. The normalized discrimination thresholds for each expression are plotted against ASD symptom severity as measured by the total AQ score in Figure 4.25. As with our ASD participants, no significant interactions were found for angry (p=0.37), happy (p=0.50), or sad (p=0.35) expressions.   Figure 4.25 Correlation between ASD symptom severity and expression discrimination thresholds for angry, happy, and sad expressions for the Control participant group. The z-scores of the expression discrimination threshold are plotted against total AQ scores. Across all three expressions, there is no significant correlation with ASD symptom severity.    Happy z-scores-1.5 -1 -0.5 0 0.5 1AQ score4681012141618Sad z-scores-3 -2 -1 0 1 2AQ score8101214161820Angry z-scores-2 -1 0 1 2AQ score01020304050Control expression discrimination ability and ASD symptom severityangry happysadr=0.24 p=0.37r=0.30 p=0.35r=0.14 p=0.50  167 Taken together, these correlation analyses indicate that while individuals with ASD score lower in measures of face identity perception and memory, these lower scores are not associated with social competence or ASD symptom severity. Impairments in face expression processing, however, did show an association with social competence for ASD and control participants alike. This association between expression discrimination ability and social competence was strongest between the empathic concern sub scale and the ability to discriminate sad expressions for the ASD group. Social competence was associated with expression discrimination abilities for sad and angry expressions in the Control group. There was no significant correlation between expression discrimination abilities and ASD symptom severity for either group.  4.6 General discussion and conclusions  As described in Chapter 1, the literature is divided as to whether individuals with ASD have an impairment in identification of faces, discrimination of facial expressions, eye gaze, or some combination of all three (Bailey et al., 2005; Barton et al., 2007; Churches et al., 2012; Falck-Ytter, 2008; Pallett et al., 2013; Simmons et al., 2009; Tanaka & Sung, 2016; S. Wallace et al., 2008b; Wolf et al., 2008). Given the high social relevance of the human face, it remains an active avenue of inquiry for researchers seeking to characterize the sources of the social impairments associated with individuals with ASD and to determine if there is a relationship to face processing deficits in this population. It has been suggested that these social impairments may be a direct consequence of difficulties in identifying faces or face expressions (Baron-Cohen et al., 2000; Dawson, Webb, & McPartland, 2005; Schultz, 2005). We sought to contribute to this discussion by assessing face memory, perception, and expression discrimination in the same cohort of individuals with ASD. We obtained detailed assessments for each participant, allowing   168 us to examine the association between social competence and face processing abilities. We tested two main hypotheses: (a) face identification and (b) face expression discrimination impairments are associated with social competence issues and symptom severity in adults with ASD. Face identification was assessed using two experimental paradigms: a perception task comparing thresholds for identifying face and house stimuli, and face memory using the upright and inverted CFMT.  The ASD group scored significantly lower in the face condition compared to Controls, but they were not worse in the house condition. On the contrary, the ASD group showed a tendency to perform better with houses than the Control group. There was no main effect for group or stimulus category. The ASD group did not show an overall visual impairment as they performed just as well as the Controls overall. The difference emerged from the relative performance between the face vs. house conditions across the two groups. Additionally, based on the standard deviation between groups for face-specific perception, ASD participants demonstrated greater variability in face identification contrast thresholds, which may reflect the heterogeneity of face processing abilities in this population (Barton, Cherkasova, Hefter, et al., 2004). These results are similar to other studies comparing face and object processing in ASD which suggest that individuals with ASD are not utilizing specialized expert face processing strategies to the same degree as neurotypical individuals (Arkush et al., 2013; Humphreys et al., 2008; Pallett et al., 2013).   We also assessed face-specific memory as a type of face identification skill in individuals with ASD. Using the CFMT, we found that individuals with ASD performed worse in the upright condition compared to Controls. Additionally, adults with ASD showed less of a face inversion effect compared to controls, indicating that there is less of an upright face advantage- a   169 hallmark of configural face processing. The CFMT has well-established norms for the general population (Duchaine & Nakayama, 2006), making it an ideal task to compare performance between individuals with ASD and the typical population. Our results agree with previous work that showed that face memory is variable, but moderately impaired for this group compared to the general population (Hedley, Brewer, & Young, 2011; O’Hearn et al., 2010). Despite this lower performance on the CFMT compared to the general population, all but 12 of our adults with ASD outperformed individuals with prosopagnosia (Duchaine & Nakayama, 2006). It is possible that the mixed results for face memory in ASD that are reported in the literature are due to their impairments not being as severe as in prosopagnosia, and so these subtle differences may not be readily observable between subjects with ASD in different studies.  We found a higher correlation between upright and inverted face memory tasks for our ASD group, indicating that expert processes utilized for face recognition are not being employed to the same degree in individuals with ASD. This suggests that not only is this population impaired with face memory compared to their neurotypical peers, but this effect is face-specific and does not generalize to other types of visual processing. Our results are supported by studies of ERP latencies to face stimuli which indicate an inability to engage specialized face processing mechanisms in ASD (Bailey et al., 2005; Dawson, Webb, & McPartland, 2005). Conversely, our finding of quantitative and qualitative differences in face identity processing is at odds with a review by Weigelt et al. (2012), which reported that quantitative differences were driving the impaired face perception typically found in individuals with ASD. However, the more recent review by Tang et al. (2015) argues that both qualitative and quantitative differences are found for face identification in ASD. Our results support this later review and suggest that face-specific memory is impaired both quantitatively and qualitatively for individuals with ASD.   170  In our face expression discrimination experiment, individuals with ASD were significantly impaired with all three expressions tested compared to the Control group. These findings are in agreement with several studies (Clark, Winkielman, & McIntosh, 2008; Eack et al., 2015; Griffiths et al., 2017; Sato et al., 2017), but are at odds with other studies of expression processing in ASD, which have found processing of happy expressions were not impaired (C. Ashwin et al., 2006; Crawford et al., 2015; Farran et al., 2011). As negative and positive expression processing were both impaired in our ASD group, our findings are also at odds with the predictions of amygdala dysfunction theories of ASD (Adolphs et al., 2001; C. Ashwin et al., 2006; Schultz, 2005). On the other hand, some studies suggest that the amygdala is not exclusively involved in negative expression processing (Wang et al., 2017; M. A. Williams, Morris, McGlone, Abbott, & Mattingley, 2004). Our results are consistent with the amygdala playing a role in processing of both positive and negative expressions alike, and consistent with the hypothesis that amygdala dysfunction is altering expression processing in ASD.  To test our first hypothesis that face identification impairments are associated with social competence, we examined potential correlations between performance in both face identity tasks with social competence and ASD symptom severity measures. In both the face-specific perception and face-specific memory tasks, we found no evidence of any association between face identification abilities and social competence. Additionally, neither measure was associated with ASD symptom severity. Our results indicate that individuals with ASD are impaired in both face-specific perception and face-specific memory measures of identity, however these difficulties are not correlated with social competence or ASD symptom severity. These results are not in agreement with an earlier study associating face processing abilities with ADOS symptom severity, although this may be a reflection of the fact that the ADOS and AQ are   171 measuring different aspects of ASD symptom severity (Hadjikhani, Joseph, Snyder, & Tager-Flusberg, 2007; Yager & Iarocci, 2013). Taken together, these results suggest that while there are face recognition impairments in ASD, they are not directly causing or being caused by social competence issues for individuals with ASD. These particular results are in agreement with Behrmann, Thomas, et al. (2006), which proposed that face issues exist independently of social issues.   In testing our second hypothesis, we find that face expression discrimination abilities are moderately associated with social competence for sad expressions in adults with ASD. This effect was driven by the empathic concern subdomain of the MSCS. For Control participants, both sad and angry expressions were correlated with social competence. In this group, the sad expression association was due to a combination of the social motivation and empathic concern subdomains. The significant association of social competence with angry expressions for Controls was driven by four subdomains: social inferencing, empathic concern, social knowledge, and nonverbal sending skills. There were no significant correlations between expression discrimination ability and social competence for happy expressions in either group. These results are consistent with the those of Poljac et al. (2013), which found impaired face expression recognition was associated with autism trait severity for negative expressions, but not happy. Additionally, our results indicate that discrimination abilities for sad expressions are associated with empathic concern in both groups, suggesting this subdomain can predict face expression discrimination abilities for sad expressions in the ASD and general population alike. Our findings are supported by G. L. Wallace et al. (2011), who reported diminished sensitivity to sad face expressions were associated with social responsiveness and adaptive functioning for   172 adolescents with ASD. Together, these findings suggest that the ability to discriminate sad expressions is influencing and/or being influenced by social abilities in this population.  Results from the current study have direct implications for face training interventions. Programs like Let’s Face It! (Wolf et al., 2008), FaceSay (Hopkins et al., 2011), and other face identification training paradigms (e.g. Faja, Aylward, Bernier, & Dawson, 2007) provide an opportunity for individuals with ASD to develop face recognition skills and have reported successful improvements for participants with ASD. However, our results suggest that face recognition impairments are not correlated with social competence issues. While social competence is correlated with face expression processing, correlation does not imply causation. It is possible that training on such programs will not lead to improvements in ASD symptoms. Should a causal relationship exist, face impairments impact developmental processes and may prevent the maturation of face processing skills. Thus, these face training programs may not improve face processing skills or social competence. If there is a causal relationship and it is found that face training programs do improve social competence (which at this point has not been demonstrated), then our results indicate that these programs should be adjusted to focus primarily on expression recognition training, rather than face identification. Other training programs already include face expression skill development (Didehbani, Allen, Kandalaft, Krawczyk, & Chapman, 2016; Golan et al., 2010), but should be adjusted to prioritize developing skills related to processing of sad expressions as these are associated with social competence. There are disparate findings regarding aspects of face processing for identity and expressions in ASD (for reviews see Harms et al., 2010; Simmons et al., 2009; Tang et al., 2015; Weigelt et al., 2012). Using carefully designed experiments combined with detailed assessment   173 of intelligence, social competence, and autism severity provide an opportunity to examine some of these discrepancies. Our study contributes to this discussion by systematically assessing identity and expression abilities in adults with ASD and carefully matched Controls, allowing for thorough analysis of any associations between social competence and face processing abilities. Future research should expand to test for other expressions and look across the lifespan to assess developmental trajectory of face processing abilities for individuals with ASD.   174 Chapter 5: Conclusion  The visual cortex processes images of varying complexity to allow for rapid and accurate perception of the surrounding environment. Behavioural studies of visual processing have suggested that individuals with ASD may process visual stimuli differently from the general population (Kéïta et al., 2014; Manning et al., 2015; O'Neill & Jones, 1997; O'Riordan, 2004; Pellicano et al., 2005). Other studies have contradicted the findings of altered visual functioning in individuals with ASD in a variety of visual tasks (Barton et al., 2007; Behrmann, Thomas, et al., 2006; Dakin & Frith, 2005; Smith, Kenny, Rudnicka, Briscoe, & Pellicano, 2016; Van Elst, Bach, Blessing, Riedel, & Bubl, 2015; Weigelt et al., 2012). The lack of consensus on these issues is partially due to differences in methodology, participant inclusion criteria, and a dearth of studies using stimuli of varying levels of complexity. These varied findings have led to a number of models and hypotheses regarding the relationship between ASD symptoms and visual perception; notably the EPF model (Mottron et al., 2006), the amygdala theory of autism (Schultz, 2005), and the social motivation hypothesis (Chevallier et al., 2012; Dawson, Webb, & McPartland, 2005). Despite extensive studies of visual perception in ASD, there remains controversy as to whether individuals on the spectrum process images similar to their neurotypical counterparts, and if not, where along the visual pathway those differences first become evident. The overall purpose of this dissertation was to study perception of visual stimuli of different levels of complexity in adults with ASD and a group of control participants. We sought to further examine the nature of any visual alterations between the two groups, and when one was found, determine if there was a relationship between visual processing differences and our measures of social competence and autism severity.    175 5.1 Optometric concerns in ASD The visual processing pathway is initiated when an image is projected through the lens onto the retina, with the clarity of the image depending on whether it is projected directly on the fovea region. Increased rates of refractive errors have been associated with children with ASD, although prevalence reports vary widely between studies (Black et al., 2013; Ikeda et al., 2013; Scharre & Creedon, 1992). The literature on rates of refractive errors in adults with ASD is scarce (Simmons et al., 2009). We sought to contribute to filling this gap in research by first asking if the rates of refractive errors are higher for adults with ASD compared to rates for the general population (Chapter 2). This information would allow researchers a greater understanding of the developmental trajectory for individuals with ASD. While incidence of refractive errors increases with age, it is possible that those rates remain stagnant for the ASD population as a whole, while the general population rates eventually catch up. Alternatively, the rates of refractive error can continue to climb for individuals with ASD throughout the lifespan, which would be indicated by higher rates in our adult subjects compared to the general population.  A doctor of optometry performed comprehensive optometric examinations for each participant and, for those requiring corrective lens prescriptions, confirmed the appropriate lenses were being worn. Refraction values were obtained for each eye. The results of these measurements indicate that adults with ASD show no differences in prevalence of refractive error compared to Controls. Overall, SPH, CYL, and AXIS values were not different between ASD and Control participants. There are variable definitions of clinically significant myopia in the literature, making it difficult to compare our findings to rates in the general population. The SphEq rates of myopia in our ASD group were similar to those in our Control group, regardless   176 of the criteria being used to define clinically significant myopia. Additionally, the rates of myopia for both groups in our study were significantly higher than the prevalence reported for the general population in previous studies (Vitale et al., 2008; Vitale et al., 2009; K. M. Williams et al., 2015).    Reports of prevalence of astigmatism vary in the literature and are at least partially driven by differences between studies in cut-off CYL values used to define clinically significant astigmatism. Rates and severity of astigmatism were not significantly different between our ASD and Control groups. Comparisons of our results to reported prevalence in the general population yield mixed results depending on the criteria used to define clinically significant astigmatism. Rates of astigmatism were higher for both our ASD and Control group than those reported for the general population in one study (K. M. Williams et al., 2015). When assessing the presence of astigmatism alone, meaning a Plano SPH value and a CYL value of -1.0 D or lower, we find that our ASD and Control groups have lower rates compared to those reported for the general population (Gomez-Salazar et al., 2017). In another study, rates of clinically significant astigmatism were lower for our ASD and Control groups compared to reported prevalence in the general population (Vitale et al., 2008). Taken together, our findings of both higher and lower rates of astigmatism in our ASD and Control groups suggest careful attention should be paid to inclusion criteria and classification cut-offs when making comparisons to prevalence rates in the general population. Whether the increased rate of refractive error in reported in children with ASD, and subsequent leveling off of rates in adulthood, is an inherent characteristic of the disorder or if it is a confound of different participant selection criteria between studies remains unclear. Our results indicate that adults with ASD have higher rates of myopia than in the general population.   177 A caveat to this is that our Control group also have higher rates of myopia than expected based on reports of national prevalence for refractive errors. It is possible that the preferred activities of individuals on the spectrum, which often include using computers, playing video games, and use of other small electronic devices (Engelhardt et al., 2017; Gillespie-Lynch et al., 2014; Mazurek et al., 2015), are driving the rates of increased incidences of myopia (Foster & Jiang, 2014; Ramamurthy et al., 2015). Increased time with near-focused activities may also be driving the higher rates of myopia in our Control group (Dolgin, 2015; Ramamurthy et al., 2015). Unfortunately, we cannot directly answer this question as we did not collect data from our ASD or Control participants regarding time spent in various daily activities. Future studies could address this issue by including detailed questionnaires assessing time spent in near-focused activities. As incidence of myopia increases with age and overall rates are increasing in the general population (Vitale et al., 2009), detailed analysis of a large number of participants are needed to determine what is driving the higher rates of myopia in adults with ASD.  We provide evidence that there were no significant differences between our ASD and Control groups for rates or severity of refractive error. Rates of myopia are higher for both groups compared to the general population, while rates of astigmatism are comparable. Given the findings that our ASD and Control groups both had similar rates of refractive error, our hypothesis that we would find greater incidence of refractive error in adults with ASD is not supported. As there were no significant differences between our participant groups for severity of refractive error, we also did not find evidence to support our hypothesis that levels of myopia, hyperopia, and astigmatism would be more severe in our ASD group compared to Controls. There were no significant differences between our participant groups for rates or severity of myopia when separated based on age, which does not support our final hypothesis that increased   178 rates of myopia would be evident across development. The precise source of the differences between our ASD and Control groups compared to reported prevalence in the general population is not immediately clear. Future optometric studies in ASD can address these questions by obtaining detailed measurements of refractive errors. Importantly, future research should include a large number of adult participants in a wide range of ages to determine the developmental time course for rates of refractive errors in the ASD population. Additionally, detailed accounts of daytime activities would allow for time spent in near-focused activities to be calculated. Analyses could then be run to determine the relationship between age, near-focused activities, and refractive errors. Understanding whether increased rates of mild and moderate myopia in ASD are due to lifestyle or are related to the condition itself will assist parents/care-givers and clinical psychologists in addressing the needs of this population.  In conclusion, we provide detailed analyses of refractive error measurements for adults with ASD, a group that has largely been neglected in the literature. These results contribute to understanding of  optometric issues in ASD by informing researchers that rates of refractive error in this population may be higher than general prevalence reported in the literature. Many studies assume participants have “normal or corrected-to-normal vision” without conducting a thorough eye exam. While our participants with ASD were no more likely to be wearing improper lenses than Controls (as is evident by the number of trial lenses loaned by the optometrist), this could be an issue if working with adults with more severe communication issues, as individuals with ASD are less likely to update their prescriptions as often as necessary or are less likely to comply with medical prescriptions in general (Logan et al., 2014). The fact that rates of loaned lenses were not different between our ASD and Control participants could be a reflection of the higher-functioning individuals who took part in this study being better able to communicate their needs   179 to caregivers and medical professionals. Altogether, we contribute three important findings to this area of research: 1) Adults with ASD are not more likely to have visual impairments stemming from optometric issues than typical observers, 2) poor acuity or other vision impairments due to refractive errors are not causing the impairments in higher-level visual tasks reported for individuals with ASD, and 3) compliance with prescriptions were not different for the high-functioning adults in our study, so detailed optometric exams may not be necessary with this population. However, studies with low-functioning individuals or children with ASD should include detailed measurements of acuity.   5.2 Low-level visual processing in ASD Continuing down the visual pathway, signals from the retina are sent to the visual cortex and begin being processed in V1. This cortical area is known to be involved in the earliest processing of images, including aspects of spatial frequency and orientation (De Valois et al., 1982; Enroth-Cugell & Robson, 1966; Hubel & Wiesel, 1960; Tootell et al., 1981; Vázquez et al., 2000). While numerous studies have indicated that individuals with ASD are biased towards processing fine details (Falter, Elliott, et al., 2012; Falter et al., 2010; Happe, 1999; Plaisted et al., 1999; Scharre & Creedon, 1992; Shah & Frith, 1983; Van der Hallen et al., 2015), the literature is split as to whether these results reflect atypical processing of the basic, low-level components of an image or if higher cortical areas are driving this bias (Bertone, 2005; Boeschoten et al., 2007; Kéïta et al., 2014; Scherf et al., 2008). For our adult participants with ASD, we chose to assess the state of the oblique effect, the classic finding of superior perception of cardinal angles compared to oblique ones. By limiting our focus to properties of orientation processing, we could compare our results to well-established findings for processing the most basic aspects of visual   180 orientation. We assessed the presence of qualitative and quantitative differences between participant groups to determine whether visual orientation processing for (a) discrimination, (b) veridical perception, and (c) detection was different in adults with ASD compared to their neurotypical peers. In our first experiment, we measured orientation discrimination thresholds for Gabor stimuli at eight separate reference orientations. Both cardinal and six oblique angles were tested to allow us to determine the status of the oblique effect in our participant groups. We assessed qualitative differences by comparing the overall response pattern between groups to see if the shape of the response curves were inconsistent with one another. We checked for quantitative differences by comparing the discrimination threshold estimates for each group. In both cases, we found no significant differences between the group with ASD and controls, indicating they were similarly precise at each tested orientation. Our second experiment asked participants to manually adjust a randomly oriented Gabor to a specified target orientation, allowing us to assess accuracy as a measure of veridical perception. We assessed vertical (90°), horizontal (180°), and oblique (45°) orientations to obtain measures of average standard error (accuracy) and standard deviation (precision). Both groups were superior at adjusting the stimuli to vertical and horizontal orientations compared to the oblique target orientation. Additionally, both groups consistently adjusted the Gabor to near veridical orientations for the vertical and horizontal conditions, indicating their perception of these cardinal orientations is accurate. In all three orientations tested, there was no significant difference between groups in either accuracy or bias. Finally, in our third experiment participants were asked to detect Gabor stimuli presented at four orientations (90°, 180°, 45°, 135°). Contrast detection thresholds were estimated to assess   181 the sensitivity for each tested orientation. As expected, for both groups sensitivity was significantly better for the cardinal angles compared to the two oblique orientations tested. There were no significant differences in sensitivity between participants with ASD and their neurotypical counterparts. Across all three experiments, our findings suggest that orientation processing is typical in ASD. We hypothesized qualitative differences for our ASD group in the form of deviations from the classic oblique effect, but the performance curves for each group demonstrated the classic oblique effect and no deviations were observed for the ASD group. For quantitative differences, we hypothesized that we would observe more precise discrimination, more accurate perception, and more sensitive detection for our ASD group, but we found no significant differences between groups on any of the three experimental tasks.  Thus, we do not find support for our hypotheses that there would be qualitative and quantitative differences between our ASD and control groups in orientation processing for experiments assessing orientation precision, accuracy, and sensitivity. These findings indicate that basic level orientation processing does not contribute to the enhanced perception reported for individuals with ASD, as we did not find any evidence that would suggest enhanced low-level orientation processing. If there was enhanced V1 processing in ASD, a likely outcome would be enhanced low-level orientation processing, but we find no evidence to support the latter.  These results are consistent with some (Brock et al., 2011; Schwarzkopf et al., 2014), but not all (Bertone et al., 2005; Dickinson et al., 2014) reports in the literature. There are some possible explanations for these discrepancies. ASD is not a homogenous disorder. Given the wide variability of individuals with this disorder, it is possible that different studies may be sampling different sub-populations of ASD. Our findings have raised the intriguing prospect that   182 certain sub-groups in ASD may have enhanced perception, such as those with a BDT peak. However, our results also show that these apparent enhancements do not apply to the ASD population as a whole. Second, it is possible that stimulus or task differences are driving the inconsistent findings between studies. There may not be perceptual enhancements in orientation processing, but instead these enhancements may apply to processing of specific spatial frequencies. It is possible that higher spatial-frequencies are the source of enhanced perception in ASD (Kéïta et al., 2014; Latham et al., 2013), which would explain the discrepancies between reports in the literature as some studies utilized stimuli with higher spatial frequencies while others did not. Taken together, these results indicate that orientation processing is not significantly different in adults with ASD. This important information suggests the source of altered visual perception in ASD should be sought downstream of V1 or in other aspects of V1 processing such as spatial frequency.  5.3 Object vs. Face-processing in ASD We next chose to examine perception of the most socially-salient class of visual images: the human face. Impairments in face processing for individuals with ASD have been reported for various aspects of face perception (Barton et al., 2007; Dawson, Webb, & McPartland, 2005; Falck-Ytter, 2008; Harms et al., 2010; Pallett et al., 2013; Poljac et al., 2013; Simmons et al., 2009; Weigelt et al., 2012; Wolf et al., 2008). We chose to assess identity and expression processing for our adult participants. An ability to remember previously seen faces is important for social functioning, as indicated by prosopagnosia studies (Barton, 2011; Barton, Cherkasova, Press, et al., 2004; Feigin et al., 2008). Expertise for faces has been an established finding in the   183 general population (Farah et al., 1998; Fox, Moon, et al., 2009; Maurer et al., 2002). The ability to rapidly and accurately discriminate between facial expressions contributes to social functioning (Narumoto, Okada, Sadato, Fukui, & Yonekura, 2001; Nichols et al., 2010; Sergent et al., 1992; Zhu et al., 2013). By obtaining data for processing of face identity and expressions, we gained a more complete picture of the relative strengths and weaknesses for each individual participant. Additionally, using our questionnaires, we could determine if ASD symptom severity and social competence were correlated with (a) face identification, and (b) expression discrimination.   In our first face recognition experiment, we utilized the CFMT (Duchaine & Nakayama, 2006) to assess face memory in our adults with and without ASD. We also used an inverted version of the task, which allowed assessment of the upright face advantage based on face-specific memory, which is believed to be a measure of holistic face processing. With thousands of participants, this online task has established robust population norms. While originally designed to be a screening tool for prosopagnosia, some studies have shown impaired face memory on the CFMT for some, but not all, individuals with ASD (Hedley et al., 2011; Wilson et al., 2010). In agreement with these studies, we found that adults with ASD performed lower in upright face memory and did not show as much of an advantage for upright over inverted faces (Hedley, Brewer, & Young, 2015). The magnitude of the face-specific memory was reduced in our participants with ASD, indicating this group is not utilizing face-specific expert processing strategies to the same degree as their neurotypical counterparts when identifying faces. When we examined the correlation between upright and inverted face memory performance, we found higher correlation for our ASD group. This suggests that expert processes commonly employed for face recognition are not being utilized to the same degree in individuals with ASD.    184 We next assessed face-specific perception in our second experiment by using a psychophysical identification paradigm for face and house stimuli. For this experiment, participants were shown a target stimulus before being asked to indicate which face they saw on a choice screen. Contrast threshold estimates were measured and compared for each task. Face-specific perception was calculated as the log threshold-ratio between face and house conditions. Control participants showed superior perception for faces as compared to house stimuli. Participants with ASD, however, had the opposite response pattern. Face-specific perception scores indicate ASD participants performed lower in faces compared to Controls. In the house condition, however, our ASD group demonstrated comparable performance to Controls. Thus, the face-specific ratio was significantly different from Controls, indicating that only face perception was impaired and the results were not due to a general deficit in recognition of high-level images. These findings suggest that individuals with ASD utilize less expert face processing strategies when completing face identification tasks. As these impairments in face-specific perception were found for the ASD group as a whole, these results support our hypothesis that face identification abilities would be diminished for adults with ASD. In our third experiment, we sought to determine expression discrimination thresholds using expression morphs to directly compare performance between angry, happy, and sad expressions. The presence of positive and negative expressions allowed us to obtain results capable of speaking to the amygdala theory of autism, which predicts impaired expression discrimination for negative expressions only (Adolphs et al., 1994; C. Ashwin et al., 2006). Discrimination thresholds were significantly impaired for all three expressions tested for participants with ASD. As both positive and negative expressions were impaired in our group with ASD, our results do not support the amygdala theory of autism. Our results support our   185 hypothesis that expression discrimination abilities would be impaired in ASD, but do not support our hypothesis that processing of negative expressions would demonstrate greater impairments than processing of positive expressions for our participants with ASD.  Taken together, our results indicate impaired performance in face identification and expression discrimination tasks. We next looked to measures of social competence and ASD symptom severity and examined associations between face abilities and social impairments. While individuals with ASD are impaired in both identification tasks, face memory and face-specific perception, we did not find that these abilities were related to social impairments. In our expression discrimination task, we found a significant correlation between social competence and expression discrimination abilities. There was a moderate association between the empathic concern subscale of the MSCS and ability to discriminate sad expressions. These results do not support our hypothesis that face identification abilities would be related to social competence or ASD symptom severity. Our face expression discrimination results support our hypothesis that performance on this task would be related to social competence but do not support that it is associated with ASD symptom severity. Consistent with previous research findings in ASD, our results suggest that there are deficits in various aspects of face processing. An influential review by Weigelt et al. (2012) argues that face identification differences in the literature reflect only quantitative differences, rather than a qualitatively different pattern altogether. This review found that face processing tasks with face memory demands were the most severely impacted for individuals with ASD. While our results show impairments in face processing for our ASD group as predicted by Weigelt et al. (2012) these differences are not solely quantitative. The results of our face-specific memory calculation indicate that there are at least some qualitative differences, as individuals   186 with ASD showed less upright-face advantage. Additionally, the correlation between the upright and inverted CFMT was higher for our ASD participants, indicating that individuals on the spectrum are using a more general-purpose visual strategy for face identification. This suggests there is less of a contribution from specialized face processing strategies compared to Controls. It is worth noting, however, that another review found both qualitative and quantitative differences in face identification in ASD (Tang et al., 2015). Moving our attention to face expression processing in adults with ASD, we find that face expression discrimination performance is moderately associated with social competence for sad expressions. Specifically, the empathic concern subdomain of the MSCS was correlated with expression discrimination abilities for sad expressions. Our Control participants, however, showed that both sad and angry expressions were correlated with social competence. Expression discrimination for sad expressions was associated with the social motivation and empathic concern subdomains for Controls. For angry expression discrimination in Controls, association with social competence was driven by four subdomains: social inferencing, empathic concern, social knowledge, and nonverbal sending skills. Expression discrimination abilities for happy expressions were not associated with social competence in either group. Another study found an association between reduced sensitivity to sad expressions and social responsiveness and adaptive functioning in ASD (G. L. Wallace et al., 2011). The current findings suggest that sad expression processing is impacting/being impacted by social competence in adults with ASD. Future studies can address gaps in the literature first by including multiple measures of face processing in ASD within the same participant group. Additionally, we are one of the few studies to gather assessment of intelligence, social competence, and autism severity to allow for potential associations between social impairments and face processing to be further assessed.   187 Researchers should expand expression research by including expression discrimination abilities for more expression types, specifically fear.  Altogether, our results indicate processing of face memory, face-specific perception, and face expression discrimination are all impaired in adults with ASD. While face identification is impaired, it is not associated with social impairments in ASD. The ability to discriminate face expressions, however, did show a moderate correlation with social competency measures, specifically for the empathic concern subdomain of the MSCS.  Our findings of impaired face memory and face-specific perception being unrelated to social competence are not surprising given that not all individuals who struggle with face recognition have impaired social competence. While individuals with prosopagnosia have some difficulties with social interactions due to an inability to recognize familiar faces (Dalrymple et al., 2014; Yardley et al., 2008), they are not diagnosed with autism (Duchaine et al., 2009). Additionally, while some of our adults with ASD were below the CFMT cut-off for prosopagnosia, these impairments were not associated with social competence or ASD symptom severity. These results are in agreement with another study exploring face recognition in social developmental disorders, which found heterogeneity in face recognition abilities in adults diagnosed with Asperger’s syndrome (Barton, Cherkasova, Hefter, et al., 2004). Together, our findings suggest that face recognition difficulties are not associated with overall social competence in ASD. Difficulties with face expression processing, however, can contribute to impairments in social competence (Frith, 2009). If an individual is unable to gauge a face as friendly or threatening (Adolphs et al., 2001; Clark et al., 2008; English et al., 2017), he or she will be disadvantaged in various social contexts and may not respond appropriately. Indeed, an inability   188 to reliably interpret face expressions has been associated with consequences for social functioning (Eack et al., 2015; Harms et al., 2010). It is debatable as to whether these difficulties with face expression processing are driving the social competence issues, or if the social competence difficulties lead to impaired expression processing abilities. It is possible that each contributes to the other, where impairments in one feed into the other throughout development. These results have important implications for rehabilitation efforts in ASD. There are several face training programs aimed at children with ASD, in the hopes that improving face processing abilities will translate to improved social function. For example, the Let’s Face It! skills battery includes training on a variety of face processing skills, including recognition of identity across changes in viewpoint, expression, and featural information (Tanaka et al., 2010; Wolf et al., 2008). Another face-training program for adults with ASD was developed to improve recognition skills, manipulating the second-order characteristics as foils for target faces, and found that recognition skills could be improved with training (Faja et al., 2007). The FaceSay social skills training program for children with ASD uses avatar assistants to practice recognizing faces and emotions, discriminating facial expressions, and attending to eye gaze (Hopkins et al., 2011). Both low- and high-functioning children with ASD demonstrated improved performance in emotion recognition, but only the high-functioning group showed improved face recognition skills. As correlation does not imply causation, and it remains unknown as to whether these training programs improve social competence, it is possible that the findings of the present study will not impact the success of such programs. If, however, there is a causal relationship and improvements of face processing skills do lead to improved social competence, then the present study suggests that the focus of these face-training programs should be altered to prioritize training for face expressions as opposed to development of face   189 identification skills. Additionally, while other training paradigms like Virtual Reality Social Cognition Training and The Transporters already focus on developing emotion recognition skills for individuals with ASD (Didehbani et al., 2016; Golan et al., 2010), our results indicate that training for “sad” expressions should be prioritized. If it is indeed possible to improve social skills and overall outcomes for individuals with ASD using real-life rehabilitation efforts, then such programs need to focus on the deficiencies that are most related to social competence in ASD. These results provide important, concrete recommendations for researchers developing such training programs and offer scientific evidence for specific dimensions that should be tailored to train particular face processing skills linked to social competence, thus providing avenues to improve the quality of life for individuals with ASD.  5.4 Overall conclusions This thesis sought to examine visual functioning in adults with ASD at three discrete stages of visual processing. At the eye, where the visual pathway is initiated, refractive errors result when the image is unable to focus directly on the retina in the fovea region. We found increased rates of myopia for both our ASD and Control groups compared to the general population. Overall rates of myopia and astigmatism did not differ between our participant groups. We assessed orientation processing, an aspect of vision long associated with V1, to examine the state of early cortical processing of visual stimuli. Our results found no qualitative or quantitative differences between participant groups, indicating orientation processing is unremarkable in ASD. Finally, we assessed processing of high-level, socially-relevant stimuli in three tasks: face memory, face-specific identification, and expression discrimination. Across all three tasks, we found that adults with ASD were impaired compared to their neurotypical counterparts. We also found that   190 processing of face expressions, but not identification, were correlated with social competence measures. In particular, the ability to discriminate sad expressions was moderately associated with the empathic concern subscale of the MSCS. Taken together, these three studies bridge a gap in the literature and provide concrete evidence to support, clarify, or refute a number of models of ASD.  5.4.1 Enhanced perceptual functioning models As described in Chapter 1, the EPF model approaches visual functioning in ASD as being rooted in a place of superior perception, rather than being the result of a deficit (Mottron et al., 2006). A fMRI meta-analysis conducted by Samson et al. (2012) found that in face, object, and letter visual domains there is increased cortical activity in temporal, parietal, and occipital regions for individuals with ASD, with reduced activity in the prefrontal cortex. The authors argue that these results indicate that individuals with ASD may have enhanced functional resource activity associated with visual processing, directly supporting the EPF model. It is possible that local hyperconnectivity may result in increased sensitivity to stimulation, as indicated by autism rat model studies (Rinaldi, 2008; Rinaldi, Silberberg, & Markram, 2007). Supporting these neuroimaging results in the auditory domain, pitch discrimination and sensitivity are both enhanced for individuals with ASD (Bonnel et al., 2010; Bonnel et al., 2003). Visual perception skills and measures of aloofness, a common ASD trait, were associated in males who participated in a large online study of the general population (Sabatino DiCriscio & Troiani, 2017). The authors argue their findings support the EPF as social difficulties associated with ASD were related to visual processing characteristics.   191 There is disagreement in the literature around some of the central tenants of EPF. In a large meta-analysis of visuospatial skills in ASD, Muth et al. (2014) found that there was no evidence for enhanced perception. They point to large group heterogeneity as being the driving force between the different findings in various studies. The authors of the EPF contend that this heterogeneity supports, instead of refutes, their model in a review of the literature. They suggested that the heterogeneity of relative strengths and weaknesses in processing sensory stimuli in the ASD population may have a common cognitive mechanism, possibly due to alterations in brain connectivity in perceptual areas (Mottron et al., 2013). The authors argue that individuals with autism have domain-specific abilities rooted in enhanced perception of sensory stimuli, and that heterogeneity is a characteristic of this population regardless of differential performance on various sensory tasks. While the EPF suggests that superior local processing skills are distinct from global ones, Smith et al. (2016) analyzed drawing production in children with ASD and found no broad distinctions between local and global processing or evidence of any trade-offs occurring at either level. This is contrary to the findings of Chamberlain, McManus, Riley, Rankin, and Brunswick (2013), who report that superior observational drawing was associated with local processing enhancements, directly supporting the EPF model. Our study provides evidence that is pertinent to the EPF. We did not find evidence for enhanced perception of low-level aspects of visual orientation in measures of discrimination, accuracy, or precision. These results seem to contradict the EPF, as it would predict superior processing for the most basic aspects of visual perception. This discrepancy may be driven by differences in participant selection. Our study included a heterogeneous sample of individuals with ASD with a wide range of IQ and symptom severity. Studies cited as supporting enhanced   192 visual perception in EPF have used individuals with a high IQ, many with a BDT peak (Bertone et al., 2005; Caron et al., 2006). Thus, enhanced perception may be related to specific types of intelligence, and the findings of superior visual perception may only be applicable to a subpopulation with ASD that possesses these relative peaks in intelligence. Indeed, when we separated out subjects with a BDT peak, there was a trend suggesting those individuals may demonstrate enhanced perception, but it was based on a very small sample. Our participants may be more representative of the ASD population as a whole and suggest that the EPF may be most descriptive for a subgroup of individuals with ASD. It is also possible that only certain properties of a visual stimulus are related to the superior processing of visual information as described in the EPF. The literature varies in its definitions of detail-based perception and stimuli characteristics such as basic and low-level. Our tasks and stimuli were carefully designed to examine the earliest stages of processing of orientation and did not include high spatial frequency components. If it is these higher spatial frequencies that are enabling enhanced perception in ASD, then our results contribute to the EPF by clarifying that only specific types of perception are enhanced. We suggest that the EPF is either due to a different aspect of early visual processing, such as spatial frequency, or is downstream of the earliest cortical areas processing visual stimuli.   5.4.2 Amygdala theory of autism As previously described, the first published account suggesting the amygdala’s role in face expression processing was based on a case study of an individual following cortical damage to the amygdala (Adolphs et al., 2001; Adolphs et al., 1994). Neuroimaging studies have largely supported these findings by demonstrating that faces activate the amygdala, particularly to   193 regulate motivation and attention to salient or atypical faces (Todorov, 2012). As face expressions offer important social cues, amygdala dysfunction has the potential to negatively impact social interactions.  The potential connection between amygdala dysfunction and social impairments led some researchers to suggest that abnormal amygdala function is at least a partial source for the social competence issues characteristic of this population (Baron-Cohen et al., 2000; Grelotti, Gauthier, & Schultz, 2002; Schultz, 2005). Amygdala function was directly related to anxiety and social deficits in a large group of adolescents with ASD (Herrington, Miller, Pandey, & Schultz, 2016), suggesting that social interactions are being modulated, at least in part, by amygdala input.  Tanaka and Sung (2016) argue that avoidance of the eyes, a hallmark of ASD, is caused in part by increased amygdala activity. To prevent the physiological stress response to eye contact, individuals with ASD avoid making eye contact with others. The cumulative effect of avoiding important social cues usually conveyed with the eyes throughout development results in reduced social competence.  Structural and functional connectivity studies have helped to uncover some issues surrounding amygdala dysfunction in ASD. A review by Bellani, Calderoni, Muratori, and Brambilla (2013) found that abnormal amygdala volume was related to social behavior impairments in ASD. A recent resting state analysis of adolescents and young adults with ASD indicate that functional connectivity is reduced between the amygdala and regions related to visuospatial processing (Rausch et al., 2016). Kleinhans et al. (2016) suggests that there are subregions of hyperconnectivity and hypoconnectivity between the amygdala and subcortical face processing systems.   194  Amygdala function has been associated with processing negative expressions by some authors (Adolphs et al., 1994; C. Ashwin et al., 2006). This suggests that individuals with ASD should be impaired when processing negative expressions. As predicted by the amygdala theory of autism, some studies have found impaired processing of the negative expressions of anger, fear, and sadness, but not for happiness (C. Ashwin et al., 2006; Back et al., 2016; Farran et al., 2011). The present study is at odds with these earlier findings. We found impaired face expression processing across all expressions, not just the negative ones. It has been suggested that processing in the amygdala is not limited to a particular class of expression (Todorov, 2012; Wang et al., 2017). It is important to note that methodological differences may be partially contributing these different findings, as we used expression morphs for an expression discrimination task rather than emotion recognition tasks like many of the authors of other studies. By maintaining the inherent difficulty of the task across all three expressions tested, we provide evidence that abilities are reduced for each individual expression tested. These impairments, however, are moderate and comparing performance across different expressions requires the use of highly-sensitive fine-grained tasks that allow us to equate difficulty across different types of expressions. Our results present a challenge to the amygdala hypothesis, which predicts that processing of negative expressions would be the most impaired (Schultz, 2005). Our results indicate that both positive and negative expressions are impaired in ASD. However, we do provide some potential support for the amygdala theory of autism, as we found an association between social competence and expression processing abilities, with the strongest association between discrimination abilities for sadness and the empathic concern subscale of the MSCS. Taken together, our results offer evidence relevant to the amygdala theory of autism, suggesting   195 that processing of positive and negative face expressions is impaired, but that only some of those impairments are related to deficits in social competence.  5.4.3 Social motivation hypothesis The social motivation hypothesis of autism has some overlaps with the amygdala theory of autism. While ultimately arguing for the amygdala theory of autism, Grelotti et al. (2002) were amongst the earliest to suggest the that reduced social interest may be impacting face specialization in ASD. The social motivation hypothesis posits that a lack of social motivation results in reduced attention to faces, possibly due to an inability to extract the reward value of social stimuli (Chevallier et al., 2012; Dawson et al., 2002; Dawson, Webb, & McPartland, 2005; Schultz, 2005). In both the amygdala theory of autism and the social motivation hypothesis, time is not spent socially engaging throughout development, which reduces face abilities due to a lack of adequate exposure. In the typical population, face processing skills develop throughout childhood, adolescence, and peak in early adulthood (Germine et al., 2011), with the greatest gains during childhood (Carey, Diamond, & Woods, 1980). In ASD, however, it is possible that reduced salience of social stimuli disrupts the maturation of face processing skills in early development (Klin, Shultz, & Jones, 2015). If social stimuli have less intrinsic reward value to individuals with ASD, a natural consequence would be less time spent focusing on faces. In children with ASD, pupillary diameter did not increase after viewing happy faces, indicating reduced reward response to social stimuli (Sepeta et al., 2012). Children with ASD spend a greater percentage of time fixating on images related to circumscribed interests compared to socially-relevant stimuli, suggesting that time spent engaging with face stimuli is reduced during early development (Sasson et al., 2010).   196 Participants with ASD spent equal amounts of viewing people and objects in another eye-tracking study, whereas controls spent a greater proportion focusing on people (Wilson et al., 2010). Adolescents with ASD had impaired face recognition memory for faces, but were unimpaired in recognition memory for houses (Arkush et al., 2013). Differences is social attention during infancy are related to social motivation traits in ASD (E. J. H. Jones et al., 2017). For younger siblings of children with ASD, face recognition ability in children could be predicted by face engagement during infancy (Klerk, Gliga, Charman, & Johnson, 2014). Social attention and attentional resources were assessed in two meta-analysis of eye-tracking studies in ASD and found both were altered, social attention was especially reduced if stimuli included more than one person (Chita-Tegmark, 2016a, 2016b). Taken together, these studies suggest there is reduced social attention during early development, which Klerk et al. (2014) suggests is linked to impaired face processing abilities later in life. In disagreement with the social motivation hypothesis, Garman et al. (2016) found that increased social motivation resulted in less face emotion recognition skills in a EEG study of adolescents with ASD. The authors suggest this discrepancy may result from the child faces used within their experiment, as it is possible that social motivation is focused on interactions with adults in high-functioning children on the spectrum. Therefore, they would be socially motivated to engage with adults, but still have little practice with the faces of children and perform poorly on their measure. They conclude that social motivation is a continuum within this population and should not be seen as an ‘all or none’ that applies uniformly to all situations and age groups.  The present study was designed to provide insight for this discussion by using assessments of face memory, perception, and expression discrimination. We obtained   197 assessments of social competence and ASD symptom severity for each participant, enabling us to examine the relationship between face processing abilities and social competence.  Our results offer partial support of the social motivation hypothesis, which predicts that a lack of social motivation impedes the development of visual experience with faces and leads to deficits in social competence. Our participants did show deficits in face memory, face-specific perception, and face expression discrimination, but not all were significantly correlated with social competence measures. Instead, only discrimination for sad expressions was related to a social competence measure: the demonstrating empathic concern subdomain of the MSCS. As there was not a significant correlation between social competence and face processing abilities for the other expressions tested or our measures of face identification skills, our results suggest that support for the social motivation hypothesis may be most apparent at certain levels of face processing for individuals with ASD.   5.4.4 Implications for general models of ASD The search for the source of altered visual perception in ASD has led to various models and hypotheses seeking to explain aspects of visual performance and behaviours associated with ASD. Some explanations have approached the issue from the earliest stages of the visual processing stream, suggesting that higher incidence of refractive errors or strabismus may be impacting development of visual processing skills (Coulter, 2009; Kaplan et al., 1999). While the adults with ASD in our study did show higher rates of myopia compared to reports of prevalence in the general population, the same was true of our Control participants. Thus, our findings do not support the assertion that rates of refractive error are influencing visual perception differences in ASD.   198 Other models of visual perception in ASD have suggested that the differences are couched in visual perception of low-level stimuli, such that the most basic aspects of processing for simple visual stimuli are altered in ASD (Bertone et al., 2005; Caron et al., 2006; Mottron et al., 2006). We examined three aspects of processing of orientation, believed to be associated with the earliest visual areas. Our findings of qualitative and quantitative similarities in orientation perception for adults with ASD and Controls suggests that this facet of low-level visual perception is not contributing to the differences in visual performance or behavioural results reported in the literature.  The robust (but not universal) findings of altered perception of face identity and expression in ASD have driven the proposal of a number of hypotheses and models in ASD seeking to link social deficits with face processing differences (Adolphs et al., 2001; Dawson, Webb, & McPartland, 2005; Schultz, 2005). In assessing face identification using face memory and face-specific perception in adults with ASD, we replicate previous findings of impaired performance, but did not find those impairments to be associated with social competence. Discrimination for sad expressions, however, was impaired in ASD and significantly correlated with social competence, specifically in the empathic concern subdomain. Importantly, no face impairments were linked with ASD symptom severity as measured by the AQ, suggesting that this questionnaire is not assessing symptoms related to face processing abilities.  Taken together, no single model of ASD can account for the results presented here. The source of the differences is not solely relegated to the eye or the earliest visual areas. Face processing deficits were found in all three face tasks, but only discrimination for sad expressions were significantly correlated with social competence in ASD, and none were related to symptom severity. It is possible that the face memory and face-specific perception tasks used in our study   199 are not tapping into the level of face perception related to social motivation. It is also possible that the relationship between face processing abilities and social motivation only apply to a subset of individuals on the spectrum.   5.5 Strengths and limitations This thesis is a comprehensive study of a group of participants with ASD that provides a broad picture of visual perception at three discrete levels along the visual pathway. With over eight hours of testing for each individual, we were able to obtain data using a variety of measures, which allowed us to gain a more complete understanding of our subjects. We could look for associations between levels of visual processing and social competence and symptom severity for individuals with ASD. Indeed, detailed analyses of the relationship between symptom severity and visual processing of faces are sorely lacking in the conversation on whether the social competence of an individual with ASD is impacted by visual processing differences (Tavassoli et al., 2017; Thye, Bednarz, Herringshaw, Sartin, & Kana, 2017). Additionally, there are relatively few studies of visual processing in adults with ASD compared to studies of adolescents and children (Simmons et al., 2009). By conducting a thorough analysis of a group of adults with ASD, we add important information to a large gap in the literature. Without data from across the lifespan, it is impossible to comment on developmental trajectories or develop effective treatment options to assist with ASD symptoms past adolescence. Finally, by using psychophysical paradigms, we are able to measure a participant’s individual thresholds for each task, thereby avoiding some of the ceiling and floor effects common to other experiment designs. We contribute a vital dataset to this line of research, one that is able to inform and contribute to a fundamental understanding of visual processing and social competence in ASD.   200  As with most studies of visual processing in individuals with ASD, this research is limited by our selection of high-functioning individuals on the spectrum as our subjects. The nature of rigorous psychophysics and IQ measures demand that participants are able to follow instructions, communicate effectively, and engage with tasks for long periods of time. All of these requirements are difficult for severely impacted individuals on the spectrum. A natural consequence of this is that findings from majority of studies, including this one, may not be representative of the ASD population as a whole. While all of our participants had a clinical diagnosis of ASD, eight individuals did not receive ADOS testing. We were using the ADOS to confirm diagnosis, we did not assess its sub-scores. Had we collected this information on all participants, we would have an opportunity to see if there were any correlations with the ADOS sub-scores and our face processing results. Additionally, while this thorough research involved a relatively high number of participants given the intensive eight-hour examination, larger participant groups would provide more evidence that could allow concrete findings to be established. Indeed, for our refractive error results, our power analyses revealed that we only had reasonable power to detect a significant difference between our Control group and the general population. A larger effect size and/or higher number of participants in each group could allow us to have higher power and thus be confident when rejecting the null hypothesis. These are common limitations in psychophysical experiments, which by necessity involve fewer numbers of participants than other testing approaches. Another limitation of our research is that while we had a large age range of participants, the majority of our participants were in their early 20’s. A more representative sample of adults would include a higher number of older adults.    201 5.6 Future directions The literature for visual perception with ASD is plagued with controversy regarding the earliest levels of processing and misunderstandings as to what constitutes a truly “simple” stimulus. Rigorous psychophysical studies like ours contribute to previous studies of aspects of low-level vision known to be processed early in the visual cortex; including spatial frequency, basic motion, and colour perception. It is only after a thorough analysis of these most basic of visual processes that a consensus can be reached as to whether information entering the visual processing stream is perceived differently from early on in the visual cortical pathway or are the result of processes occurring later in the visual system.  Additionally, neuroimaging studies in conjunction with psychophysical paradigms would provide invaluable insight into the various brain regions activated during a given task. Functional MRI studies offer the unique opportunity to observe different areas operate in synchrony, while EEG allows researchers to understand more about the temporal dynamics when processing stimuli of varying complexity. Thorough analysis of behavioural results compared to spatial and temporal neuroimaging findings has the potential to contribute a great deal to our understanding of visual perception in individuals with ASD. Processing of visual information is one of many aspects of perceptual processing in ASD requiring additional research. Multiple sensory modalities are indicated as being altered in ASD, with indications of atypical integration between multiple senses (Falter, Elliott, et al., 2012; Falter, Noreika, Wearden, & Bailey, 2012; Iarocci & McDonald, 2006; Noel, De Niear, Stevenson, Alais, & Wallace, 2017; O'Neill & Jones, 1997; Wiggins et al., 2009). Carefully-designed studies exploring multisensory integration and sensitivity in ASD would allow us to examine possible associations with measures of symptom severity and social competence. These   202 analyses would provide additional insight into whether reported difficulties are impacting and/or being impacted by social competence or severity of symptoms associated with ASD.  Indeed, with ASD being a disorder that impacts social functioning, it is necessary to develop treatment options that directly address the social competence difficulties experienced by this population. Training programs like Let’s Face It! (Tanaka et al., 2012), The Transporters (Golan et al., 2010), and Virtual Reality Social Cognition training (Didehbani et al., 2016) are designed with these deficiencies in mind; allowing individuals with ASD to practice social skills in a low-risk environment. While these treatments offer a path forward in ASD treatment that is specific to the social competence issues, they can be informed to address the sensory integration issues associated with this disorder. With more studies combining sensory issues and measures of symptom severity and social competence, treatment options can be designed to address the myriad of issues facing individuals with ASD, instead of focusing on just one aspect of this complex disorder. In conclusion, our findings contribute valuable insight into the nature of visual processing in ASD and provide concrete recommendations for rehabilitation efforts and training programs seeking to improve social skills and overall outcomes for individuals with this disorder.      203 Bibliography  Adolphs, R., Sears, L., & Piven, J. (2001). Abnormal processing of social information from faces in autism. J Cogn Neurosci, 13(2), 232-240.  Adolphs, R., Tranel, D., Damasio, H., & Damasio, A. (1994). Impaired recognition of emotion in facial expressions following bilateral damage to the human amygdala. Nature, 372(6507), 669-672. doi:10.1038/372669a0 Amaral, D. G., Schumann, C. M., & Nordahl, C. W. Neuroanatomy of autism. Trends in Neurosciences, 31(3), 137-145. doi:10.1016/j.tins.2007.12.005 American Psychiatric Association. (2000). Diagnostic criteria from DSM-IV-TR. Washington, D. C.: American Psychiatric Association. American Psychiatric Association (Ed.) (2013). Diagnostic and statistical manual of mental disorders (DSM-5®): American Psychiatric Pub. Anketell, P. M., Saunders, K. J., Gallagher, S., Bailey, C., & Little, J.-A. (2016). Profile of refractive errors in European Caucasian children with Autistic Spectrum Disorder; increased prevalence and magnitude of astigmatism. Ophthalmic and Physiological Optics, 36(4), 395-403. doi:10.1111/opo.12286 Anney, R., Klei, L., Pinto, D., Regan, R., Conroy, J., Magalhaes, T. R., . . . Hallmayer, J. (2010). A genome-wide scan for common alleles affecting risk for autism. Hum Mol Genet, 19(20), 4072-4082. doi:10.1093/hmg/ddq307 Appelle, S. (1972). Perception and discrimination as a function of stimulus orientation: the" oblique effect" in man and animals. Psychological bulletin, 78(4), 266.  Applegate, R. A. (2000). Limits to vision: can we do better than nature? J Refract Surg, 16(5), S547-551.  Arkush, L., Smith-Collins, A. P. R., Fiorentini, C., & Skuse, D. H. (2013). Recognition of Face and Non-Face Stimuli in Autistic Spectrum Disorder. Autism Research, 6(6), 550-560. doi:10.1002/aur.1318 Ashwin, C., Chapman, E., Colle, L., & Baron-Cohen, S. (2006). Impaired recognition of negative basic emotions in autism: a test of the amygdala theory. Soc Neurosci, 1(3-4), 349-363. doi:10.1080/17470910601040772 Ashwin, E., Ashwin, C., Rhydderch, D., Howells, J., & Baron-Cohen, S. (2009). Eagle-Eyed Visual Acuity: An Experimental Investigation of Enhanced Perception in Autism. Biological Psychiatry, 65(1), 17-21. doi:http://dx.doi.org/10.1016/j.biopsych.2008.06.012 Assaf, M., Jagannathan, K., Calhoun, V. D., Miller, L., Stevens, M. C., Sahl, R., . . . Pearlson, G. D. (2010). Abnormal functional connectivity of default mode sub-networks in autism spectrum disorder patients. NeuroImage, 53(1), 247-256. doi:10.1016/j.neuroimage.2010.05.067 Back, D. J., Francis, S. M., Skankland, E., Wasserburg, L. H., & Jacob, S. (2016). 1.4 SOCIAL INFORMATION PROCESSING IN AUTISM SPECTRUM DISORDER. Journal of the American Academy of Child & Adolescent Psychiatry, 55(10), S100-S101. doi:10.1016/j.jaac.2016.09.005 Bailey, A. J., Braeutigam, S., Jousmäki, V., & Swithenby, S. J. (2005). Abnormal activation of face processing systems at early and intermediate latency in individuals with autism spectrum disorder: a magnetoencephalographic study. European Journal of Neuroscience, 21(9), 2575-2585. doi:10.1111/j.1460-9568.2005.04061.x   204 Bakroon, A., & Lakshminarayanan, V. (2016). Visual function in autism spectrum disorders: a critical review. Clinical and Experimental Optometry, 99(4), 297-308. doi:10.1111/cxo.12383 Baron-Cohen, S., Ring, H. A., Bullmore, E. T., Wheelwright, S., Ashwin, C., & Williams, S. C. R. (2000). The amygdala theory of autism. Neuroscience & Biobehavioral Reviews, 24(3), 355-364. doi:https://doi.org/10.1016/S0149-7634(00)00011-7 Baron-Cohen, S., Wheelwright, S., Skinner, R., Martin, J., & Clubley, E. (2001). The autism-spectrum quotient (AQ): Evidence from asperger syndrome/high-functioning autism, males and females, scientists and mathematicians. Journal of Autism and Developmental Disorders, 31(1), 5-17.  Barraclough, N. E., & Perrett, D. I. (2011). From single cells to social perception. Philosophical Transactions: Biological Sciences, 366(1571), 1739-1752.  Barton, J. J. S. (2008). Structure and function in acquired prosopagnosia: lessons from a series of 10 patients with brain damage. J Neuropsychol, 2(Pt 1), 197-225.  Barton, J. J. S. (2011). Disorder of higher visual function. Curr Opin Neurol, 24(1), 1-5. doi:10.1097/WCO.0b013e328341a5c2 Barton, J. J. S., Cherkasova, M., & O'Connor, M. (2001). Covert recognition in acquired and developmental prosopagnosia. Neurology, 57(7), 1161-1168.  Barton, J. J. S., Cherkasova, M. V., Hefter, R. L., Cox, T. A., O’Connor, M., & Manoach, D. S. (2004). Are patients with social developmental disorders prosopagnosic? Perceptual heterogeneity in the Asperger and socio-emotional processing disorders. Brain, 127(8), 1706-1716. doi:10.1093/brain/awh194 Barton, J. J. S., Cherkasova, M. V., Press, D. Z., Intriligator, J. M., & O'Connor, M. (2004). Perceptual functions in prosopagnosia. Perception, 33(8), 939-956. doi:10.1068/p5243 Barton, J. J. S., Cherkasova, M. V., Press, D. Z., Intriligator, J. M., & O’Connor, M. (2003). Developmental prosopagnosia: A study of three patients. Brain and Cognition, 51(1), 12-30. doi:10.1016/s0278-2626(02)00516-x Barton, J. J. S., Hefter, R. L., Cherkasova, M. V., & Manoach, D. S. (2007). Investigations of face expertise in the social developmental disorders. Neurology, 69(9), 860-870.  Bayliss, A. P., di Pellegrino, G., & Tipper, S. P. (2005). Sex differences in eye gaze and symbolic cueing of attention. The Quarterly Journal of Experimental Psychology Section A, 58(4), 631-650. doi:10.1080/02724980443000124 Behrmann, M. (2003). Neuropsychological approaches to perceptual organization. Perception of faces, objects and scenes: Analytic and holistic processes, 295-334.  Behrmann, M., Avidan, G., Leonard, G. L., Kimchi, R., Luna, B., Humphreys, K., & Minshew, N. (2006). Configural processing in autism and its relationship to face processing. Neuropsychologia, 44(1), 110-129. doi:https://doi.org/10.1016/j.neuropsychologia.2005.04.002 Behrmann, M., Richler, J. J., Avidan, G., & Kimchi, R. (2015). Holistic face perception. In I. J. W. (Ed.) (Ed.), The Oxford Handbook of Perceptual Organization (pp. 758-774). Oxford, UK: Oxford University Press. Behrmann, M., Thomas, C., & Humphreys, K. (2006). Seeing it differently: visual processing in autism. Trends in Cognitive Sciences, 10(6), 258-264. doi:http://dx.doi.org/10.1016/j.tics.2006.05.001 Bellani, M., Calderoni, S., Muratori, F., & Brambilla, P. (2013). Brain anatomy of autism spectrum disorders II. Focus on amygdala (Vol. 22).   205 Bentin, S., Allison, T., Puce, A., Perez, E., & McCarthy, G. (1996). Electrophysiological Studies of Face Perception in Humans. Journal of cognitive neuroscience, 8(6), 551-565. doi:10.1162/jocn.1996.8.6.551 Bentin, S., & Deouell, L. Y. (2000). Structural encoding and identification in face processing: erp evidence for separate mechanisms. Cogn Neuropsychol, 17(1), 35-55. doi:10.1080/026432900380472 Benton, A. L., & Van Allen, M. W. (1968). Impairment in Facial Recognition in Patients with Cerebral Disease. Cortex, 4(4), 344-IN341. doi:http://dx.doi.org/10.1016/S0010-9452(68)80018-8 Berkley, M. A., Kitterle, F., & Watkins, D. W. (1975). Grating visibility as a function of orientation and retinal eccentricity. Vision Research, 15(2), 239-244. doi:https://doi.org/10.1016/0042-6989(75)90213-8 Bertone, A. (2005). Enhanced and diminished visuo-spatial information processing in autism depends on stimulus complexity. Brain, 128(10), 2430-2441. doi:10.1093/brain/awh561 Bertone, A., Mottron, L., Jelenic, P., & Faubert, J. (2005). Enhanced and diminished visuo-spatial information processing in autism depends on stimulus configuration. Brain, 128. doi:10.1093/brain/awh561 Bilalić, M., Langner, R., Ulrich, R., & Grodd, W. (2011). Many Faces of Expertise: Fusiform Face Area in Chess Experts and Novices. The Journal of Neuroscience, 31(28), 10206-10214. doi:10.1523/jneurosci.5727-10.2011 Black, K., McCarus, C., Collins, M. L. Z., & Jensen, A. (2013). Ocular Manifestations of Autism in Ophthalmology. Strabismus, 21(2), 98-102. doi:10.3109/09273972.2013.786733 Blake, R., Turner, L. M., Smoski, M. J., Pozdol, S. L., & Stone, W. L. (2003). Visual Recognition of Biological Motion Is Impaired in Children with Autism. Psychological Science, 14(2), 151-157.  Bloom, R. I., Friedman, I. B., & Chuck, R. S. (2010). Increasing rates of myopia: the long view. Curr Opin Ophthalmol, 21(4), 247-248. doi:10.1097/ICU.0b013e328339f1dd Boeschoten, M. A., Kenemans, J. L., Engeland, H. v., & Kemner, C. (2007). Abnormal spatial frequency processing in high-functioning children with pervasive developmental disorder (PDD). Clinical Neurophysiology, 118(9), 2076-2088. doi:10.1016/j.clinph.2007.05.004 Bonnel, A., McAdams, S., Smith, B., Berthiaume, C., Bertone, A., Ciocca, V., . . . Mottron, L. (2010). Enhanced pure-tone pitch discrimination among persons with autism but not Asperger syndrome. Neuropsychologia, 48(9), 2465-2475. doi:https://doi.org/10.1016/j.neuropsychologia.2010.04.020 Bonnel, A., Mottron, L., Peretz, I., Trudel, M., Gallun, E., & Bonnel, A.-M. (2003). Enhanced pitch sensitivity in individuals with autism: A signal detection analysis. Journal of cognitive neuroscience, 15(2), 226-235.  Boucher, J., & Lewis, V. (1992). Unfamiliar Face Recognition in Relatively Able Autistic Children. Journal of Child Psychology and Psychiatry, 33(5), 843-859. doi:10.1111/j.1469-7610.1992.tb01960.x Brainard, D. H. (1997). The psychophysics toolbox. Spatial vision, 10, 433-436.  Breidenstein, B. G. (2015). Ophthalmic Assessment in Children with Autism Spectrum Disorders. Current Ophthalmology Reports, 3(4), 219-224. doi:10.1007/s40135-015-0085-5   206 Brock, J., Xu, J. Y., & Brooks, K. R. (2011). Individual differences in visual search: Relationship to autistic traits, discrimination thresholds, and speed of processing. Perception, 40(6), 739-742.  Burack, J. A., Iarocci, G., Flanagan, T. D., & Bowler, D. M. (2004). On mosaics and melting pots: Conceptual considerations of comparison and matching strategies. Journal of Autism and Developmental Disorders, 34(1), 65-73.  Camisa, J., Randolph, B., & Sandra, L. (1977). The Effects of Temporal Modulation on the Oblique Effect in Humans. Perception, 6(2), 165-171. doi:doi:10.1068/p060165 Campbell, F. W., Kulikowski, J. J., & Levinson, J. (1966). The effect of orientation on the visual resolution of gratings. The Journal of physiology, 187(2), 427-436.  Campbell, F. W., & Robson, J. G. (1968). Application of Fourier analysis to the visibility of gratings. The Journal of physiology, 197(3), 551.  Canada, S. (2016). 2016 Census of Canada, Vancouver [Census metropolitan ara], British Columbia [Province], Topic: Ethnic origin.  Retrieved from http://www12.statcan.gc.ca/census-recensement/2016/dp-pd/prof/details/page.cfm?Lang=E&Geo1=CMACA&Code1=933&Geo2=PR&Code2=59&Data=Count&SearchText=vancouver&SearchType=Begins&SearchPR=01&B1=Ethnic%20origin&TABID=1. Carey, S., Diamond, R., & Woods, B. (1980). Development of face recognition: A maturational component? Developmental Psychology, 16(4), 257-269. doi:10.1037/0012-1649.16.4.257 Caron, M. J., Mottron, L., Berthiaume, C., & Dawson, M. (2006). Cognitive mechanisms, specificity and neural underpinnings of visuospatial peaks in autism. Brain, 129(7), 1789-1802. doi:10.1093/brain/awl072 Chamberlain, R., McManus, I. C., Riley, H., Rankin, Q., & Brunswick, N. (2013). Local processing enhancements associated with superior observational drawing are due to enhanced perceptual functioning, not weak central coherence. The Quarterly Journal of Experimental Psychology, 66(7), 1448-1466. doi:10.1080/17470218.2012.750678 Champely, S. (2012). Pwr: Basic functions for power analysis. R package version 1.1.1.  Charman, T., Young, G. S., Brian, J., Carter, A., Carver, L. J., Chawarska, K., . . . Zwaigenbaum, L. (2017). Non-ASD outcomes at 36 months in siblings at familial risk for autism spectrum disorder (ASD): A baby siblings research consortium (BSRC) study. Autism Research, 10(1), 169-178. doi:10.1002/aur.1669 Chen, X., Fu, Z., Yu, J., Ding, H., Bai, J., Chen, J., . . . Liu, H. (2016). Prevalence of amblyopia and strabismus in Eastern China: results from screening of preschool children aged 36–72 months. British Journal of Ophthalmology, 100(4), 515-519. doi:10.1136/bjophthalmol-2015-306999 Cheng, W., Rolls, E. T., Zhang, J., Sheng, W., Ma, L., Wan, L., . . . Feng, J. (2017). Functional connectivity decreases in autism in emotion, self, and face circuits identified by Knowledge-based Enrichment Analysis. NeuroImage, 148, 169-178. doi:https://doi.org/10.1016/j.neuroimage.2016.12.068 Chevallier, C., Kohls, G., Troiani, V., Brodkin, E. S., & Schultz, R. T. (2012). The social motivation theory of autism. Trends Cogn Sci, 16. doi:10.1016/j.tics.2012.02.007 Chita-Tegmark, M. (2016a). Attention Allocation in ASD: a Review and Meta-analysis of Eye-Tracking Studies. Review Journal of Autism and Developmental Disorders, 3(3), 209-223. doi:10.1007/s40489-016-0077-x   207 Chita-Tegmark, M. (2016b). Social attention in ASD: A review and meta-analysis of eye-tracking studies. Research in Developmental Disabilities, 48, 79-93. doi:https://doi.org/10.1016/j.ridd.2015.10.011 Christensen, D. L., Baio, J., Van Naarden Braun, K., Bilder, D., Charles, J., Constantino, J. N., . . . Yeargin-Allsopp, M. (2016). Prevalence and Characteristics of Autism Spectrum Disorder Among Children Aged 8 Years--Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2012. MMWR Surveill Summ, 65(3), 1-23. doi:10.15585/mmwr.ss6503a1 Churches, O., Baron-Cohen, S., & Ring, H. (2012). The psychophysiology of narrower face processing in autism spectrum conditions. NeuroReport, 23(6), 395-399. doi:10.1097/WNR.0b013e3283525bc8 Clark, T. F., Winkielman, P., & McIntosh, D. N. (2008). Autism and the extraction of emotion from briefly presented facial expressions: stumbling at the first step of empathy. Emotion, 8(6), 803-809. doi:10.1037/a0014124 Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, N.J.: L. Erlbaum Associates. Coulter, R. A. (2009). Understanding the Visual Symptoms of Individuals with Autism Spectrum Disorder (ASD). Optometry & Vision Development, 40(3).  Cousins, R. (2013). Prosopagnosia after stroke: potentials for impairment and treatment. Topics in Stroke Rehabilitation, 20, 471+.  Cranwell, M. B., Pearce, B., Loveridge, C., & Hurlbert, A. C. (2015). Performance on the Farnsworth-Munsell 100-Hue Test Is Significantly Related to Nonverbal IQ. Invest Ophthalmol Vis Sci, 56(5), 3171-3178. doi:10.1167/iovs.14-16094 Crawford, H., Moss, J., Anderson, G., M., Oliver, C., & McCleery, J., P. . (2015). Implicit Discrimination of Basic Facial Expressions of Positive/Negative Emotion in Fragile X Syndrome and Autism Spectrum Disorder. American Journal on Intellectual and Developmental Disabilities, 120(4), 328-345. doi:10.1352/1944-7558-120.4.328 Crick, R. P., & Khaw, P. T. (1997). A Textbook of Clinical Ophthalmology: A Practical Guide to Disorders of the Eyes and Their Management , 2nd Edition. Critchley, H. D., Daly, E. M., Bullmore, E. T., Williams, S. C. R., Van Amelsvoort, T., Robertson, D. M., . . . Murphy, D. G. M. (2000). The functional neuroanatomy of social behaviour: Changes in cerebral blood flow when people with autistic disorder process facial expressions. Brain, 123(11), 2203-2212. doi:10.1093/brain/123.11.2203 Daitch, J. M., & Green, D. G. (1969). Contrast sensitivity of the human peripheral retina. Vision Research, 9(8), 947-952.  Dakin, S. C., & Frith, U. (2005). Vagaries of Visual Perception in Autism. Neuron, 48(3), 497-507. doi:10.1016/j.neuron.2005.10.018 Dalrymple, K. A., Fletcher, K., Corrow, S., das Nair, R., Barton, J. J., Yonas, A., & Duchaine, B. (2014). "A room full of strangers every day": the psychosocial impact of developmental prosopagnosia on children and their families. J Psychosom Res, 77(2), 144-150. doi:10.1016/j.jpsychores.2014.06.001 Dalrymple, K. A., Oruç, I., Duchaine, B., Pancaroglu, R., Fox, C. J., Iaria, G., . . . Barton, J. J. S. (2011). The anatomic basis of the right face-selective N170 IN acquired prosopagnosia: A combined ERP/fMRI study. Neuropsychologia, 49(9), 2553-2563. doi:https://doi.org/10.1016/j.neuropsychologia.2011.05.003   208 Damasio, A. R., Damasio, H., & Van Hoesen, G. W. (1982). Prosopagnosia: anatomic basis and behavioral mechanisms. Neurology, 32(4), 331-341.  Dawson, G., Carver, L., Meltzoff, A. N., Panagiotides, H., McPartland, J., & Webb, S. J. (2002). Neural correlates of face and object recognition in young children with autism spectrum disorder, developmental delay, and typical development. Child Dev, 73. doi:10.1111/1467-8624.00433 Dawson, G., Webb, S. J., & McPartland, J. (2005). Understanding the nature of face processing impairment in autism: insights from behavioral and electrophysiological studies. Dev Neuropsychol, 27(3), 403-424. doi:10.1207/s15326942dn2703_6 Dawson, G., Webb, S. J., Wijsman, E., Schellenberg, G., Estes, A., Munson, J., & Faja, S. (2005). Neurocognitive and electrophysiological evidence of altered face processing in parents of children with autism: Implications for a model of abnormal development of social brain circuitry in autism. Dev Psychopathol, 17. doi:10.1017/S0954579405050327 De Renzi, E., Faglioni, P., Grossi, D., & Nichelli, P. (1991). Apperceptive and associative forms of prosopagnosia. Cortex, 27(2), 213-221.  De Valois, R. L., William Yund, E., & Hepler, N. (1982). The orientation and direction selectivity of cells in macaque visual cortex. Vision Research, 22(5), 531-544. doi:http://dx.doi.org/10.1016/0042-6989(82)90112-2 Denis, D., Burillon, C., Livet, M. O., & Burguière, O. (1997). [Ophthalmologic signs in children with autism]. Journal francais d'ophtalmologie, 20(2), 103-110.  DeYoe, E. A., & Van Essen, D. C. (1988). Concurrent processing streams in monkey visual cortex. Trends in Neurosciences, 11(5), 219-226. doi:http://dx.doi.org/10.1016/0166-2236(88)90130-0 Dickinson, A., Bruyns-Haylett, M., Smith, R., Jones, M., & Milne, E. (2016). Superior orientation discrimination and increased peak gamma frequency in autism spectrum conditions. Journal of Abnormal Psychology, 125(3), 412-422. doi:10.1037/abn0000148 Dickinson, A., Jones, M., & Milne, E. (2014). Oblique orientation discrimination thresholds are superior in those with a high level of autistic traits. Journal of Autism and Developmental Disorders, 44(11), 2844-2850.  Didehbani, N., Allen, T., Kandalaft, M., Krawczyk, D., & Chapman, S. (2016). Virtual Reality Social Cognition Training for children with high functioning autism. Computers in Human Behavior, 62(Supplement C), 703-711. doi:https://doi.org/10.1016/j.chb.2016.04.033 Dolgin, E. (2015). The myopia boom. Nature, 519(7543), 276.  Duchaine, B., Murray, H., Turner, M., White, S., & Garrido, L. (2009). Normal social cognition in developmental prosopagnosia. Cognitive Neuropsychology, 26(7), 620-634. doi:10.1080/02643291003616145 Duchaine, B., & Nakayama, K. (2006). The Cambridge Face Memory Test: Results for neurologically intact individuals and an investigation of its validity using inverted face stimuli and prosopagnosic participants. Neuropsychologia, 44(4), 576-585. doi:10.1016/j.neuropsychologia.2005.07.001 Eack, S. M., Mazefsky, C. A., & Minshew, N. J. (2015). Misinterpretation of facial expressions of emotion in verbal adults with autism spectrum disorder. Autism, 19(3), 308-315. doi:10.1177/1362361314520755 Ehlers, S., Nydén, A., Gillberg, C., Sandberg, A. D., Dahlgren, S. O., Hjelmquist, E., & Odén, A. (1997). Asperger Syndrome, Autism and Attention Disorders: A Comparative Study of   209 the Cognitive Profiles of 120 Children. Journal of Child Psychology and Psychiatry, 38(2), 207-217. doi:10.1111/j.1469-7610.1997.tb01855.x Emsley, H. H. (1925). Irregular astigmatism of the eye: effect of correcting lenses. Transactions of the Optical Society, 27(1), 28.  Engelhardt, C. R., Mazurek, M. O., & Hilgard, J. (2017). Pathological game use in adults with and without Autism Spectrum Disorder. PeerJ, 5, e3393.  English, M. C. W., Maybery, M. T., & Visser, T. A. W. (2017). Threatening faces fail to guide attention for adults with autistic-like traits. Autism Research, 10(2), 311-320. doi:10.1002/aur.1658 Enroth-Cugell, C., & Robson, J. G. (1966). The contrast sensitivity of retinal ganglion cells of the cat. The Journal of physiology, 187(3), 517-552.  Essock, E. A. (1980). The Oblique Effect of Stimulus Identification Considered with Respect to Two Classes of Oblique Effects. Perception, 9(1), 37-46. doi:doi:10.1068/p090037 Essock, E. A., & Lehmkuhle, S. (1982). The Oblique Effects of Pattern and Flicker Sensitivity: Implications for Mixed Physiological Input. Perception, 11(4), 441-455. doi:10.1068/p110441 Ezegwui, I. R., Lawrence, L., Aghaji, A. E., Okoye, O. I., Okoye, O., Onwasigwe, E. N., & Ebigbo, P. O. (2014). Refractive errors in children with autism in a developing country. Nigerian journal of clinical practice, 17(4), 467-470.  Faja, S., Aylward, E., Bernier, R., & Dawson, G. (2007). Becoming a Face Expert: A Computerized Face-Training Program for High-Functioning Individuals With Autism Spectrum Disorders. Dev Neuropsychol, 33(1), 1-24. doi:10.1080/87565640701729573 Falck-Ytter, T. (2008). Face inversion effects in autism: a combined looking time and pupillometric study. Autism Research, 1(5), 297-306. doi:10.1002/aur.45 Falkmer, M., Stuart, G. W., Danielsson, H., Bram, S., Lönebrink, M., & Falkmer, T. (2011). Visual Acuity in Adults with Asperger’s Syndrome: No Evidence for “Eagle-Eyed” Vision. Biological Psychiatry, 70(9), 812-816. doi:http://dx.doi.org/10.1016/j.biopsych.2011.07.025 Falter, C. M., Elliott, M. A., & Bailey, A. J. (2012). Enhanced visual temporal resolution in Autism Spectrum Disorders. PLoS ONE, 7. doi:10.1371/journal.pone.0032774 Falter, C. M., Grant, K. C. P., & Davis, G. (2010). Object-based attention benefits reveal selective abnormalities of visual integration in autism. Autism Research, 3(3), 128-136. doi:10.1002/aur.134 Falter, C. M., Noreika, V., Wearden, J. H., & Bailey, A. J. (2012). More consistent, yet less sensitive: Interval timing in autism spectrum disorders. Quarterly Journal of Experimental Psychology, 65(11), 2093-2107. doi:10.1080/17470218.2012.690770 Farah, M. J., Wilson, K. D., Drain, M., & Tanaka, J. N. (1998). What is 'special' about face perception? Psychological Review, 105(3), 482-498. doi:10.1037/0033-295X.105.3.482 Farah, M. J., Wilson, K. D., Drain, M. H., & Tanaka, J. R. (1995). The inverted face inversion effect in prosopagnosia: Evidence for mandatory, face-specific perceptual mechanisms. Vision Research, 35(14), 2089-2093. doi:http://dx.doi.org/10.1016/0042-6989(94)00273-O Farran, E. K., Branson, A., & King, B. J. (2011). Visual search for basic emotional expressions in autism; impaired processing of anger, fear and sadness, but a typical happy face advantage. Research in Autism Spectrum Disorders, 5(1), 455-462. doi:http://dx.doi.org/10.1016/j.rasd.2010.06.009   210 Fechner, G. (1860). Elemente der Psychophysik (Elements of Psychophysics)(HE Adler, Trans.). New York: Holt, Rinehart and Winston. Inc.(1860/1966).  Feigin, V. L., Barker-Collo, S., McNaughton, H., Brown, P., & Kerse, N. (2008). Long-Term Neuropsychological and Functional Outcomes in Stroke Survivors: Current Evidence and Perspectives for New Research. International Journal of Stroke, 3(1), 33-40. doi:doi:10.1111/j.1747-4949.2008.00177.x Foss-Feig, J. H., McGugin, R. W., Gauthier, I., Mash, L. E., Ventola, P., & Cascio, C. J. (2016). A functional neuroimaging study of fusiform response to restricted interests in children and adolescents with autism spectrum disorder. Journal of neurodevelopmental disorders, 8(1), 15. doi:10.1186/s11689-016-9149-6 Foster, P. J., & Jiang, Y. (2014). Epidemiology of myopia. Eye, 28(2), 202-208. doi:10.1038/eye.2013.280 Fox, C., Iaria, G., & Barton, J. J. S. (2009). Defining the face processing network: Optimization of the functional localizer in fMRI. Human Brain Mapping, 30(5), 1637-1651. doi:10.1002/hbm.20630 Fox, C., Moon, S., Iaria, G., & Barton, J. (2009). The correlates of subjective perception of identity and expression in the face network: An fMRI adaptation study. NeuroImage, 44(2), 569-580. doi:10.1016/j.neuroimage.2008.09.011 Franklin, A., Sowden, P., Burley, R., Notman, L., & Alder, E. (2008). Color perception in children with autism. J Autism Dev Disord, 38(10), 1837-1847. doi:10.1007/s10803-008-0574-6 Franklin, A., Sowden, P., Notman, L., Gonzalez-Dixon, M., West, D., Alexander, I., . . . White, A. (2010). Reduced chromatic discrimination in children with autism spectrum disorders. Developmental Science, 13(1), 188-200. doi:10.1111/j.1467-7687.2009.00869.x French, A. N., Morgan, I. G., Mitchell, P., & Rose, K. A. (2013). Patterns of myopigenic activities with age, gender and ethnicity in Sydney schoolchildren. Ophthalmic Physiol Opt, 33(3), 318-328. doi:10.1111/opo.12045 Freyberg, J., Robertson, C. E., & Baron-Cohen, S. (2016). Typical magnitude and spatial extent of crowding in autism. Journal of Vision, 16(5), 17-17. doi:10.1167/16.5.17 Friedman, D. S., Repka, M. X., Katz, J., Giordano, L., Ibironke, J., Hawse, P., & Tielsch, J. M. (2009). Prevalence of Amblyopia and Strabismus in White and African American Children Aged 6 through 71 Months: The Baltimore Pediatric Eye Disease Study. Ophthalmology, 116(11), 2128-2134.e2122. doi:https://doi.org/10.1016/j.ophtha.2009.04.034 Friston, K. J., Buechel, C., Fink, G. R., Morris, J., Rolls, E., & Dolan, R. J. (1997). Psychophysiological and modulatory interactions in neuroimaging. NeuroImage, 6(3), 218-229.  Frith, C. (2009). Role of facial expressions in social interactions. Philosophical Transactions of the Royal Society B: Biological Sciences, 364(1535), 3453-3458. doi:10.1098/rstb.2009.0142 Fujita, I., Tanaka, K., Ito, M., & Cheng, K. (1992). Columns for visual features of objects in monkey inferotemporal cortex. Nature, 360(6402), 343-346.  Fujita, T., Yamasaki, T., Kamio, Y., Hirose, S., & Tobimatsu, S. (2011). Parvocellular pathway impairment in autism spectrum disorder: Evidence from visual evoked potentials. Research in Autism Spectrum Disorders, 5(1), 277-285. doi:https://doi.org/10.1016/j.rasd.2010.04.009   211 Garman, H. D., Spaulding, C. J., Webb, S. J., Mikami, A. Y., Morris, J. P., & Lerner, M. D. (2016). Wanting it Too Much: An Inverse Relation Between Social Motivation and Facial Emotion Recognition in Autism Spectrum Disorder. Child Psychiatry & Human Development, 47(6), 890-902. doi:10.1007/s10578-015-0620-5 Gauthier, I., Skudlarski, P., Gore, J. C., & Anderson, A. W. (2000). Expertise for cars and birds recruits brain areas involved in face recognition. Nature Neuroscience, 3, 191+.  Gauthier, I., Tarr, M. J., Moylan, J., Skudlarski, P., Gore, J. C., & Anderson, A. W. (2000). The Fusiform “Face Area” is Part of a Network that Processes Faces at the Individual Level. Journal of cognitive neuroscience, 12(3), 495-504. doi:10.1162/089892900562165 Germine, L. T., Duchaine, B., & Nakayama, K. (2011). Where cognitive development and aging meet: Face learning ability peaks after age 30. Cognition, 118(2), 201-210. doi:https://doi.org/10.1016/j.cognition.2010.11.002 Gillespie-Lynch, K., Kapp, S. K., Shane-Simpson, C., Smith, D. S., & Hutman, T. (2014). Intersections Between the Autism Spectrum and the Internet: Perceived Benefits and Preferred Functions of Computer-Mediated Communication. Intellectual and Developmental Disabilities, 52(6), 456-469. doi:10.1352/1934-9556-52.6.456 Giovedí, S., Corradi, A., Fassio, A., & Benfenati, F. (2014). Involvement of Synaptic Genes in the Pathogenesis of Autism Spectrum Disorders: The Case of Synapsins. Frontiers in Pediatrics, 2, 94. doi:10.3389/fped.2014.00094 Gliga, T., Bedford, R., Charman, T., Johnson, Mark H., Baron-Cohen, S., Bolton, P., . . . Tucker, L. (2015). Enhanced Visual Search in Infancy Predicts Emerging Autism Symptoms. Current Biology, 25(13), 1727-1730. doi:http://dx.doi.org/10.1016/j.cub.2015.05.011 Gobbini, M. I., & Haxby, J. V. (2007). Neural systems for recognition of familiar faces. Neuropsychologia, 45(1), 32-41. doi:http://dx.doi.org/10.1016/j.neuropsychologia.2006.04.015 Golan, O., Ashwin, E., Granader, Y., McClintock, S., Day, K., Leggett, V., & Baron-Cohen, S. (2010). Enhancing emotion recognition in children with autism spectrum conditions: an intervention using animated vehicles with real emotional faces. J Autism Dev Disord, 40(3), 269-279. doi:10.1007/s10803-009-0862-9 Gomez-Salazar, F., Campos-Romero, A., Gomez-Campaña, H., Cruz-Zamudio, C., Chaidez-Felix, M., Leon-Sicairos, N., . . . Canizalez-Roman, A. (2017). Refractive errors among children, adolescents and adults attending eye clinics in Mexico. International Journal of Ophthalmology, 10(5), 796-802. doi:10.18240/ijo.2017.05.23 Goodale, M. A., & Milner, A. D. (1992). Separate visual pathways for perception and action. Trends in Neurosciences, 15(1), 20-25. doi:http://dx.doi.org/10.1016/0166-2236(92)90344-8 Goodale, M. A., Milner, A. D., Jakobson, L. S., & Carey, D. P. (1991). A neurological dissociation between perceiving objects and grasping them. Nature, 349(6305), 154-156. doi:10.1038/349154a0 Gowen, E., & Hamilton, A. (2013). Motor abilities in autism: a review using a computational context. Journal of Autism and Developmental Disorders, 43(2), 323-344. doi:10.1007/s10803-012-1574-0 Grelotti, D. J., Gauthier, I., & Schultz, R. T. (2002). Social interest and the development of cortical face specialization: what autism teaches us about face processing. Dev Psychobiol, 40. doi:10.1002/dev.10028   212 Griffiths, S., Jarrold, C., Penton-Voak, I. S., Woods, A. T., Skinner, A. L., & Munafò, M. R. (2017). Impaired Recognition of Basic Emotions from Facial Expressions in Young People with Autism Spectrum Disorder: Assessing the Importance of Expression Intensity. Journal of Autism and Developmental Disorders. doi:10.1007/s10803-017-3091-7 Grill-Spector, K. (2003). The neural basis of object perception. Curr Opin Neurobiol, 13(2), 159-166.  Grill-Spector, K., & Malach, R. (2004). The Human Visual Cortex. Annual Review of Neuroscience, 27(1), 649-677. doi:10.1146/annurev.neuro.27.070203.144220 Guillon, Q., Hadjikhani, N., Baduel, S., & Rogé, B. (2014). Visual social attention in autism spectrum disorder: Insights from eye tracking studies. Neuroscience & Biobehavioral Reviews, 42, 279-297. doi:https://doi.org/10.1016/j.neubiorev.2014.03.013 Gunton, K. B., Wasserman, B. N., & DeBenedictis, C. (2015). Strabismus. Primary Care: Clinics in Office Practice, 42(3), 393-407. doi:http://dx.doi.org/10.1016/j.pop.2015.05.006 Guo, X., Duan, X., Long, Z., Chen, H., Wang, Y., Zheng, J., . . . Chen, H. (2016). Decreased amygdala functional connectivity in adolescents with autism: A resting-state fMRI study. Psychiatry Research: Neuroimaging, 257, 47-56. doi:https://doi.org/10.1016/j.pscychresns.2016.10.005 Guo, X. M., Oruc, I., & Barton, J. J. (2009). Cross-orientation transfer of adaptation for facial identity is asymmetric: a study using contrast-based recognition thresholds. Vision Res, 49(18), 2254-2260. doi:10.1016/j.visres.2009.06.012 Guy, J., Mottron, L., Berthiaume, C., & Bertone, A. (2016). The developmental trajectory of contrast sensitivity in autism spectrum disorder. Autism Research, 9(8), 866-878. doi:10.1002/aur.1579 Hadjikhani, N., Chabris, C. F., Joseph, R. M., Clark, J., McGrath, L., Aharon, I., . . . Harris, G. J. (2004). Early visual cortex organisation in autism: an fMRI study. NeuroReport, 15. doi:10.1097/00001756-200402090-00011 Hadjikhani, N., Joseph, R. M., Snyder, J., & Tager-Flusberg, H. (2007). Abnormal activation of the social brain during face perception in autism. Human Brain Mapping, 28(5), 441-449. doi:10.1002/hbm.20283 Happe, F. (1996). Studying weak central coherence at low levels: children with autism do not succumb to visual illusions. A research note. Journal of Child Psychology and Psychiatry, 37(7), 873-877.  Happe, F. (1999). Autism: cognitive deficit or cognitive style? Trends in Cognitive Sciences, 3(6), 216-222. doi:http://dx.doi.org/10.1016/S1364-6613(99)01318-2 Happe, F., & Frith, U. (2006). The weak coherence account: detail-focused cognitive style in autism spectrum disorders. J Autism Dev Disord, 36(1), 5-25. doi:10.1007/s10803-005-0039-0 Harms, M. B., Martin, A., & Wallace, G. L. (2010). Facial Emotion Recognition in Autism Spectrum Disorders: A Review of Behavioral and Neuroimaging Studies. Neuropsychology Review, 20(3), 290-322. doi:10.1007/s11065-010-9138-6 Haxby, J. V., Hoffman, E. A., & Gobbini, M. I. (2000). The distributed human neural system for face perception. Trends in Cognitive Sciences, 4(6), 223-233. doi:http://dx.doi.org/10.1016/S1364-6613(00)01482-0   213 Heaton, P., Ludlow, A., & Roberson, D. (2008). When less is more: Poor discrimination but good colour memory in autism. Research in Autism Spectrum Disorders, 2(1), 147-156. doi:https://doi.org/10.1016/j.rasd.2007.04.004 Hedley, D., Brewer, N., & Young, R. (2011). Face recognition performance of individuals with Asperger syndrome on the Cambridge Face Memory Test. Autism Res, 4(6), 449-455. doi:10.1002/aur.214 Hedley, D., Brewer, N., & Young, R. (2015). The effect of inversion on face recognition in adults with autism spectrum disorder. J Autism Dev Disord, 45(5), 1368-1379. doi:10.1007/s10803-014-2297-1 Heidary, G., Ying, G. S., Maguire, M. G., & Young, T. L. (2005). The association of astigmatism and spherical refractive error in a high myopia cohort. Optom Vis Sci, 82(4), 244-247.  Herrington, J. D., Miller, J. S., Pandey, J., & Schultz, R. T. (2016). Anxiety and social deficits have distinct relationships with amygdala function in autism spectrum disorder. Social Cognitive & Affective Neuroscience, 11(6), 907-914. doi:10.1093/scan/nsw015 Herrington, J. D., Riley, M. E., Grupe, D. W., & Schultz, R. T. (2015). Successful face recognition is associated with increased prefrontal cortex activation in autism spectrum disorder. J Autism Dev Disord, 45(4), 902-910. doi:10.1007/s10803-014-2233-4 Hoffman, E. A., & Haxby, J. V. (2000). Distinct representations of eye gaze and identity in the distributed human neural system for face perception. Nature Neuroscience, 3, 80+.  Hoffmann, E., Brück, C., Kreifelts, B., Ethofer, T., & Wildgruber, D. (2016). Reduced functional connectivity to the frontal cortex during processing of social cues in autism spectrum disorder. Journal of Neural Transmission, 123(8), 937-947. doi:10.1007/s00702-016-1544-3 Hole, G. J. (1994). Configurational Factors in the Perception of Unfamiliar Faces. Perception, 23(1), 65-74. doi:doi:10.1068/p230065 Hopkins, I. M., Gower, M. W., Perez, T. A., Smith, D. S., Amthor, F. R., Wimsatt, F. C., & Biasini, F. J. (2011). Avatar assistant: improving social skills in students with an ASD through a computer-based intervention. J Autism Dev Disord, 41(11), 1543-1555. doi:10.1007/s10803-011-1179-z Howard, M. A., Cowell, P. E., Boucher, J., Broks, P., Mayes, A., Farrant, A., & Roberts, N. (2000). Convergent neuroanatomical and behavioural evidence of an amygdala hypothesis of autism. NeuroReport, 11(13), 2931-2935.  Hubel, D. H., & Wiesel, T. N. (1960). Receptive fields of optic nerve fibres in the spider monkey. J Physiol, 154, 572-580.  Hubel, D. H., & Wiesel, T. N. (1968). Receptive fields and functional architecture of monkey striate cortex. J Physiol, 195(1), 215-243.  Humphreys, K., Hasson, U., Avidan, G., Minshew, N., & Behrmann, M. (2008). Cortical patterns of category-selective activation for faces, places and objects in adults with autism. Autism Research, 1(1), 52-63. doi:10.1002/aur.1 Iarocci, G., & Armstrong, K. (2014). Age-Related Changes in Conjunctive Visual Search in Children with and without ASD. Autism Research, 7(2), 229-236. doi:10.1002/aur.1359 Iarocci, G., Burack, J. A., Shore, D. I., Mottron, L., & Enns, J. T. (2006). Global–Local Visual Processing in High Functioning Children with Autism: Structural vs. Implicit Task Biases. Journal of Autism and Developmental Disorders, 36(1), 117-129. doi:10.1007/s10803-005-0045-2   214 Iarocci, G., & McDonald, J. (2006). Sensory Integration and the Perceptual Experience of Persons with Autism. Journal of Autism and Developmental Disorders, 36(1), 77-90. doi:10.1007/s10803-005-0044-3 Ikeda, J., Davitt, B. V., Ultmann, M., Maxim, R., & Cruz, O. A. (2013). Brief Report: Incidence of Ophthalmologic Disorders in Children with Autism. Journal of Autism and Developmental Disorders, 43(6), 1447-1451. doi:10.1007/s10803-012-1475-2 Ishai, A., Schmidt, C. F., & Boesiger, P. (2005). Face perception is mediated by a distributed cortical network. Brain Research Bulletin, 67(1–2), 87-93. doi:http://dx.doi.org/10.1016/j.brainresbull.2005.05.027 Ishida, R., Kamio, Y., & Nakamizo, S. (2009). PERCEPTUAL DISTORTIONS OF VISUAL ILLUSIONS IN CHILDREN WITH HIGH-FUNCTIONING AUTISM SPECTRUM DISORDER. PSYCHOLOGIA, 52(3), 175-187. doi:10.2117/psysoc.2009.175 Ishitobi, M., Kosaka, H., Omori, M., Matsumura, Y., Munesue, T., Mizukami, K., . . . Wada, Y. (2011). Differential amygdala response to lower face in patients with autistic spectrum disorders: An fMRI study. Research in Autism Spectrum Disorders, 5(2), 910-919. doi:https://doi.org/10.1016/j.rasd.2010.10.005 Jemel, B., Mimeault, D., Saint-Amour, D., Hosein, A., & Mottron, L. (2010). VEP contrast sensitivity responses reveal reduced functional segregation of mid and high filters of visual channels in autism. J Vis, 10. doi:10.1167/10.6.13 Jiang, X., Bollich, A., Cox, P., Hyder, E., James, J., Gowani, S. A., . . . Riesenhuber, M. (2013). A quantitative link between face discrimination deficits and neuronal selectivity for faces in autism. Neuroimage Clin, 2, 320-331. doi:10.1016/j.nicl.2013.02.002 Jolliffe, T., & Baron-Cohen, S. (1997). Are people with autism and Asperger syndrome faster than normal on the Embedded Figures Test? Journal of Child Psychology and Psychiatry, 38(5), 527-534.  Jones, E. J. H., Venema, K., Earl, R. K., Lowy, R., & Webb, S. J. (2017). Infant social attention: an endophenotype of ASD-related traits? Journal of Child Psychology and Psychiatry, 58(3), 270-281. doi:10.1111/jcpp.12650 Jones, W., & Klin, A. (2013). Attention to eyes is present but in decline in 2-6-month-old infants later diagnosed with autism. Nature, 504(7480), 427-431. doi:10.1038/nature12715 Joseph, J. E., Zhu, X., Gundran, A., Davies, F., Clark, J. D., Ruble, L., . . . Bhatt, R. S. (2015). Typical and atypical neurodevelopment for face specialization: an FMRI study. J Autism Dev Disord, 45(6), 1725-1741. doi:10.1007/s10803-014-2330-4 Kabatas, E. U., Ozer, P. A., Ertugrul, G. T., Kurtul, B. E., Bodur, S., & Alan, B. E. (2015). Initial Ophthalmic Findings in Turkish Children with Autism Spectrum Disorder. Journal of Autism and Developmental Disorders, 45(8), 2578-2581. doi:10.1007/s10803-015-2428-3 Kanner, L. (1943). Autistic disturbances of affective contact.  Kanwisher, N., McDermott, J., & Chun, M. M. (1997). The Fusiform Face Area: A Module in Human Extrastriate Cortex Specialized for Face Perception. The Journal of Neuroscience, 17(11), 4302-4311.  Kanwisher, N., & Yovel, G. (2006). The fusiform face area: a cortical region specialized for the perception of faces. Philosophical Transactions of the Royal Society B: Biological Sciences, 361(1476), 2109-2128. doi:10.1098/rstb.2006.1934 Kaplan, M., Rimland, B., & Edelson, S. M. (1999). Strabismus in autism spectrum disorder. Focus on Autism and Other Developmental Disabilities, 14(2), 101-105.    215 Kawuma, M., & Mayeku, R. (2002). A survey of the prevalence of refractive errors among children in lower primary schools in Kampala district. African Health Sciences, 2(2), 69-72.  Kéïta, L., Guy, J., Berthiaume, C., Mottron, L., & Bertone, A. (2014). An early origin for detailed perception in Autism Spectrum Disorder: biased sensitivity for high-spatial frequency information. Scientific reports, 4, 5475.  Kéïta, L., Mottron, L., & Bertone, A. (2010). Far visual acuity is unremarkable in autism: Do we need to focus on crowding? Autism Research, 3(6), 333-341. doi:10.1002/aur.164 Kendon, A. (1967). Some functions of gaze-direction in social interaction. Acta Psychologica, 26, 22-63. doi:https://doi.org/10.1016/0001-6918(67)90005-4 Kim, S.-Y., Choi, U.-S., Park, S.-Y., Oh, S.-H., Yoon, H.-W., Koh, Y.-J., . . . Lee, C.-U. (2015). Abnormal Activation of the Social Brain Network in Children with Autism Spectrum Disorder: An fMRI Study. Psychiatry Investigation, 12(1), 37-45. doi:10.4306/pi.2015.12.1.37 Kleinhans, N. M., Reiter, M. A., Neuhaus, E., Pauley, G., Martin, N., Dager, S., & Estes, A. (2016). Subregional differences in intrinsic amygdala hyperconnectivity and hypoconnectivity in autism spectrum disorder. Autism Research, 9(7), 760-772. doi:10.1002/aur.1589 Kleinhans, N. M., Richards, T., Johnson, L. C., Weaver, K. E., Greenson, J., Dawson, G., & Aylward, E. (2011). fMRI evidence of neural abnormalities in the subcortical face processing system in ASD. NeuroImage, 54(1), 697-704. doi:https://doi.org/10.1016/j.neuroimage.2010.07.037 Kleinhans, N. M., Richards, T., Weaver, K., Johnson, L. C., Greenson, J., Dawson, G., & Aylward, E. (2010). Association between amygdala response to emotional faces and social anxiety in autism spectrum disorders. Neuropsychologia, 48(12), 3665-3670. doi:https://doi.org/10.1016/j.neuropsychologia.2010.07.022 Kleinstein, R. N., Jones, L. A., Hullett, S., & et al. (2003). Refractive error and ethnicity in children. Archives of Ophthalmology, 121(8), 1141-1147. doi:10.1001/archopht.121.8.1141 Klerk, C. C. J. M., Gliga, T., Charman, T., & Johnson, M. H. (2014). Face engagement during infancy predicts later face recognition ability in younger siblings of children with autism. Developmental Science, 17(4), 596-611. doi:10.1111/desc.12141 Klin, A., Shultz, S., & Jones, W. (2015). Social visual engagement in infants and toddlers with autism: Early developmental transitions and a model of pathogenesis. Neuroscience and biobehavioral reviews, 0, 189-203. doi:10.1016/j.neubiorev.2014.10.006 Koh, H. C., Milne, E., & Dobkins, K. (2010). Spatial Contrast Sensitivity in Adolescents with Autism Spectrum Disorders. Journal of Autism and Developmental Disorders, 40(8), 978-987. doi:10.1007/s10803-010-0953-7 Koretz, J. F., Cook, C. A., & Kaufman, P. L. (1997). Accommodation and presbyopia in the human eye. Changes in the anterior segment and crystalline lens with focus. Investigative Ophthalmology & Visual Science, 38(3), 569-578.  Koretz, J. F., Cook, C. A., & Kaufman, P. L. (2001). Aging of the human lens: changes in lens shape at zero-diopter accommodation. Journal of the Optical Society of America A, 18(2), 265-272. doi:10.1364/JOSAA.18.000265   216 Kracke, I. (1994). DEVELOPMENTAL PROSOPAGNOSIA IN ASPERGER SYNDROME: PRESENTATION AND DISCUSSION OF AN INDIVIDUAL CASE. Developmental Medicine & Child Neurology, 36(10), 873-886. doi:10.1111/j.1469-8749.1994.tb11778.x Latham, K., Chung, S. L., Allen, P. M., Tavassoli, T., & Baron-Cohen, S. (2013). Spatial localisation in autism: evidence for differences in early cortical visual processing. Molecular Autism, 4(1), 4. doi:10.1186/2040-2392-4-4 Le Couteur, A., Bailey, A., Goode, S., Pickles, A., Gottesman, I., Robertson, S., & Rutter, M. (1996). A Broader Phenotype of Autism: The Clinical Spectrum in Twins. Journal of Child Psychology and Psychiatry, 37(7), 785-801. doi:10.1111/j.1469-7610.1996.tb01475.x Leekam, S., Nieto, C., Libby, S., Wing, L., & Gould, J. (2007). Describing the Sensory Abnormalities of Children and Adults with Autism. Journal of Autism and Developmental Disorders, 37(5), 894-910. doi:10.1007/s10803-006-0218-7 Lennie, P. (1974). Head orientation and meridional variations in acuity. Vision Research, 14(1), 107-111. doi:https://doi.org/10.1016/0042-6989(74)90123-0 Leonard, C. M., Rolls, E. T., Wilson, F. A. W., & Baylis, G. C. (1985). Neurons in the amygdala of the monkey with responses selective for faces. Behavioural Brain Research, 15(2), 159-176. doi:http://dx.doi.org/10.1016/0166-4328(85)90062-2 Leopold, D. A., Bondar, I. V., & Giese, M. A. (2006). Norm-based face encoding by single neurons in the monkey inferotemporal cortex. Nature, 442, 572.  Liu, R. R., Pancaroglu, R., Hills, C. S., Duchaine, B., & Barton, J. J. S. (2016). Voice Recognition in Face-Blind Patients. Cerebral Cortex (New York, NY), 26(4), 1473-1487. doi:10.1093/cercor/bhu240 Logan, S. L., Carpenter, L., Leslie, R. S., Hunt, K. S., Garrett-Mayer, E., Charles, J., & Nicholas, J. S. (2014). Rates and predictors of adherence to psychotropic medications in children with autism spectrum disorders. J Autism Dev Disord, 44(11), 2931-2948. doi:10.1007/s10803-014-2156-0 Lord, C., Risi, S., Lambrecht, L., Cook, E. H., Leventhal, B. L., DiLavore, P. C., . . . Rutter, M. (2000). The Autism Diagnostic Observation Schedule—Generic: A Standard Measure of Social and Communication Deficits Associated with the Spectrum of Autism. Journal of Autism and Developmental Disorders, 30(3), 205-223. doi:10.1023/a:1005592401947 Lord, C., Risi, S., Lambrecht, L., Cook, E. H., Leventhal, B. L., DiLavore, P. C., & Rutter, M. (2000). The autism diagnostic observation schedule-generic: a standard measure of social and communication deficits associated with the spectrum of autism. J Autism Dev Disord, 30. doi:10.1023/A:1005592401947 Lundqvist, D., & Litton, J. E. (1998). The Averaged Karolinksa Directed Emotional Faces–AKDEF. CD ROM from Clinical Neuroscience, psychology section, Karolinska Instituet.  Malach, R., Reppas, J. B., Benson, R. R., Kwong, K. K., Jiang, H., Kennedy, W. A., . . . Tootell, R. B. (1995). Object-related activity revealed by functional magnetic resonance imaging in human occipital cortex. Proceedings of the National Academy of Sciences of the United States of America, 92(18), 8135-8139.  Manning, C., Tibber, M. S., Charman, T., Dakin, S. C., & Pellicano, E. (2015). Enhanced Integration of Motion Information in Children With Autism. The Journal of Neuroscience, 35(18), 6979-6986. doi:10.1523/jneurosci.4645-14.2015   217 Manning, C., Tibber, M. S., & Dakin, S. C. (2017). Visual integration of direction and orientation information in autistic children. Autism & Developmental Language Impairments, 2, 2396941517694626. doi:10.1177/2396941517694626 Maurer, D., Le Grand, R., & Mondloch, C. J. (2002). The many faces of configural processing. Trends in Cognitive Sciences, 6(6), 255-260.  Mazurek, M. O., Engelhardt, C. R., & Clark, K. E. (2015). Video games from the perspective of adults with autism spectrum disorder. Comput. Hum. Behav., 51(PA), 122-130. doi:10.1016/j.chb.2015.04.062 McKean-Cowdin, R., Cotter, S. A., Tarczy-Hornoch, K., Wen, G., Kim, J., Borchert, M., & Varma, R. (2013). Prevalence of Amblyopia or Strabismus in Asian and Non-Hispanic White Preschool Children: Multi-Ethnic Pediatric Eye Disease Study. Ophthalmology, 120(10), 2117-2124. doi:https://doi.org/10.1016/j.ophtha.2013.03.001 McKernan, E. P., Russo, N., Burnette, C., & Kates, W. R. (2017). ASD concordance of twins across DSM-IV-TR and DSM-5 diagnostic criteria. Research in Autism Spectrum Disorders, 41-42(Supplement C), 51-56. doi:https://doi.org/10.1016/j.rasd.2017.08.004 McKone, E., Kanwisher, N., & Duchaine, B. C. (2007). Can generic expertise explain special processing for faces? Trends in Cognitive Sciences, 11(1), 8-15. doi:10.1016/j.tics.2006.11.002 McPartland, J. C., Reichow, B., & Volkmar, F. R. (2012). Sensitivity and specificity of proposed DSM-5 diagnostic criteria for autism spectrum disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 51(4), 368-383.  McPartland, J. C., Webb, S. J., Keehn, B., & Dawson, G. (2011). Patterns of visual attention to faces and objects in autism spectrum disorder. J Autism Dev Disord, 41(2), 148-157. doi:10.1007/s10803-010-1033-8 Meilleur, A. A. S., Berthiaume, C., Bertone, A., & Mottron, L. (2014). Autism-Specific Covariation in Perceptual Performances: "g'' or "p'' Factor? PLoS ONE, 9(8), e103781. doi:10.1371/journal.pone.0103781 Meinhardt-Injac, B., Persike, M., Imhof, M., & Meinhardt, G. (2015). The sensitivity to replacement and displacement of the eyes region in early adolescence, young and later adulthood. Frontiers in Psychology, 6(1164). doi:10.3389/fpsyg.2015.01164 Milne, E., Swettenham, J., Hansen, P., Campbell, R., Jeffries, H., & Plaisted, K. (2002). High motion coherence thresholds in children with autism. Journal of Child Psychology and Psychiatry, 43(2), 255-263. doi:10.1111/1469-7610.00018 Milner, A. D., & Goodale, M. A. (2008). Two visual systems re-viewed. Neuropsychologia, 46(3), 774-785. doi:http://dx.doi.org/10.1016/j.neuropsychologia.2007.10.005 Mishkin, M., Ungerleider, L. G., & Macko, K. A. (1983). Object vision and spatial vision: two cortical pathways. Trends in Neurosciences, 6, 414-417. doi:http://dx.doi.org/10.1016/0166-2236(83)90190-X Mitchell, D. E., Freeman, R. D., & Westheimer, G. (1967). Effect of Orientation on the Modulation Sensitivity for Interference Fringes on the Retina*. Journal of the Optical Society of America, 57(2), 246-249. doi:10.1364/JOSA.57.000246 Mondloch, C., Le Grand, R., & Maurer, D. (2003). Early visual experience is necessary for the development of some—but not all—aspects of face processing. Monk, C. (2010). Neural circuitry of emotional face processing in autism spectrum disorders. Journal of Psychiatry and Neuroscience, 35(2), 105-114. doi:10.1503/jpn.090085   218 Moore, B. D., Ausburger, A. R., Ciner, E. B., Cockrell, D. A., Fern, K. D., & Harb, E. (2008). Optometric Clinical Practice Guideline: Care of the patient with hyperopia. St. Louis, MO: American Optometric Association. Moore, D. J., Heavey, L., & Reidy, J. (2012). Attentional processing of faces in ASD: a Dot-Probe study. J Autism Dev Disord, 42(10), 2038-2045. doi:10.1007/s10803-012-1449-4 Moscovitch, M., Winocur, G., & Behrmann, M. (1997, 1997/09//). What is special about face recognition? Nineteen experiments on a person with visual object agnosia and dyslexia but normal face recognition. Journal of cognitive neuroscience, 9, 555+. Mottron, L., & Belleville, S. (1993). A study of perceptual analysis in a high-level autistic subject with exceptional graphic abilities. Brain Cogn, 23(2), 279-309. doi:10.1006/brcg.1993.1060 Mottron, L., Bouvet, L., Bonnel, A., Samson, F., Burack, J. A., Dawson, M., & Heaton, P. (2013). Veridical mapping in the development of exceptional autistic abilities. Neuroscience & Biobehavioral Reviews, 37(2), 209-228. doi:https://doi.org/10.1016/j.neubiorev.2012.11.016 Mottron, L., Dawson, M., Soulières, I., Hubert, B., & Burack, J. (2006). Enhanced Perceptual Functioning in Autism: An Update, and Eight Principles of Autistic Perception. Journal of Autism and Developmental Disorders, 36(1), 27-43. doi:10.1007/s10803-005-0040-7 Mukamel, R., & Fried, I. (2012). Human intracranial recordings and cognitive neuroscience. Annu Rev Psychol, 63, 511-537. doi:10.1146/annurev-psych-120709-145401 Muth, A., Honekopp, J., & Falter, C. M. (2014). Visuo-spatial performance in autism: a meta-analysis. J Autism Dev Disord, 44(12), 3245-3263. doi:10.1007/s10803-014-2188-5 Narumoto, J., Okada, T., Sadato, N., Fukui, K., & Yonekura, Y. (2001). Attention to emotion modulates fMRI activity in human right superior temporal sulcus. Cognitive Brain Research, 12(2), 225-231.  Navon, D. (1977). Forest before trees: The precedence of global features in visual perception. Cognitive Psychology, 9(3), 353-383. doi:http://dx.doi.org/10.1016/0010-0285(77)90012-3 Nele, D., Ellen, D., Petra, W., & Herbert, R. (2015). Social information processing in infants at risk for ASD at 5 months of age: The influence of a familiar face and direct gaze on attention allocation. Research in Autism Spectrum Disorders, 17, 95-105. doi:https://doi.org/10.1016/j.rasd.2015.06.006 Nichols, D. F., Betts, L. R., & Wilson, H. R. (2010). Decoding of Faces and Face Components in Face-Sensitive Human Visual Cortex. Frontiers in Psychology, 1, 28. doi:10.3389/fpsyg.2010.00028 Nickl-Jockschat, T., Rottschy, C., Thommes, J., Schneider, F., Laird, A. R., Fox, P. T., & Eickhoff, S. B. (2015). Neural networks related to dysfunctional face processing in autism spectrum disorder. Brain structure & function, 220(4), 2355-2371. doi:10.1007/s00429-014-0791-z Noel, J.-P., De Niear, M. A., Stevenson, R., Alais, D., & Wallace, M. T. (2017). Atypical rapid audio-visual temporal recalibration in autism spectrum disorders. Autism Research, 10(1), 121-129. doi:10.1002/aur.1633 Nomi, J. S., & Uddin, L. Q. (2015). Face processing in autism spectrum disorders: from brain regions to brain networks. Neuropsychologia, 71, 201-216. doi:10.1016/j.neuropsychologia.2015.03.029   219 O'Neill, M., & Jones, R. S. P. (1997). Sensory-perceptual abnormalities in autism: a case for more research? Journal of Autism and Developmental Disorders, 27(3), 283-293.  O'Riordan, M. A. (2004). Superior Visual Search in Adults with Autism. Autism, 8(3), 229-248. doi:10.1177/1362361304045219 O'Riordan, M. A., Plaisted, K. C., Driver, J., & Baron-Cohen, S. (2001). Superior visual search in autism. Journal of Experimental Psychology: Human Perception and Performance, 27(3), 719-730. doi:10.1037/0096-1523.27.3.719 O’Hearn, K., Schroer, E., Minshew, N., & Luna, B. (2010). Lack of developmental improvement on a face memory task during adolescence in autism. Neuropsychologia, 48(13), 3955-3960. doi:https://doi.org/10.1016/j.neuropsychologia.2010.08.024 Orsmond, G. I., & Kuo, H.-Y. (2011). The daily lives of adolescents with an autism spectrum disorder. Autism, 15(5), 579-599. doi:10.1177/1362361310386503 Oruc, I., & Barton, J. J. S. (2010a). Critical frequencies in the perception of letters, faces, and novel shapes: Evidence for limited scale invariance for faces. Journal of Vision, 10(12), 20-20. doi:10.1167/10.12.20 Oruc, I., & Barton, J. J. S. (2010b). A novel face aftereffect based on recognition contrast thresholds. Vision Research, 50(18), 1845-1854. doi:10.1016/j.visres.2010.06.005 Oruc, I., Krigolson, O., Dalrymple, K., Nagamatsu, L. S., Handy, T. C., & Barton, J. J. (2011). Bootstrap analysis of the single subject with event related potentials. Cogn Neuropsychol, 28(5), 322-337. doi:10.1080/02643294.2011.648176 Pallett, P. M., Cohen, S. J., & Dobkins, K. R. (2013). Face and Object Discrimination in Autism, and Relationship to IQ and Age. Journal of Autism and Developmental Disorders, 44(5), 1039-1054. doi:10.1007/s10803-013-1955-z Pan, C. W., Ramamurthy, D., & Saw, S. M. (2012). Worldwide prevalence and risk factors for myopia. Ophthalmic Physiol Opt, 32(1), 3-16. doi:10.1111/j.1475-1313.2011.00884.x Pelli, D. G. (1997). The VideoToolbox software for visual psychophysics: transforming numbers into movies. Spat Vis, 10(4), 437-442.  Pellicano, E., Gibson, L., Maybery, M., Durkin, K., & Badcock, D. R. (2005). Abnormal global processing along the dorsal visual pathway in autism: a possible mechanism for weak visuospatial coherence? Neuropsychologia, 43(7), 1044-1053. doi:10.1016/j.neuropsychologia.2004.10.003 Peters, H. B. (1961). VISION SCREENING WITH A SNELLEN CHART*. Optometry and Vision Science, 38(9), 487-505.  Pinsk, M. A., DeSimone, K., Moore, T., Gross, C. G., & Kastner, S. (2005). Representations of faces and body parts in macaque temporal cortex: A functional MRI study. Proceedings of the National Academy of Sciences of the United States of America, 102(19), 6996-7001. doi:10.1073/pnas.0502605102 Pitcher, D., Walsh, V., & Duchaine, B. (2011). The role of the occipital face area in the cortical face perception network. Experimental Brain Research, 209(4), 481-493. doi:10.1007/s00221-011-2579-1 Plaisted, K., O'Riordan, M., & Baron-Cohen, S. (1998a). Enhanced Visual Search for a Conjunctive Target in Autism: A Research Note. Journal of Child Psychology and Psychiatry, 39(5), 777-783. doi:10.1111/1469-7610.00376 Plaisted, K., O'Riordan, M., & Baron-Cohen, S. (1998b). Enhanced discrimination of novel, highly similar stimuli by adults with autism during a perceptual learning task. Journal of Child Psychology and Psychiatry, 39(5), 765-775.    220 Plaisted, K., Saksida, L., Alcantara, J., & Weisblatt, E. (2003). Towards an Understanding of the Mechanisms of Weak Central Coherence Effects: Experiments in Visual Configural Learning and Auditory Perception. Philosophical Transactions: Biological Sciences, 358(1430), 375-386.  Plaisted, K., Swettenham, J., & Rees, L. (1999). Children with autism show local precedence in a divided attention task and global precedence in a selective attention task. Journal of Child Psychology and Psychiatry, 40(05), 733-742.  Poljac, E., Poljac, E., & Wagemans, J. (2013). Reduced accuracy and sensitivity in the perception of emotional facial expressions in individuals with high autism spectrum traits. Autism, 17(6), 668-680. doi:10.1177/1362361312455703 Poole, J. H., Tobias, F. C., & Vinogradov, S. (2000). The functional relevance of affect recognition errors in schizophrenia. Journal of the International Neuropsychological Society, 6(6), 649-658. doi:undefined Puce, A., Allison, T., Bentin, S., Gore, J. C., & McCarthy, G. (1998). Temporal Cortex Activation in Humans Viewing Eye and Mouth Movements. The Journal of Neuroscience, 18(6), 2188-2199.  Puglia, M. H., & Morris, J. P. (2017). Neural Response to Biological Motion in Healthy Adults Varies as a Function of Autistic-Like Traits. Frontiers in Neuroscience, 11(404). doi:10.3389/fnins.2017.00404 Quiroga, R. Q., Reddy, L., Kreiman, G., Koch, C., & Fried, I. (2005). Invariant visual representation by single neurons in the human brain. Nature, 435(7045), 1102-1107. doi:http://www.nature.com/nature/journal/v435/n7045/suppinfo/nature03687_S1.html Ramamurthy, D., Lin Chua, S. Y., & Saw, S. M. (2015). A review of environmental risk factors for myopia during early life, childhood and adolescence. Clin Exp Optom, 98(6), 497-506. doi:10.1111/cxo.12346 Ramessur, R., Williams, K. M., & Hammond, C. J. (2015). Risk factors for myopia in a discordant monozygotic twin study. Ophthalmic and Physiological Optics, 35(6), 643-651. doi:10.1111/opo.12246 Rausch, A., Zhang, W., Haak, K. V., Mennes, M., Hermans, E. J., van Oort, E., . . . Groen, W. B. (2016). Altered functional connectivity of the amygdaloid input nuclei in adolescents and young adults with autism spectrum disorder: a resting state fMRI study. Molecular Autism, 7.  Resnikoff, S., Pascolini, D., Mariotti, S. P., & Pokharel, G. P. (2008). Global magnitude of visual impairment caused by uncorrected refractive errors in 2004. Bulletin of the World Health Organization, 86, 63-70.  Rey, H. G., Ison, M. J., Pedreira, C., Valentin, A., Alarcon, G., Selway, R., . . . Quian Quiroga, R. (2015). Single-cell recordings in the human medial temporal lobe. Journal of Anatomy, 227(4), 394-408. doi:10.1111/joa.12228 Rinaldi, T. (2008). Hyper-connectivity and hyper-plasticity in the medial prefrontal cortex in the valproic acid animal model of autism. Frontiers in Neural Circuits, 2. doi:10.3389/neuro.04.004.2008 Rinaldi, T., Silberberg, G., & Markram, H. (2007). Hyperconnectivity of Local Neocortical Microcircuitry Induced by Prenatal Exposure to Valproic Acid. Cerebral Cortex, 18(4), 763-770. doi:10.1093/cercor/bhm117   221 Rivest, J. B., Jemel, B., Bertone, A., McKerral, M., & Mottron, L. (2013). Luminance- and texture-defined information processing in school-aged children with autism. PLoS ONE, 8(10), e78978. doi:10.1371/journal.pone.0078978 Roberts, J., & Rowland, M. (1978). Refraction status and motility defects of persons 4-74 years, United States, 1971-1972.  Roe, Anna W., Chelazzi, L., Connor, Charles E., Conway, Bevil R., Fujita, I., Gallant, Jack L., . . . Vanduffel, W. (2012). Toward a Unified Theory of Visual Area V4. Neuron, 74(1), 12-29. doi:https://doi.org/10.1016/j.neuron.2012.03.011 Romanski, L. M., & Diehl, M. M. (2011). Neurons responsive to face-view in the primate ventrolateral prefrontal cortex. Neuroscience, 189, 223-235. doi:http://dx.doi.org/10.1016/j.neuroscience.2011.05.014 Rosenberg, R. E., Law, J., Yenokyan, G., McGready, J., Kaufmann, W. E., & Law, P. A. (2009). Characteristics and concordance of autism spectrum disorders among 277 twin pairs. Archives of Pediatrics & Adolescent Medicine, 163(10), 907-914. doi:10.1001/archpediatrics.2009.98 Rossi, E. A., & Roorda, A. (2010). The relationship between visual resolution and cone spacing in the human fovea. Nature Neuroscience, 13, 156+.  Rossion, B. (2008a). Constraining the cortical face network by neuroimaging studies of acquired prosopagnosia. NeuroImage, 40(2), 423-426.  Rossion, B. (2008b). Picture-plane inversion leads to qualitative changes of face perception. Acta Psychologica, 128(2), 274-289. doi:http://dx.doi.org/10.1016/j.actpsy.2008.02.003 Rossion, B. (2013). The composite face illusion: A whole window into our understanding of holistic face perception. Visual Cognition, 21(2), 139-253. doi:10.1080/13506285.2013.772929 Rossion, B., Caldara, R., Seghier, M., Schuller, A. M., Lazeyras, F., & Mayer, E. (2003). A network of occipito-temporal face-sensitive areas besides the right middle fusiform gyrus is necessary for normal face processing. Brain, 126(11), 2381-2395. doi:10.1093/brain/awg241 Rossion, B., Hanseeuw, B., & Dricot, L. (2012). Defining face perception areas in the human brain: A large-scale factorial fMRI face localizer analysis. Brain and Cognition, 79(2), 138-157. doi:10.1016/j.bandc.2012.01.001 Rossion, B., & Jacques, C. (2008). Does physical interstimulus variance account for early electrophysiological face sensitive responses in the human brain? Ten lessons on the N170. NeuroImage, 39(4), 1959-1979. doi:10.1016/j.neuroimage.2007.10.011 Rossion, B., Joyce, C. A., Cottrell, G. W., & Tarr, M. J. (2003). Early lateralization and orientation tuning for face, word, and object processing in the visual cortex. NeuroImage, 20(3), 1609-1624. doi:http://dx.doi.org/10.1016/j.neuroimage.2003.07.010 Russell, R., Duchaine, B., & Nakayama, K. (2009). Super-recognizers: People with extraordinary face recognition ability. Psychonomic Bulletin & Review, 16(2), 252-257. doi:10.3758/pbr.16.2.252 Rutherford, M. D., Walsh, J. A., & Lee, V. (2015). Brief Report: Infants Developing with ASD Show a Unique Developmental Pattern of Facial Feature Scanning. J Autism Dev Disord, 45(8), 2618-2623. doi:10.1007/s10803-015-2396-7 Sabatino DiCriscio, A., & Troiani, V. (2017). Brief Report: Autism-like Traits are Associated With Enhanced Ability to Disembed Visual Forms. Journal of Autism and Developmental Disorders, 47(5), 1568-1576. doi:10.1007/s10803-017-3053-0   222 Sah, P., Faber, E. S. L., Lopez De Armentia, M., & Power, J. (2003). The Amygdaloid Complex: Anatomy and Physiology. Physiological Reviews, 83(3), 803-834. doi:10.1152/physrev.00002.2003 Said, C. P., Moore, C. D., Engell, A. D., Todorov, A., & Haxby, J. V. (2010). Distributed representations of dynamic facial expressions in the superior temporal sulcus. Journal of Vision, 10(5), 11-11. doi:10.1167/10.5.11 Samson, F., Mottron, L., Soulières, I., & Zeffiro, T. A. (2012). Enhanced visual functioning in autism: An ALE meta-analysis. Human Brain Mapping, 33(7), 1553-1581. doi:10.1002/hbm.21307 Sasson, N. J., Elison, J. T., Turner-Brown, L. M., Dichter, G. S., & Bodfish, J. W. (2010). Brief Report: Circumscribed Attention in Young Children with Autism. Journal of Autism and Developmental Disorders, 41(2), 242-247. doi:10.1007/s10803-010-1038-3 Sato, W., Sawada, R., Uono, S., Yoshimura, S., Kochiyama, T., Kubota, Y., . . . Toichi, M. (2017). Impaired detection of happy facial expressions in autism. Scientific reports, 7(1), 13340.  Scharre, J. E., & Creedon, M. P. (1992). Assessment of visual function in autistic children. Optometry & Vision Science, 69(6), 433-439.  Schauder, K. B., Park, W. J., Tadin, D., & Bennetto, L. (2017). Larger Receptive Field Size as a Mechanism Underlying Atypical Motion Perception in Autism Spectrum Disorder. Clinical Psychological Science, 5(5), 827-842. doi:10.1177/2167702617707733 Scherf, K. S., Luna, B., Kimchi, R., Minshew, N., & Behrmann, M. (2008). Missing the big picture: impaired development of global shape processing in autism. Autism Res, 1(2), 114-129. doi:10.1002/aur.17 Schiefer, U., Kraus, C., Baumbach, P., Ungewiß, J., & Michels, R. (2016). Refractive errors: Epidemiology, Effects and Treatment Options. Deutsches Ärzteblatt International, 113(41), 693-702. doi:10.3238/arztebl.2016.0693 Schultz, R. T. (2005). Developmental deficits in social perception in autism: the role of the amygdala and fusiform face area. Int J Dev Neurosci, 23(2-3), 125-141. doi:10.1016/j.ijdevneu.2004.12.012 Schultz, R. T., Gauthier, I., Klin, A., Fulbright, R. K., Anderson, A. W., Volkmar, F., & Gore, J. C. (2000). Abnormal ventral temporal cortical activity during face discrimination among individuals with autism and Asperger syndrome. Arch Gen Psychiatry, 57. doi:10.1001/archpsyc.57.4.331 Schwarzkopf, D. S., Anderson, E. J., de Haas, B., White, S. J., & Rees, G. (2014). Larger Extrastriate Population Receptive Fields in Autism Spectrum Disorders. The Journal of Neuroscience, 34(7), 2713-2724. doi:10.1523/jneurosci.4416-13.2014 Senju, A., & Johnson, M. H. (2009). The eye contact effect: mechanisms and development. Trends in Cognitive Sciences, 13(3), 127-134. doi:https://doi.org/10.1016/j.tics.2008.11.009 Senju, A., Vernetti, A., Ganea, N., Hudry, K., Tucker, L., Charman, T., & Johnson, Mark H. (2015). Early Social Experience Affects the Development of Eye Gaze Processing. Current Biology, 25(23), 3086-3091. doi:https://doi.org/10.1016/j.cub.2015.10.019 Sepeta, L., Tsuchiya, N., Davies, M. S., Sigman, M., Bookheimer, S. Y., & Dapretto, M. (2012). Abnormal social reward processing in autism as indexed by pupillary responses to happy faces. Journal of neurodevelopmental disorders, 4(1), 1.    223 Sergent, J., Ohta, S., & Macdonald, B. (1992). Functional neuroanatomy of face and object processing. Brain, 115(1), 15-36.  Servadio, M., Vanderschuren, L. J., & Trezza, V. (2015). Modeling autism-relevant behavioral phenotypes in rats and mice: Do 'autistic' rodents exist? Behav Pharmacol, 26(6), 522-540. doi:10.1097/fbp.0000000000000163 Services, W. P. (2001). ADOS Manual. Shafai, F., Armstrong, K., Iarocci, G., & Oruc, I. (2015). Visual orientation processing in autism spectrum disorder: No sign of enhanced early cortical function. Journal of Vision, 15(18), 1-15. doi:10.1167/15.15.18 Shah, A., & Frith, U. (1983). AN ISLET OF ABILITY IN AUTISTIC CHILDREN: A RESEARCH NOTE. Journal of Child Psychology and Psychiatry, 24(4), 613-620. doi:10.1111/j.1469-7610.1983.tb00137.x Shepherd, S. (2010). Following Gaze: Gaze-Following Behavior as a Window into Social Cognition. Frontiers in Integrative Neuroscience, 4(5). doi:10.3389/fnint.2010.00005 Sherwin, J. C., Reacher, M. H., Keogh, R. H., Khawaja, A. P., Mackey, D. A., & Foster, P. J. (2012). The association between time spent outdoors and myopia in children and adolescents: a systematic review and meta-analysis. Ophthalmology, 119(10), 2141-2151. doi:10.1016/j.ophtha.2012.04.020 Siegel, D. J., Minshew, N. J., & Goldstein, G. (1996). Wechsler IQ profiles in diagnosis of high-functioning autism. Journal of Autism and Developmental Disorders, 26(4), 389-406. doi:10.1007/bf02172825 Simmons, D. R., Robertson, A. E., McKay, L. S., Toal, E., McAleer, P., & Pollick, F. E. (2009). Vision in autism spectrum disorders. Vision Research, 49(22), 2705-2739. doi:10.1016/j.visres.2009.08.005 Smerdon, D. (2000). Anatomy of the eye and orbit. Current Anaesthesia & Critical Care, 11(6), 286-292. doi:http://dx.doi.org/10.1054/cacc.2000.0296 Smith, A. D., Kenny, L., Rudnicka, A., Briscoe, J., & Pellicano, E. (2016). Drawing Firmer Conclusions: Autistic Children Show No Evidence of a Local Processing Bias in a Controlled Copying Task. Journal of Autism and Developmental Disorders, 46(11), 3481-3492. doi:10.1007/s10803-016-2889-z Szatmari, P., Paterson, A. D., Zwaigenbaum, L., Roberts, W., Brian, J., Liu, X.-Q., . . . Senman, L. (2007). Mapping autism risk loci using genetic linkage and chromosomal rearrangements. Nature Genetics, 39(3), 319-328.  Tanaka, J. W., & Farah, M. J. (1993). Parts and wholes in face recognition. Q J Exp Psychol A, 46(2), 225-245.  Tanaka, J. W., & Sung, A. (2016). The “eye avoidance” hypothesis of autism face processing. Journal of Autism and Developmental Disorders, 46(5), 1538-1552. doi:10.1007/s10803-013-1976-7 Tanaka, J. W., Wolf, J. M., Klaiman, C., Koenig, K., Cockburn, J., Herlihy, L., . . . Schultz, R. T. (2010). Using computerized games to teach face recognition skills to children with autism spectrum disorder: the Let's Face It! program. J Child Psychol Psychiatry, 51(8), 944-952. doi:10.1111/j.1469-7610.2010.02258.x Tanaka, J. W., Wolf, J. M., Klaiman, C., Koenig, K., Cockburn, J., Herlihy, L., . . . Schultz, R. T. (2012). The perception and identification of facial emotions in individuals with autism spectrum disorders using the Let’s Face It! Emotion Skills Battery. Journal of Child Psychology and Psychiatry, 53(12), 1259-1267. doi:10.1111/j.1469-7610.2012.02571.x   224 Tang, J., Falkmer, M., Horlin, C., Tan, T., Vaz, S., & Falkmer, T. (2015). Face Recognition and Visual Search Strategies in Autism Spectrum Disorders: Amending and Extending a Recent Review by Weigelt et al. PLoS ONE, 10(8), e0134439. doi:10.1371/journal.pone.0134439 Tavassoli, T., Latham, K., Bach, M., Dakin, S. C., & Baron-Cohen, S. (2011). Psychophysical measures of visual acuity in autism spectrum conditions. Vision Research, 51(15), 1778-1780. doi:10.1016/j.visres.2011.06.004 Tavassoli, T., Miller, L. J., Schoen, S. A., Jo Brout, J., Sullivan, J., & Baron-Cohen, S. (2017). Sensory reactivity, empathizing and systemizing in autism spectrum conditions and sensory processing disorder. Developmental Cognitive Neuroscience. doi:https://doi.org/10.1016/j.dcn.2017.05.005 Thompson, P. (1980). Margaret Thatcher: a new illusion. Perception, 9(4), 483-484.  Thye, M. D., Bednarz, H. M., Herringshaw, A. J., Sartin, E. B., & Kana, R. K. (2017). The impact of atypical sensory processing on social impairments in autism spectrum disorder. Developmental Cognitive Neuroscience. doi:https://doi.org/10.1016/j.dcn.2017.04.010 Tick, B., Bolton, P., Happé, F., Rutter, M., & Rijsdijk, F. (2016). Heritability of autism spectrum disorders: a meta-analysis of twin studies. Journal of Child Psychology and Psychiatry, 57(5), 585-595. doi:10.1111/jcpp.12499 Todorov, A. (2012). The role of the amygdala in face perception and evaluation. Motivation and Emotion, 36(1), 16-26. doi:10.1007/s11031-011-9238-5 Tootell, R., Dale, A., Sereno, M., & Malach, R. (1996). New images from human visual cortex. Trends in Neurosciences, 19(11), 481-489. doi:10.1016/s0166-2236(96)10053-9 Tootell, R., Silverman, M., & De Valois, R. (1981). Spatial Frequency Columns in Primary Visual Cortex. Science, 214(4522), 813-815.  Tye, C., Battaglia, M., Bertoletti, E., Ashwood, K. L., Azadi, B., Asherson, P., . . . McLoughlin, G. (2014). Altered neurophysiological responses to emotional faces discriminate children with ASD, ADHD and ASD + ADHD. Biological Psychology, 103, 125-134. doi:https://doi.org/10.1016/j.biopsycho.2014.08.013 Uljarevic, M., & Hamilton, A. (2013). Recognition of Emotions in Autism: A Formal Meta-Analysis. Journal of Autism and Developmental Disorders, 43(7), 1517-1526. doi:10.1007/s10803-012-1695-5 Valentine, T. (1988). Upside-down faces: A review of the effect of inversion upon face recognition. British journal of psychology, 79(4), 471-491.  Van Belle, G., De Graef, P., Verfaillie, K., Rossion, B., & Lefèvre, P. (2010). Face inversion impairs holistic perception: Evidence from gaze-contingent stimulation. Journal of Vision, 10(5), 10-10. doi:10.1167/10.5.10 Van der Hallen, R., Evers, K., Brewaeys, K., Van den Noortgate, W., & Wagemans, J. (2015). Global processing takes time: A meta-analysis on local–global visual processing in ASD. Psychological bulletin, 141(3), 549-573. doi:10.1037/bul0000004 Van Elst, L. T., Bach, M., Blessing, J., Riedel, A., & Bubl, E. (2015). Normal Visual Acuity and Electrophysiological Contrast Gain in Adults with High-Functioning Autism Spectrum Disorder. Frontiers in Human Neuroscience.  Vázquez, P., Cano, M., & Acuña, C. (2000). Discrimination of line orientation in humans and monkeys. Journal of Neurophysiology, 83(5), 2639-2648.    225 Vitale, S., Ellwein, L., Cotch, M., Ferris, F. L., Iii, & Sperduto, R. (2008). Prevalence of refractive error in the united states, 1999-2004. Archives of Ophthalmology, 126(8), 1111-1119. doi:10.1001/archopht.126.8.1111 Vitale, S., Sperduto, R. D., Ferris, F. L., & Iii. (2009). Increased prevalence of myopia in the united states between 1971-1972 and 1999-2004. Archives of Ophthalmology, 127(12), 1632-1639. doi:10.1001/archophthalmol.2009.303 Vogan, V. M., Morgan, B. R., Lee, W., Powell, T. L., Smith, M. L., & Taylor, M. J. (2014). The neural correlates of visuo-spatial working memory in children with autism spectrum disorder: effects of cognitive load. Journal of neurodevelopmental disorders, 6, 19.  Vogels, R. (1999). Categorization of complex visual images by rhesus monkeys. Part 2: single-cell study. European Journal of Neuroscience, 11(4), 1239-1255. doi:10.1046/j.1460-9568.1999.00531.x Vogels, R., & Orban, G. A. (1986). Decision processes in visual discrimination of line orientation. Journal of Experimental Psychology: Human Perception and Performance, 12(2), 115.  Volkmar, F. R., & McPartland, J. C. (2014). From Kanner to DSM-5: autism as an evolving diagnostic concept. Annual review of clinical psychology, 10, 193-212.  Wada, Y., & Yamamoto, T. (2001). Selective impairment of facial recognition due to a haematoma restricted to the right fusiform and lateral occipital region. Journal of Neurology, Neurosurgery, and Psychiatry, 71(2), 254-257. doi:10.1136/jnnp.71.2.254 Wallace, G. L., Case, L. K., Harms, M. B., Silvers, J. A., Kenworthy, L., & Martin, A. (2011). Diminished Sensitivity to Sad Facial Expressions in High Functioning Autism Spectrum Disorders is Associated with Symptomatology and Adaptive Functioning. Journal of Autism and Developmental Disorders, 41(11), 1475-1486. doi:10.1007/s10803-010-1170-0 Wallace, S., Coleman, M., & Bailey, A. (2008a). Face and object processing in autism spectrum disorders. Autism Research, 1(1), 43-51. doi:doi:10.1002/aur.7 Wallace, S., Coleman, M., & Bailey, A. (2008b). An investigation of basic facial expression recognition in autism spectrum disorders. Cognition and Emotion, 22(7), 1353-1380. doi:10.1080/02699930701782153 Wallace, S., Sebastian, C., Pellicano, E., Parr, J., & Bailey, A. (2010). Face processing abilities in relatives of individuals with ASD. Autism Research, 3(6), 345-349. doi:10.1002/aur.161 Walsh, J. A., Creighton, S. E., & Rutherford, M. D. (2016). Emotion Perception or Social Cognitive Complexity: What Drives Face Processing Deficits in Autism Spectrum Disorder? Journal of Autism and Developmental Disorders, 46(2), 615-623. doi:10.1007/s10803-015-2606-3 Wang, S., Yu, R., Tyszka, J. M., Zhen, S., Kovach, C., Sun, S., . . . Rutishauser, U. (2017). The human amygdala parametrically encodes the intensity of specific facial emotions and their categorical ambiguity. Nature Communications, 8, 14821. doi:10.1038/ncomms14821 Warrington, E. K. (1984). Recognition Memory Test: Rmt.(Words). Test Booklet 1: NFER-Nelson Publishing Company. Watson, A. B., & Pelli, D. G. (1983). QUEST: A Bayesian adaptive psychometric method. Perception & psychophysics, 33(2), 113-120.    226 Wechsler, D., & Hsiao-pin, C. (2011). WASI-II: Wechsler abbreviated scale of intelligence: Pearson. Wechsler, D., & Zhou, X. (2011). WASI-II (Wechsler Abbreviated Scale of Intelligence – second edition) manual. Bloomington MN: Pearson. Weigelt, S., Koldewyn, K., & Kanwisher, N. (2012). Face identity recognition in autism spectrum disorders: A review of behavioral studies. Neuroscience & Biobehavioral Reviews, 36(3), 1060-1084. doi:10.1016/j.neubiorev.2011.12.008 Westheimer, G., & Beard, B. L. (1998). Orientation dependency for foveal line stimuli: detection and intensity discrimination, resolution, orientation discrimination and vernier acuity. Vision Res, 38(8), 1097-1103.  Wieckowski, A. T., & White, S. W. (2017). Eye-Gaze Analysis of Facial Emotion Recognition and Expression in Adolescents with ASD. Journal of Clinical Child & Adolescent Psychology, 46(1), 110-124. doi:10.1080/15374416.2016.1204924 Wiggins, L. D., Robins, D. L., Bakeman, R., & Adamson, L. B. (2009). Breif report: sensory abnormalities as distinguishing symptoms of autism spectrum disorders in young children. Journal of Autism and Developmental Disorders, 39(7), 1087-1091.  Williams, K. M., Verhoeven, V. J. M., Cumberland, P., Bertelsen, G., Wolfram, C., Buitendijk, G. H. S., . . . Hammond, C. J. (2015). Prevalence of refractive error in Europe: the European Eye Epidemiology (E3) Consortium. European Journal of Epidemiology, 30(4), 305-315. doi:10.1007/s10654-015-0010-0 Williams, M. A., Morris, A. P., McGlone, F., Abbott, D. F., & Mattingley, J. B. (2004). Amygdala responses to fearful and happy facial expressions under conditions of binocular suppression. J Neurosci, 24(12), 2898-2904. doi:10.1523/jneurosci.4977-03.2004 Wilmer, J. B., Germine, L., Chabris, C. F., Chatterjee, G., Williams, M., Loken, E., . . . Duchaine, B. (2010). Human face recognition ability is specific and highly heritable. Proceedings of the National Academy of Sciences, 107(11), 5238-5241. doi:10.1073/pnas.0913053107 Wilson, C. E., Brock, J., & Palermo, R. (2010). Attention to social stimuli and facial identity recognition skills in autism spectrum disorder. Journal of Intellectual Disability Research, 54(12), 1104-1115. doi:10.1111/j.1365-2788.2010.01340.x Winston, J. S., Henson, R. N. A., Fine-Goulden, M. R., & Dolan, R. J. (2004). fMRI-Adaptation Reveals Dissociable Neural Representations of Identity and Expression in Face Perception. Journal of Neurophysiology, 92(3), 1830-1839. doi:10.1152/jn.00155.2004 Wolf, J. M., Tanaka, J. W., Klaiman, C., Cockburn, J., Herlihy, L., Brown, C., . . . Schultz, R. T. (2008). Specific impairment of face-processing abilities in children with autism spectrum disorder using theLet's Face It!skills battery. Autism Research, 1(6), 329-340. doi:10.1002/aur.56 Wolff, S. (1991). Asperger's syndrome. Archives of Disease in Childhood, 66(2), 178-179. doi:10.1136/adc.66.2.178 Yager, J., & Iarocci, G. (2013). The development of the Multidimensional Social Competence Scale: a standardized measure of social competence in autism spectrum disorders. Autism Res, 6(6), 631-641. doi:10.1002/aur.1331 Yang, N., Shafai, F., & Oruc, I. (2014). Size determines whether specialized expert processes are engaged for recognition of faces. J Vis, 14(8), 17. doi:10.1167/14.8.17   227 Yardley, L., McDermott, L., Pisarski, S., Duchaine, B., & Nakayama, K. (2008). Psychosocial consequences of developmental prosopagnosia: a problem of recognition. J Psychosom Res, 65(5), 445-451. doi:10.1016/j.jpsychores.2008.03.013 Yin, R. K. (1969). Looking at upside-down faces. Journal of Experimental Psychology, 81(1), 141-145. doi:10.1037/h0027474 Young, A., W. , Hellawell, D., & Hay, D. C. (1987). Configurational Information in Face Perception. Perception, 42(11), 1166-1178. doi:doi:10.1068/p160747n Young, S. G., Slepian, M. L., Wilson, J. P., & Hugenberg, K. (2014). Averted eye-gaze disrupts configural face encoding. Journal of Experimental Social Psychology, 53, 94-99. doi:https://doi.org/10.1016/j.jesp.2014.03.002 Zachi, E. C., Costa, T. L., Barboni, M. T. S., Costa, M. F., Bonci, D. M. O., & Ventura, D. F. (2017). Color Vision Losses in Autism Spectrum Disorders. Frontiers in Psychology, 8(1127). doi:10.3389/fpsyg.2017.01127 Zeck, G. M., Xiao, Q., & Masland, R. H. (2005). The spatial filtering properties of local edge detectors and brisk–sustained retinal ganglion cells. European Journal of Neuroscience, 22(8), 2016-2026. doi:10.1111/j.1460-9568.2005.04390.x Zeki, S., Watson, J., Lueck, C., Friston, K., Kennard, C., & Frackowiak, R. (1991). A direct demonstration of functional specialization in human visual cortex. The Journal of Neuroscience, 11(3), 641-649. doi:10.1523/jneurosci.11-03-00641.1991 Zhao, J., Pan, X., Sui, R., Munoz, S. R., Sperduto, R. D., & Ellwein, L. B. (2000). Refractive error study in children: results from Shunyi District, China. American Journal of Ophthalmology, 129(4), 427-435. doi:http://dx.doi.org/10.1016/S0002-9394(99)00452-3 Zhu, Q., Nelissen, K., Van den Stock, J., De Winter, F.-L., Pauwels, K., de Gelder, B., . . . Vandenbulcke, M. (2013). Dissimilar processing of emotional facial expressions in human and monkey temporal cortex. NeuroImage, 66, 402-411. doi:http://dx.doi.org/10.1016/j.neuroimage.2012.10.083  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0368844/manifest

Comment

Related Items