Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effects of attention on cortical auditory evoked potentials during a gap detection task Cozzi, Jennifer 2017

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_february_cozzi_jennifer.pdf [ 6.78MB ]
Metadata
JSON: 24-1.0360659.json
JSON-LD: 24-1.0360659-ld.json
RDF/XML (Pretty): 24-1.0360659-rdf.xml
RDF/JSON: 24-1.0360659-rdf.json
Turtle: 24-1.0360659-turtle.txt
N-Triples: 24-1.0360659-rdf-ntriples.txt
Original Record: 24-1.0360659-source.json
Full Text
24-1.0360659-fulltext.txt
Citation
24-1.0360659.ris

Full Text

EFFECTS OF ATTENTION ON CORTICAL AUDITORY EVOKED POTENTIALS DURING A GAP DETECTION TASK by  Jennifer Cozzi  B.Sc., Queen’s University, 2015  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Audiology and Speech Sciences)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  November 2017  © Jennifer Cozzi, 2017   ii Abstract  Gap-detection testing is a measure of temporal resolution, a component of auditory processing that is sometimes affected in Central Auditory Processing Disorder (CAPD). Using behavioural gap-detection tasks as part of a CAPD battery can be confounded by attention, in that it is difficult to distinguish whether poor performance is a result of attention deficits or central auditory processing deficits. Researchers have assessed the utility of cortical auditory evoked potentials (CAEPs) as an objective measure of gap-detection thresholds. The present study aimed to compare CAEPs during a passive and active gap-detection task to assess the effects of attention on electrophysiological gap-detection testing. The results showed no significant differences in gap-detection thresholds obtained behaviourally, obtained electrophysiologically during the passive task, and obtained behaviourally or electrophysiologically during the active task. N1-P2 amplitudes were found to be significantly larger in the active gap-detection task compared to the passive task. In conditions where a 0 ms duration or subthreshold duration gap was presented, an N2b waveform was evoked. The N2b is an endogenous event-related potential (ERP) that is usually evoked when a prepotent response must be withheld, such as during the no-go trials of a go/no-go task. Because the gap-detection task could be considered as a go/no-go paradigm, the N2b was likely evoked due to the unfulfilled expectation of a gap occurring with predictable timing and high probability. The goal of experiment 2 was to investigate the occurrence of this endogenously-evoked no-go N2b as previous literature has focused primarily on the N2b evoked by exogenous stimuli. The results found that all participants had an N2b wave to no-gap and subthreshold gap conditions. There were no significant correlations between N2b-P3a amplitudes and behavioural thresholds or between N2b-P3a amplitudes or N2b   iii latencies and reaction time or reaction time standard deviation. There was a significant correlation between P3a amplitudes and ex-Gaussian reaction time standard deviation. No significant differences were found between N2b amplitudes for correct rejections or misses. Overall, this study demonstrated that the no-go N2b can be evoked by an endogenous signal in the form of the omission of an expected gap in noise.         iv Lay Summary  In central auditory processing disorder, there is sometimes a deficit in the brain’s ability to process the timing of sounds. This can be assessed with tests of gap detection (detecting a silent gap within a sound). An issue is when patients cannot pay attention, making it difficult to know if the test results are accurate. Therefore, researchers have assessed gap detection using brain waves, called cortical auditory evoked potentials (CAEPs), that are produced to sound. This study investigated if CAEPs would be affected by the participants paying attention. Overall, specific brain waves were larger but gap detection results were unaffected. A brain wave called the N2b was also seen when there was no gap in the sound. The N2b likely occurred because the participants expected to hear a gap that did not happen, meaning the N2b was produced to an internal signal in the brain rather than an external sound.    v Preface  The data used in this thesis were collected by R. Angel and A. Herdman in the BRANE lab at the University of Columbia, Vancouver campus as part of a Master’s thesis. The data collection as well as the analysis of the passive-task cortical auditory evoked potentials performed by R. Angel and A. Herdman are described in Chapter 2, Section 2.1 to 2.2.1. The methods were approved by the Behavioural Research Ethics Board of the University of British Columbia. The ethics certificate number is H14-00441. All other analyses were performed by J. Cozzi and A. Herdman. This thesis is the original work of the author, J. Cozzi, and thesis supervisor, A. Herdman.      vi Table of Contents  Abstract .......................................................................................................................................... ii	Lay Summary ............................................................................................................................... iv	Preface .............................................................................................................................................v	Table of Contents ......................................................................................................................... vi	List of Tables ..................................................................................................................................x	List of Figures ............................................................................................................................... xi	List of Abbreviations .................................................................................................................. xii	Acknowledgements .................................................................................................................... xiii	Dedication ................................................................................................................................... xiv	Chapter 1: Introduction of Experiment 1 ....................................................................................1	1.1	 Temporal Processing ....................................................................................................... 1	1.2	 Central Auditory Processing Disorder ............................................................................ 1	1.2.1	 Gap Detection ......................................................................................................... 2	1.2.2	 Cortical Auditory Evoked Potentials ...................................................................... 3	1.3	 Goals of Study................................................................................................................. 5	1.3.1	 Purpose .................................................................................................................... 5	1.3.2	 Hypotheses .............................................................................................................. 6	Chapter 2: Methods of Experiment 1 ...........................................................................................8	2.1	 Data Acquisition ............................................................................................................. 8	2.1.1	 Participants .............................................................................................................. 8	2.1.2	 Procedure ................................................................................................................ 8	  vii 2.1.2.1	 Audiometric Assessment .................................................................................... 9	2.1.2.2	 Behavioural Gap-Detection Testing ................................................................... 9	2.1.2.3	 Gap-Detection Testing using Cortical Auditory Evoked Potentials ................. 10	2.2	 Analysis......................................................................................................................... 11	2.2.1	 Analysis of Passive Data ....................................................................................... 11	2.2.2	 Analysis of Active Data ........................................................................................ 12	2.2.2.1	 Behavioural Accuracy During Active CAEP Recording .................................. 12	2.2.2.2	 CAEP Rating ..................................................................................................... 13	2.2.2.3	 Inter-Rater Reliability ....................................................................................... 17	2.2.2.4	 Comparison of Gap-Detection Thresholds ....................................................... 17	2.2.2.5	 Comparison of N1-P2 Amplitudes .................................................................... 18	Chapter 3: Results of Experiment 1 ...........................................................................................21	3.1	 Behavioural Accuracy During Active CAEP Recording .............................................. 21	3.2	 ROC Curve.................................................................................................................... 22	3.3	 Inter-Rater Reliability ................................................................................................... 23	3.4	 Comparison of Gap-Detection Thresholds ................................................................... 23	3.5	 Comparison of N1-P2 Amplitudes ................................................................................ 24	3.6	 Topography of N1 vs. N2b Waveforms ........................................................................ 25	Chapter 4: Discussion of Experiment 1 .....................................................................................27	4.1	 Interpretation of Results ................................................................................................ 27	4.1.1	 Behavioural Accuracy ........................................................................................... 27	4.1.2	 Inter-Rater Reliability ........................................................................................... 28	4.1.3	 Gap-Detection Thresholds .................................................................................... 28	  viii 4.1.4	 Effect of Attention on N1-P2 Amplitudes ............................................................ 30	4.2	 Clinical Implications ..................................................................................................... 30	4.3	 Conclusions ................................................................................................................... 33	Chapter 5: Introduction of Experiment 2 ..................................................................................34	5.1	 N2 Background ............................................................................................................. 34	5.1.1	 N2b Component .................................................................................................... 35	5.1.1.1	 Go/No-Go and Stop Signal Paradigms ............................................................. 35	5.1.1.2	 Stimulus Modality ............................................................................................. 36	5.1.1.3	 Inhibition Theory vs. Conflict Monitoring Theory ........................................... 37	5.2	 Exogenous vs. Endogenous Stimuli .............................................................................. 40	5.3	 Purpose .......................................................................................................................... 42	5.4	 Hypotheses .................................................................................................................... 43	Chapter 6: Methods of Experiment 2 .........................................................................................44	6.1	 Data Acquisition ........................................................................................................... 44	6.1.1	 Paradigm ............................................................................................................... 44	6.2	 Data Analysis ................................................................................................................ 45	6.2.1	 CAEP Rating ......................................................................................................... 45	6.2.2	 Inter-Rater Reliability ........................................................................................... 46	6.2.3	 Correlation Between N2b-P3a Amplitudes and Behavioural Thresholds ............ 46	6.2.4	 Correlation Between N2b-P3a Amplitudes and Latencies and Reaction Times .. 48	6.2.5	 Comparison of N2b-P3a Amplitudes Between Correct Rejections and Misses ... 49	Chapter 7: Results of Experiment 2 ...........................................................................................50	7.1	 Inter-Rater Reliability ................................................................................................... 50	  ix 7.2	 Presence of N2b Across Participants ............................................................................ 51	7.3	 Correlation Between N2b-P3a Amplitudes and Behavioural Thresholds .................... 53	7.4	 Correlation Between N2b-P3a Amplitudes and Latencies and Reaction Times .......... 53	7.5	 Comparison of N2b-P3a Amplitudes Between Correct Rejections and Misses ........... 55	Chapter 8: Discussion of Experiment 2 .....................................................................................56	8.1	 Interpretation of Results ................................................................................................ 56	8.1.1	 Inter-Rater Reliability ........................................................................................... 56	8.1.2	 N2b-P3a Amplitudes and Behavioural Gap-Detection Thresholds ...................... 57	8.1.3	 N2b-P3a Amplitudes and Latencies and Reaction Time ...................................... 58	8.1.4	 N2b-P3a Amplitudes for 0 ms Gaps Compared to 2 ms Gaps ............................. 59	8.2	 Conclusions on the N2b ERP ........................................................................................ 59	8.2.1	 Presence of N2b Across Participants .................................................................... 59	8.2.2	 Conflict Monitoring vs. Response Inhibition Theories ......................................... 60	8.2.3	 N2b to Endogenous Signals .................................................................................. 62	8.3	 Conclusions ................................................................................................................... 64	References .....................................................................................................................................65	   x List of Tables  Table 3.1: ANOVA Among Gap Detection Thresholds. .............................................................. 24	Table 3.2: N1-P2 Amplitudes in the Passive and Active Tasks ................................................... 24	Table 7.1: Correlations Between N2b-P3a Amplitude and Behavioural Thresholds Using Two Procedures ..................................................................................................................................... 53	Table 7.2: Results of Multiple Linear Regression Analyses ......................................................... 54	   xi List of Figures  Figure 2.1: N1 Rating for Hits and Correct Rejections ................................................................ 15	Figure 2.2: N1 Rating for Misses and False Alarms ..................................................................... 16	Figure 3.1: Behavioural Gap Detection Accuracy in 35 Participants During CAEP Recording .. 21	Figure 3.2: ROC Curve for Behavioural Gap Detection During CAEP Recording ..................... 22	Figure 3.3: Consensus Among Raters for the Presence of N1 Waves .......................................... 23	Figure 3.4: N1-P2 Amplitudes in the Passive Compared to Active Task ..................................... 25	Figure 3.5: Topographies .............................................................................................................. 26	Figure 7.1: Consensus Among Raters for the Presence of the N2b and P3a Waves and CNEs ... 51	Figure 7.2: N2b Presence Across Participants .............................................................................. 52	Figure 7.3: Correlation between P3a Latency and exGaussian Reaction Time Standard Deviation....................................................................................................................................................... 54	Figure 7.4: Comparison of N2b-P3a Amplitudes for 0 ms and 2 ms Gap Durations ................... 55	   xii List of Abbreviations  ACC: Anterior Cingulate Cortex ADHD: Attention Deficit Hyperactivity Disorder ANOVA: Analysis of Variance CAEPs: Cortical Auditory Evoked Potentials CAPD: Central Auditory Processing Disorder CNE: Could Not Evaluate dB HL: deciBel Hearing Level dB SPL: deciBel Sound Pressure Level dlPFC: Dorsolateral Prefrontal Cortex EEG: Electroencephalography ERN/Ne: Error Related Negativity  ERPs: Event-Related Potentials GIN: Gaps-in-Noise MMN: Mismatch Negativity ROC: Receiver Operating Characteristic SSRT: Stop Signal Reaction Time TBI: Traumatic Brain Injury TMTF: Temporal Modulation Transfer Function UBC: University of British Columbia vlPFC: Ventrolateral Prefrontal Cortex    xiii Acknowledgements  First, I would like to extend my deepest gratitude to my thesis supervisor, Tony Herdman. His expertise and contagious enthusiasm were integral to this process. Thank you for being so generous with your time.  I would also like to thank my thesis committee, Sasha Brown and Navid Shahnaz. Your guidance and input were very much appreciated.  Next, I would like to extend my thanks to Rebecca Angel, for kindly allowing me to expand upon her research.  A special thank you to my friend and lab mate, Kelsey Meagher, for her constant encouragement and reassurance during this process. Finally, thank you to my friend Colleen Jackson, for her endless encouragement and support.       xiv Dedication  To my family, for their generous support and unwavering confidence.  1 Chapter 1: Introduction of Experiment 1  1.1 Temporal Processing Speech perception, and consequently the ability to communicate with others, is dependent on the central auditory system’s ability to process various auditory cues, including temporal cues. One aspect of temporal processing is temporal resolution or discrimination, which is an important cue that helps the central auditory system discriminate between stop consonants that have the same place of articulation. This discrimination is based on a feature known as voice onset time, the time delay that occurs between the release burst of the stop consonant and the onset of phonation or vibration of the vocal cords. In normal listeners, voice onset times of less than 20 ms result in the perception of a voiced consonant such as in /ba/, whereas voice onset times greater than 20 ms result in the perception of a voiceless consonant such as in /pa/ (Abramson & Lisker, 1970). Thus, a deficit in temporal resolution could lead to difficulty with speech understanding.   1.2 Central Auditory Processing Disorder Central auditory processing disorder (CAPD) was defined by the American Speech-Language-Hearing Association (ASHA) in 2005 as a difficulty in the perceptual processing of auditory information by the central nervous system, characterized by poor performance in sound localization or lateralization, auditory discrimination, auditory pattern recognition, temporal aspects of audition including temporal integration, temporal discrimination (gap detection), temporal ordering, and temporal masking, and/or auditory performance with competing or degraded acoustic signals. CAPD affects approximately 2 to 3% of children and 10 to 20% of adults, with the prevalence potentially increasing to 70% in adults over the age of 60 (Chermak   2 & Musiek, 1997; Cooper & Gates, 1991; Stach, Spretnjak, & Jerger, 1990). CAPD can result from dysfunction anywhere along the auditory processing pathway, but can also occur with global deficits such as traumatic brain injuries (TBIs). In fact, in a study from 2005, 58% of the participants with a previous TBI were found to have CAPD (Bergemalm & Lyxell, 2005). There is also evidence that CAPD may result from auditory deprivation caused by chronic otitis media in childhood (Moore, Hartley, & Hogan, 2003). CAPD manifests in difficulties with following conversations or multi-step directions, remembering spoken information, listening in noise or with competing speech, and reading or spelling. However, there are other clinical disorders with similar symptoms, such as attention deficit hyperactivity disorder (ADHD), specific language impairment, learning disabilities, and autism spectrum disorder, which can make diagnosing CAPD problematic (Jerger & Musiek, 2000).   1.2.1 Gap Detection CAPD is currently diagnosed using a behavioural test battery including a test of gap detection, which is a measure of auditory temporal resolution. Other measures of auditory temporal resolution do exist, such as the Temporal Modulation Transfer Function (TMTF), which assesses a listener’s threshold for detecting amplitude modulation of a sound based on the rate of modulation (Viemeister, 1979). However, gap-detection testing is the more common measure due to the availability of more efficient test procedures (Shen & Richards, 2013). Gap-detection tasks involve the participant listening for gaps of silence within a broadband noise or a pure-tone stimulus of a particular frequency. With broadband noise stimuli at supra-threshold intensity levels, normal listeners can typically perceive gaps of 2 to 3 ms (Plomp, 1964). Available clinical tests of gap detection include the Random Gap Detection Test (Keith, 2000) and the Gaps-In-  3 Noise test (Musiek et al., 2005). Gap-detection ability matures by 6 years of age, and thresholds between 2 and 20 ms are typically considered normal (Keith, 2000). Interestingly, this normal threshold cutoff of 20 ms is very near to the phonetic boundary of approximately 30 ms, where stop consonants are perceived as either voiced or voiceless in a categorical manner depending on which side of the phonetic boundary the voice onset time falls (Phillips, 1999). Research has also shown that if a child cannot perceive a gap of 20 ms, they are likely to have difficulty with speech perception (Cestnick & Jerger, 2000). Although gap detection represents only one of the tests in the CAPD battery, there are still challenges associated with it. Some confounding factors of behavioural gap-detection tasks are motivation, attention, cooperation, and understanding the instructions. These are issues that could certainly be seen in the population undergoing CAPD testing, which includes those with attention deficits and those with TBIs who may be in the process of obtaining medico-legal compensation.   1.2.2 Cortical Auditory Evoked Potentials In an effort to avoid some of the confounds of cognition associated with behavioural tests of CAPD, researchers have looked into the utility of using objective measures, specifically electrophysiological responses such as cortical auditory evoked potentials (CAEPs) (Angel, 2016; Michalewski et al., 2005; Palmer & Musiek, 2014). CAEPs are neural responses recorded from the scalp that reflect a change in the frequency, intensity, or timing of a sound. The P1-N1-P2 waveform, or acoustic change complex, has been used to find audiometric thresholds in those who cannot reliably respond behaviourally, as it has been found to correlate well with behavioural audiometric thresholds (Stapells, 2002). Recent research has shown that in normal listeners, passive CAEPs can also be used to estimate gap-detection thresholds (Angel, 2016).   4 This suggests the possibility that CAEPs could be used as an objective measure in populations who are undergoing CAPD evaluation but cannot reliably respond in behavioural gap-detection tests, such as medico-legal cases or those with cognitive deficits. The normative data collected found that the electrophysiological gap-detection thresholds were within 1 ms of behavioural thresholds, and furthermore, that 85% of participants had responses to gaps of 10 ms (Angel, 2016). Therefore, electrophysiology is a promising alternative method to help in differentiating between cognitive and perceptual deficits that can affect auditory temporal processing.   While CAEPs have been shown to estimate behavioral gap-detection thresholds in normal listeners in passive tasks, there is currently little evidence on the effect attention would have on electrophysiological gap-detection thresholds. Attention is known to have an effect on the amplitude of the N1-P2 waveform of CAEPs, and furthermore, a P3 response is typically seen when a person has consciously attended to a stimulus (Picton & Hillyard, 1974). This P3 response can be divided into a P3a, which results from frontal attention mechanisms, and a P3b, which results from temporal-parietal activity associated with memory processing (Polich, 2007). Because of the potential confound of attention on CAPD assessments, obtaining more information on the effects of attention on CAEP gap-detection thresholds could be useful in making differential diagnoses on the source of temporal processing difficulties. CAEP gap-detection thresholds are highly variable, and in addition, only 90% of normal listeners have electrophysiological responses to 12 ms gaps, significantly lower than the number who have behavioural responses, which raises the question of whether attention could be a factor (Angel, 2016).     5 1.3 Goals of Study Inattentiveness and distractibility are common symptoms of ADHD and CAPD (Chermak, Somers, & Seikel, 1998). There is much debate over whether ADHD and CAPD are clinically distinct, comorbid, or reflect a single disorder. Chermak, Hall, and Musiek (1999) postulated that CAPD and ADHD are separate clinical entities, where attention deficits may lead to difficulties with auditory processing in a top-down manner, and impaired auditory processing could result in poor attention in a bottom-up manner. Teasing apart deficits in attention (whether from ADHD, TBI, non-compliance, or other origins) and deficits in auditory processing has proven to be challenging, but if possible it may provide more confidence when diagnosing or ruling out CAPD. Thus, the goal of the present research was to compare CAEPs during passive and active gap-detection tasks in normal listeners. This normative data may provide information on the effect attention has on CAEPs, which could possibly be utilized in the future to eliminate the confound of attention on gap-detection testing as part of a CAPD test battery. This may aid in differentiating between CAPD and cognitive dysfunctions that could manifest with similar behaviours.  1.3.1 Purpose The purpose of this research was to assess the effects of attention on CAEPs during a gap-detection task in normal listeners. This included comparing gap-detection thresholds and N1-P2 amplitudes between passive and active gap-detection tasks. In addition, this research looked into the possibility of “misses” in the active task, which are proposed present cortical responses (N1-P2) without a corresponding conscious awareness and button press by the participant (absent P3a-P3b). These “misses” could indicate that attention was a confound for a particular   6 participant. This could provide information that may help differentiate between someone who performed poorly on a gap-detection test due to attention and someone who performed poorly due to temporal processing deficits, which could be used in the future to provide clinicians with more evidence and reassurance when diagnosing or ruling out primary temporal processing deficits as part of a CAPD test battery. Information that aids clinicians in the differential diagnosis of CAPD and attention deficits or cognitive dysfunctions that manifest in similar ways to CAPD could lead to more appropriate follow-up and treatment for clients.  1.3.2 Hypotheses I hypothesized that the amplitudes of the N1-P2 responses would be larger in the active compared to the passive task because N1-P2 amplitudes are known to be affected by attention (Picton & Hillyard, 1974). I also hypothesized that overall, gap-detection thresholds would not be affected by attention because the participants in this study represented a normal population. Furthermore, I hypothesized the occasional presence of a “miss” in the active task, or a present N1-P2 response without the associated button press and P3a-P3b waveform. This may represent a situation where the participant’s attention wandered but the central auditory nervous system still detected a response, similar to what would be expected in the passive task. This may provide some guidance on what would be seen if a person was performing poorly on temporal processing tasks due to deficits in attention rather than deficits in auditory processing. While the present study collected normative data on the effect of attention on CAEPs during gap detection, future research could look at the patterns that may be seen between the passive and active tasks in populations with CAPD compared to attention deficits. For instance, in a participant with primarily auditory processing difficulties, the expected results would be poor performance (i.e.   7 high gap-detection thresholds) in both the passive and active tasks. However, it is possible that if normal gap-detection thresholds are found in the passive task that are incongruent with poor behavioural results (i.e. many “misses” or not pressing the button to the same gap durations in the active task), this may represent someone whose primary issue lies in attention. If only behavioural testing was performed with this hypothetical participant, it would be difficult to determine whether poor performance was due to auditory processing or attention deficits. If only passive CAEP testing was performed on this hypothetical participant, it may be difficult to determine whether the participant’s normal performance was representative of their abilities or difficulties in the real world. The additional information of the active CAEP condition may be a useful sign in differentiating between different underlying causes of temporal processing deficits that present with similar behavioural manifestations. In other words, it may help to eliminate the confounding factor of attention in gap-detection testing as part of the CAPD test battery.    8 Chapter 2: Methods of Experiment 1  2.1 Data Acquisition The data used in this research were collected as part of a previous study (Angel, 2016). The methods were reviewed by the Behavioural Research Ethics Board at the University of British Columbia.  2.1.1 Participants The study consisted of 47 right-handed participants (31 female, 16 male) aged 18 to 40, with normal hearing and no history of perceptual or cognitive problems, learning or communication disorders, head injury, otitis media, or use of ototoxic medications. Participants were informed of the study procedures and then provided their consent. Of the 47 participants, 37 had complete datasets including both passive- and active-task EEG data. Two participant’s datasets were rejected from further analyses; one was judged by raters to be too noisy, and another was found to be an outlier based on active-task behavioural results. Thus, the remaining 35 participants’ datasets were used for further analyses in the present study.  2.1.2 Procedure The procedures performed included audiometric assessment, behavioural gap-detection testing, and recording of CAEPs during gap-detection testing. These procedures are discussed in greater detail below and in Angel (2016).     9 2.1.2.1 Audiometric Assessment Otoscopy was performed on each participant, followed by tympanometry with a conventional 226 Hz probe-tone to confirm normal middle-ear function. Next, pure-tone hearing screening was performed via air-conduction using ER-3A insert earphones in a sound booth in the BRANE lab at UBC. Hearing was considered to be normal if behavioural responses were obtained at or below an intensity of 20 dB HL from 500 Hz to 4000 Hz bilaterally.   2.1.2.2 Behavioural Gap-Detection Testing Behavioural gap-detection testing was performed on each participant in the sound booth. The stimuli used were broadband noise bursts 1 second in duration with gaps of 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, and 20 ms that were presented randomly at 500 ms post-noise onset. The second halves of the noise bursts were reduced by the duration of the gap to ensure that overall stimulus duration remained constant. Stimuli were generated in a custom Matlab program and were calibrated using a SoundPro sound level meter. The stimuli were then sent from the computer to an Interacoustics audiometer and presented to the participant’s left ear through an ER-3A insert earphone at 60 dB SPL while the right ear was occluded with a foam earplug to prevent crossover of sound. Participants were instructed to press a button when they perceived a gap in the noise. A modified Hughson-Westlake or bracketing procedure was used, such that if the participant correctly responded to two consecutive gaps of the same duration, the gap duration of the subsequent noise burst was decreased by 1 ms, and if the participant did not respond to a gap on a single trial (or incorrectly identified a gap), the gap duration of the subsequent noise burst was increased by 1 ms. The behavioural gap-detection threshold was considered to be the shortest gap duration the participant detected 4 out of 6 presentations.    10  2.1.2.3 Gap-Detection Testing using Cortical Auditory Evoked Potentials Electroencephalography (EEG) was used to record cortical auditory evoked potentials from each participant in a sound booth. An ActiView2 64-channel system from BioSemi was used to collect the EEG data. Electrode caps were used to place the 64 scalp electrodes, and additional electrodes were placed bilaterally on the mastoid and around the eyes at the outer canthi, infra-orbital margins, and supra-orbital margins to detect eye blinks. EEG was sampled online at a rate of 1024 Hz and band-pass filtered between 0.16 to 208 Hz.   Stimuli used were 1-second noise bursts with gaps of 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, or 20 ms in duration. Gaps always occurred at 500 ms following the onset of the noise burst, however the durations of the gaps presented were randomly ordered. In addition, the duration of the noise burst after the gap was reduced by the duration of the gap in order to keep the overall stimulus duration to 1 second. This ensured that participants were not using stimulus duration information to make their judgements regarding presence or absence of a gap. Stimulus onset asynchrony was randomly varied between 1850 and 2150 ms. Overall, 100 trials were collected for each gap duration for each participant. As with the behavioural test, the stimuli were calibrated using a SoundPro sound-level meter with a 2 cc coupler and presented to each participant’s left ear at 60 dB SPL. The right ear was occluded with a foam earplug.   In the passive task, participants were awake but instructed to ignore the stimuli while they watched a closed-captioned movie. In the active task, participants were instructed to pay attention to the stimuli and to press a button when they perceived a gap in the noise. This active   11 task was very similar to the response procedure of the Gaps-in-Noise (GIN) test (Musiek et al., 2005). Unlike the behavioural gap-detection testing described above (see Section 2.1.2.2), the order of gap durations was random and did not depend on the participants’ responses.  2.2 Analysis  2.2.1 Analysis of Passive Data The EEG was bandpass filtered from 1 to 20 Hz using an impulse-response Butterworth filter. Event-related potentials (ERPs) seen from -500 ms to 1500 ms were time-locked to the onset of the first noise burst, therefore the gap-evoked CAEP was time-locked to the onset of the gap at 500 ms. Any ERP trials which exceeded ±100 µV between -100 and 1000 ms were rejected. The non-rejected trials were split into two buffers for each gap duration. Each buffer was then averaged to obtain two ERP replications for each gap duration. Difference ERP waveforms were also calculated for each gap duration by subtracting the 0 ms gap condition from all other gap conditions (2 to 20 ms). These ERPs for all participants were plotted for a Cz electrode recording and placed in an ERP bank.   Three raters were required to judge whether a gap-evoked CAEP (i.e. an N1-P2 waveform) was present between 550 ms and 800 ms post-noise burst onset. They viewed a computer screen which showed two replications as well as the average of the ERPs for a given gap duration (0 to 20 ms). These waveforms were overlaid on a baseline of two replications as well as the average of the ERPs for the 0 ms gap condition. Below these ERPs on the screen, two replications as well as the average waveform for the difference ERPs (gap minus no-gap conditions) were shown.   12 The raters were blinded to the gap duration that was being presented to avoid bias. In addition, simulated 0 ms gap waveforms were added to the ERP bank for 47 sham participants to increase the probability of seeing a no-gap compared to a gap-evoked waveform to 50:50. This was done to reduce rater bias. The raters were also able to reject the ERP if the residual noise was judged to be too high to confidently decide whether a gap-evoked CAEP was present or absent. Overall, eleven participants were rejected, leaving a total of 36 participants who were included in further passive analyses. Gap-evoked CAEP thresholds were then determined, including both the individual rater-determined threshold (the lowest gap duration judged as having a present N1-P2 response by 1 out of 3 raters) and the group rater-determined threshold (the lowest gap duration judged as having a present N1-P2 response by 2 out of 3 raters).  2.2.2 Analysis of Active Data  2.2.2.1 Behavioural Accuracy During Active CAEP Recording A plot showing the percent of hits for different gap durations across participants was created using Matlab in order to determine the accuracy of participants in behaviourally detecting gaps in noise by pressing a button during recording of CAEPs. A plateau in percent of hits across 36 participants (excluding one participant that did have an active dataset but was rejected during the passive analysis due to high residual noise) was found by setting a criterion that the slope for every individual must be ≤5% change in the percent of hits over a 2 ms duration (i.e. slope must be ≤0.025/ms) to be considered as a plateau. This resulted in a plateau in percent of hits found starting at 12 ms gap duration. Next, the mean percent hits of all participants from 12 to 20 ms gap durations was found and the standard deviation was calculated. A one-tailed t-test was used   13 to determine if there were any participants with a percent of hits ±2.5 standard deviations from the mean (i.e. lower than 95% confidence limits) during the plateau period that would therefore be considered to be outliers. One participant was eliminated in this way. In addition, a Receiver Operating Characteristic (ROC) curve was created using Matlab for the 35 participants’ behavioural hits across all gap durations and false alarms for 0 ms gap trials during active CAEP recording.   2.2.2.2 CAEP Rating I initially intended for the active-task ERPs to be rated for the presence or absence of gap-evoked CAEPs using the same procedure as described for the passive-task ERP rating (i.e. raters blinded to the gap duration information with the 0 ms gap condition used as a baseline). However, in the preliminary rating of the data, an unexpected N2 waveform was seen for the 0 ms gap conditions within the active task. Because the N2 waveform is similar in morphology and occurs in the same approximate time window as the N1 (approximately 100 to 150 ms post-stimulus onset), it was difficult for raters to determine if the large negative-to-positive deflection seen between 500 and 800 ms was actually an N1 to a gap or an N2 to a no-gap condition. Thus, the no-gap condition in the active task could not be used as a baseline for raters to compare ERPs to gaps. Therefore, a different method of waveform rating was adopted for the active task in the present study, with raters not blinded to gap duration information. For more information regarding the N2 waveform, see Chapter 5.  Two raters judged the active-task ERPs for the presence or absence of N1 waveforms, as well as the N2b and P3a waveforms that were seen unexpectedly in preliminary rating of data. ERP trials   14 from -500 to 1500 ms, which did not exceed ±100 µV, were split into two buffers for each gap duration. The buffers were then averaged to produce two ERP replications for each gap condition. The two ERP replications for each participant were displayed on a computer screen, with the 11 gap durations (0 to 20 ms) shown in descending order from the top to the bottom of the screen in a plot of ERP amplitude across time (see Figure 2.1). Waveforms for adjacent gap durations were shown in alternating colours of red and blue to minimize the raters’ confusion if waveforms of different durations overlapped. The two ERP replications for each gap duration in the active task were overlaid over two replications of the no-gap condition in the passive task (with false alarms excluded). The no-gap condition from the passive task was used as the baseline because the active task no-gap condition contained N2 waves that were similar in morphology and timing compared to the N1 waves. The ERPs were plotted for a Pz, Cz, FCz, and Fz electrode recording from left to right across the screen in order to provide some spatial information of the distribution of the N1 and N2 responses. This helped raters better identify the N2b and P3a waves that have, respectively, more frontal and parietal distributions. ERPs for each participant were displayed on two screens, one for correct responses (i.e. hits and correct rejections), and one for incorrect responses (i.e. misses and false alarms), with the aim of aiding raters in differentiation of the N1 and N2b (see Figure 2.1 and Figure 2.2). Gap durations with fewer than 2 trials for either correct responses or incorrect responses were not displayed. Otherwise, ERPs to gap durations for each participant were displayed on the screen in a vertically staggered manner, allowing the raters to use a more clinical method of judging the presence or absence of a response. This decreased the possibility of raters incorrectly judging an N1 to be present for a 0 ms gap duration waveform when in fact they were viewing an N2b.     15 Overall, raters judged waveforms for 35 participants. Raters were required to make a judgment on the presence or absence of the N1 for each gap duration for each participant, in both the Hits and Correct Rejections and the Misses and False Alarms screens. The on-screen toggle button to mark the N1 as present or absent was located to the right of each ERP waveform plotted for the Cz electrode (see Figure 2.1). Raters also had the option of marking any gap duration as Could Not Evaluate (CNE) if residual noise was too high to make a confident visual judgment. The CNE toggle button was located on the far-right side of the screen.  Figure 2.1: N1 Rating for Hits and Correct Rejections Example of the screen that was displayed to raters for judgment of CAEPs with correct responses (hits for trials with gaps of varying durations and correct rejections of trials with no gaps) for one participant. Waveforms were displayed from the top to bottom of the screen in descending order of gap duration, and different electrode recordings were displayed in columns from left to right for electrode position going from the back to the front of the scalp. The toggle buttons for raters to indicate the presence or absence of N1, N2b, and P3a responses for each gap   16 condition were located beside each waveform. The toggle button for raters to indicate they could not evaluate the waveform (CNE) is located on the far right of the screen. For the hits and correct rejections for this participant, N1 responses were observed at around 650 ms (filled arrow heads) for gaps between 6 to 20 ms and a large N2b-P3a was also observed starting at about 650 ms (open arrow heads) for the 0 ms gap condition.    Figure 2.2: N1 Rating for Misses and False Alarms Example of the screen that was displayed to raters for judgment of CAEPs with incorrect responses (misses for trials with gaps of varying durations and false alarms to trials with no gaps) for one participant. Waveforms were displayed from the top to bottom of the screen in descending order of gap duration, and different electrode recordings were displayed in columns from left to right for electrode positions going from the back to the front of the scalp. Missing waveforms for certain gap durations indicated that there were insufficient (n<2) trials for displaying two replications. The toggle buttons for raters to indicate the presence or absence of N1, N2b, and P3a responses for each gap condition were located beside each waveform. The toggle button for raters to indicate they could not evaluate the waveform (CNE) is located on the far right of the screen.  For the misses and false alarms for   17 this participant, N2b-P3a responses were observed starting at about 650 ms (open arrow heads) for the misses at 2 and 4 ms gap conditions.  2.2.2.3 Inter-Rater Reliability Inter-rater reliability between the two raters was determined for judging the presence or absence of the N1 wave in the active task. Percent agreement was first calculated as an index of inter-rater reliability by dividing the number of agreed ratings between the two raters by the overall number of ratings and then converting to percentage. However, this method does not take into account agreement that may occur between two raters due to chance, which is very likely when there are only two categories of responding (i.e. present or absent), meaning that percent agreement could be high but actual agreement may be low (Fleiss & Cohen, 1973). Therefore, Cohen’s kappa (k) was also calculated as a measure of inter-rater reliability on N1 waveform rating. Cohen’s kappa is used when there are two raters and data can be considered nominal, thus it is an appropriate measure for the present study (Cohen, 1960). It is calculated by subtracting the proportion of agreement expected by chance (𝑝𝑒) from the proportion of actual agreement between the two raters (𝑝𝑜), and then dividing the resulting number by the proportion of agreement expected by chance subtracted from 1, as in the following equation: 𝑘 = &'(&)*(&)        (equation 1)  2.2.2.4 Comparison of Gap-Detection Thresholds Average gap-detection thresholds were obtained and compared across the following four conditions for 33 participants (note: two additional participants were considered by raters to have high residual noise in their recordings and thus were rejected from further analyses). During the   18 modified Hughson-Westlake procedure, the average behavioural gap detection threshold was considered to be the average of the lowest gap durations each participant responded to four out of six presentations. During the passive task, the average electrophysiological gap detection threshold was considered to be the average of all three raters’ lowest gap duration where the N1 was judged to be present for each participant. During the active task, the average electrophysiological gap detection threshold was considered to be the average of both raters’ lowest gap duration where the N1 was judged to be present for each participant. During the active task, the average behavioural gap detection threshold obtained through the modified GIN procedure was considered to be the average of the lowest gap duration each participant responded to 50% of the time. This was achieved by using an exponential fit for each participant’s behavioural data (percent hits by gap duration), then finding the gap duration where the percent of hits was equal to 50. Gap-detection thresholds across these four conditions (modified Hughson-Westlake procedure, passive-task CAEPs, active-task CAEPs, and modified GIN procedure) were then compared using a four-way Analysis of Variance (ANOVA).   2.2.2.5 Comparison of N1-P2 Amplitudes N1-P2 amplitudes were found and compared between the passive and active tasks for 33 participants (excluding participants who were rejected due to high residual noise during waveform rating). Passive and active N1-P2 amplitudes for all gap durations (excluding 0 ms) were determined by an experienced observer picking the N1 and P2 peaks in the ERPs. A plot of amplitude by time was displayed on a computer screen with the mean ERP Cz waveform for each gap duration (in black) along with two split buffer replications (in pink and green) for each participant overlaid over all other gap durations (in grey) for the same participant. The time   19 window of 500 ms to 1000 ms post-stimulus onset was marked by vertical red dotted lines on the plot to designate the interval in which N1 and P2 responses could be selected. An automatic peak-picking algorithm initially found possible N1 and P2 peaks, which were displayed as coloured asterisks (i.e. peak markers) on ERP waveforms. Scalp topographies at the time points for the N1 and P2 markers were displayed on the screen below the amplitude-time plots. Topography plots were colour-coded, where blue represents negative and red represents positive voltages. The ERP waveforms and topographies were used by the observer to confirm the correct identification of the N1 and P2 peaks. The criteria set to identify the N1 was to choose the earliest peak in the time window that replicated with the N1 waveform morphology of the other gap durations for the same participant. In addition, N1 peak identification was always checked by comparing the gap-evoked N1 topography at about 600 ms to the noise-burst onset N1 topography at about 100 ms. The criteria set to identify the P2 peak included choosing the first positive peak following the N1 that replicated with the waveform morphology of the other gap durations for the same participant. Again, P2 peak identification was always checked by comparing the gap-evoked P2 topography to the noise-burst onset P2 topography. In the case of a double-peaked P2, the first peak was chosen if the topography better matched the onset P2, otherwise the average latency between the two peaks was chosen if the topographies of the double peaks were similar. If no clear response was present or visual replication was poor for a specific gap duration for a particular participant, the waveform was not marked and the N1 and P2 responses were considered to be absent.  Subsequently, N1-P2 amplitudes between the passive and active tasks were compared using a paired samples t-test. For each participant, if there were any gap durations in either the passive or   20 active task where no N1-P2 amplitude had been determined, this particular gap duration for the specific participant was excluded from the sample. Overall, 70 data points were excluded across participants and gap conditions. A paired samples t-test was used in order to take into account any correlation between the same participant on the passive and active task.   21 Chapter 3: Results of Experiment 1  3.1 Behavioural Accuracy During Active CAEP Recording The average accuracy of 35 participants was found and shows that a plateau of 100% hits was reached starting at a gap duration of 12 ms, as seen in Figure 3.1. This indicates that it was justified to use this data for further analyses as participants could complete the task accurately.  Figure 3.1: Behavioural Gap Detection Accuracy in 35 Participants During CAEP Recording Percent behavioural hits versus gap duration across 35 participants. Individual participants are shown with grey lines and the average accuracy across participants in shown with a black line.   22 3.2 ROC Curve A d’ of 2.4685 was found, indicating that participants were able to distinguish between a gap being present or absent and performed well above chance level. A high sensitivity or hit rate was seen for all participants with a more variable specificity or false alarm rate, as shown in Figure 3.2. This suggests that it was justified to use data from this task in further analyses because the participants’ ability to perform the task was not a confound.   Figure 3.2: ROC Curve for Behavioural Gap Detection During CAEP Recording Hit rate versus false alarm rate for 35 participants.   23 3.3 Inter-Rater Reliability The percent agreement between the two raters for judgement of the N1 waves as present or absent was 97%. Cohen’s kappa for the two raters’ judgements of the N1 waves was 0.93. This indicates near-perfect inter-rater reliability (McHugh, 2012). Figure 3.3 shows the consensus between the two raters.  Figure 3.3: Consensus Among Raters for the Presence of N1 Waves Percent of participants rated as having a present N1 wave across gap durations for both correct (left) and incorrect (right) responses. Rater 1’s judgements are shown in blue, Rater 2’s judgements are shown in green, and consensus between the two raters is shown in red.   3.4 Comparison of Gap-Detection Thresholds A 4-way ANOVA for 33 participants revealed no significant differences in gap-detection thresholds obtained through the modified Hughson-Westlake procedure, N1 waves seen during the passive task, N1 waves seen during the active task, and the modified GIN procedure during the active task (see Table 3.1). Therefore, it can be concluded that for normal hearing   24 participants, behavioural and electrophysiological gap-detection thresholds obtained through various methods are not significantly different.  Source Sum of Squares Degrees of Freedom Mean Squares F  p-value Groups 8.325 3 2.77472 1.51 0.2156 Error 235.544 128 1.84018   Total 243.868 131     Table 3.1: ANOVA Among Gap Detection Thresholds. 4-way analysis of variance between gap detection threshold procedures for 33 participants. The F statistic was smaller than the critical value with a p-value of 0.2156, indicating no significant differences between groups.  3.5 Comparison of N1-P2 Amplitudes A comparison of N1-P2 amplitudes between the active and passive task was completed for 33 participants using a paired-samples t-test. The results showed that the N1-P2 amplitudes, shown in Table 3.2, in the active task were significantly larger (p = 2.43 x 10(-.) than the amplitudes during the passive task (see Figure 3.4).    Passive Active Mean 3.0842 uV 5.5711 uV Standard Deviation 1.6824 2.2883  Table 3.2: N1-P2 Amplitudes in the Passive and Active Tasks N1-P2 amplitudes were significantly larger in the active compared to the passive task (p<0.05).   25  Figure 3.4: N1-P2 Amplitudes in the Passive Compared to Active Task N1-P2 amplitudes were significantly larger in the active compared to the passive task for all gap durations.  3.6 Topography of N1 vs. N2b Waveforms Average-referenced (64 scalp and two mastoid electrodes with electrooculography electrodes removed) ERPs were plotted showing the group mean for the 20 ms gap duration trials in the passive task, the 20 ms gap duration trials (hits only) in the active task, and the 0 ms gap duration trials (correct rejections only) in the active task (see Figure 3.5). The topographies of the N1 and P2 (for the 20 ms gap duration in the passive and active task) and the N2b and P3a (for the 0 ms gap duration in the active task) were plotted at the latencies that corresponded to their peak mean amplitudes. The topography plots were colour-coded (where red symbolizes positivity and blue symbolizes negativity), and electrode locations were labeled. The N1 and P2 have both different latencies and different topographies compared to the N2b and P3a (see Figure 3.5). This helps to increase the confidence that the waveforms seen in the 0 ms gap duration trials in the active task were in fact N2b waves and not N1 waves.   26                                                                   Figure 3.5: Topographies Group-mean ERP waveforms and mean topographies for N1, P2, N2b, and P3a responses. The black dotted line denotes the onset of the noise burst (0 ms) and the green dotted lines show the onset and offset of the gap, if applicable (500 ms; 520 ms). Top left: Group-mean ERP for the 20 ms gap duration conditions in the passive task. The red lines show the latencies of the peak N1 (614 ms) and P2 (680 ms) waveforms. The average N1 and P2 topographies are shown below. Top right: Group-mean ERP for the 20 ms gap duration conditions in the active task. The red lines show the latencies of the peak N1 (614 ms) and P2 (680 ms) waveforms. The average N1 and P2 topographies are shown below. Bottom right: Group-mean ERP for the 0 ms gap duration conditions in the active task. The red lines show the latencies of the peak N2b (635 ms) and P3a (794 ms) waveforms. The average N2b and P3a topographies are shown below.   27 Chapter 4: Discussion of Experiment 1  4.1 Interpretation of Results An important part of the CAPD test battery is gap-detection testing, which assesses temporal resolution. An issue with behavioural gap-detection testing is the confound of attention, in that it is difficult to distinguish between poor results due to central auditory processing deficits or due to attention deficits. Researchers have investigated the utility of using CAEPs to assess gap-detection ability in a more objective manner (Angel, 2016; Michalewski et al., 2005; Palmer & Musiek, 2014). The purpose of the present study was to compare CAEPs during a passive compared to an active gap-detection task in normal listeners in order to collect normative data on the effects of attention on electrophysiological gap-detection testing. Information that aids clinicians in distinguishing between poor gap-detection thresholds due to central auditory processing deficits or due to attention could lead to more confident diagnoses and more appropriate treatment for clients.  4.1.1 Behavioural Accuracy The results of the present study showed that 35 participants reached a plateau of nearly 100% hits by a gap duration of 12 ms using the modified GIN procedure during the active-task EEG session. Normal gap-detection thresholds are considered to be between 2 and 20 ms, thus confirming that the participants in the present study fell within the normal range (Keith, 2000).     28 4.1.2 Inter-Rater Reliability An excellent Cohen’s kappa of 0.93 was found for the inter-rater reliability for the two raters’ judgements of N1 waves as being present or absent. A high consensus was seen between the two raters at all gap durations. High inter-rater reliability indicated that raters did not have substantial difficulty in identifying N1 waves. This increased my confidence in interpreting that the N1 waves were in fact electrophysiological responses.  Inter-rater reliability was higher than that seen for the three raters in the passive task (Angel, 2016). This is likely because of the difference in CAEP rating procedure in the present study for the active task. Due to the presence of the N2b waves, which could be easily mistaken for N1 waves without knowledge of the gap duration, raters judged ERP waves in the active task for all gap durations of each participant on the same screen. This protocol, which is more similar to clinical practices, allowed raters to more clearly see the decrease in N1 amplitude with decreasing gap duration, down to threshold. In comparison, raters judging ERPs in the passive task were blinded to gap duration and only viewed one gap duration at a time. Therefore, the raters judging the active-task ERPs could be considered to have an advantage in waveform judgement compared to the raters judging the passive-task ERPs. This advantage is likely the cause of the improvement in inter-rater reliability from the passive task to the active task.   4.1.3 Gap-Detection Thresholds The results of a 4-way ANOVA for 33 participants showed that there were no significant differences in behavioural gap-detection thresholds obtained with the modified Hughson-Westlake procedure, electrophysiological gap-detection thresholds obtained using the N1   29 thresholds in the passive task, electrophysiological gap-detection thresholds obtained using the N1 thresholds in the active task, and behavioural gap-detection thresholds obtained using the modified GIN procedure during the active task. These results are in agreement with the hypothesis that electrophysiological gap-detection thresholds would not be significantly affected by attention within a normal population.   Previous research has also found that behavioural and electrophysiological gap-detection thresholds are correlated. Michalewski et al. (2005) showed that normal hearing participants had an average N1 gap-detection threshold of 5 ms, which agreed with their behavioural gap-detection thresholds. In addition, Palmer and Musiek (2014) found that both younger and older adults had behavioural and electrophysiological gap-detection thresholds that agreed within 2 ms of each other. Angel (2016) showed that passive CAEP gap-detection thresholds agreed with behavioural thresholds within 0.8 ms, using a larger sample size than previous studies. Results from the present study add to this finding that agreement between behavioural and electrophysiological gap-detection thresholds (for normal listeners) occurs whether or not the participant is paying attention.   I hypothesized that some participants would have present N1-P2 waves without a corresponding button press, or in other words, higher behavioural thresholds than electrophysiological thresholds, due to waxing and waning of attention. In the present study, however, there were no significant differences between participants’ behavioural and electrophysiological gap-detection thresholds, and no hypothesized “misses” (i.e. N1-P2 waves without a button press) were judged to be present. As this data was collected using normal listeners, future research could collect   30 normative data on electrophysiological gap-detection thresholds for populations with attention deficits and central auditory processing deficits. If poor behavioural gap-detection thresholds with normal electrophysiological thresholds were seen in the attention deficit population (i.e. “misses”), but thresholds were equally poor behaviourally and electrophysiologically in the CAPD population, then this could help in differentiating between the two populations. Thus, CAEP recording during an active task may possibly have a clinical utility, but further research is needed before making such a claim.   4.1.4 Effect of Attention on N1-P2 Amplitudes A paired-samples t-test revealed that N1-P2 amplitudes were significantly larger in the active than the passive task. The mean N1-P2 amplitude in the active task was 5.57 uV, compared to 3.08 uV in the passive task. These results are in agreement with the hypothesis that N1-P2 amplitudes would be higher in the active task. As expected, this is consistent with previous research findings that attention has an effect on CAEPs, specifically that attention increases N1-P2 amplitudes (Picton & Hillyard, 1974). Hillyard et al. (1973) suggested that the increased N1 amplitude to tones that occurs with selective attention is related to increased neural activity in the primary auditory cortex.  4.2 Clinical Implications The purpose of this study was to assess the effect of attention on gap-evoked CAEPs, with the goal of potentially identifying a clinical relevance of using electrophysiological gap-detection testing while participants perform an active task in order to identify attentional confounds in CAPD evaluation. However, in the active task there was an unexpected finding that generated a   31 confound in using the original design. For the 0 ms gap duration ERPs as well as subthreshold gap duration ERPs, an N2b wave was seen for all participants. This N2b wave was identified as being distinct from the N1 wave using topography (as shown in Figure 3.5), but it is clear that the N2b wave could easily be mistaken for the N1 due to similarities in morphology and latency in the time-domain waveform. Clinically, the N2b would be a confound for identifying gap-detection thresholds using CAEPs during an active task. This is because at lower gap durations an N2b wave may be judged as an N1, thus giving an erroneously lower gap-detection threshold. This could result in a high number of type II errors, as people with high gap-detection thresholds may be missed. Therefore, with the paradigm that was used in the present study, the active task would be very difficult to use to evaluate gap-detection threshold.    A major limitation that led to the confound of the N2b wave being present was that the gaps always occurred at 500 ms post-stimulus onset. This created an expectancy for participants that a gap would probably occur at 500 ms and that they would need to respond at this time. When this expectancy was not fulfilled, as in the case of the 0 ms gap and subthreshold gap conditions, an N2b-P3a wave was generated (discussed further in Chapter 6). In order to use active-task CAEP paradigms clinically, the timing of gap presentation would need to be unpredictable to the participants rather than occurring at a set onset time during the noise burst. This would likely demolish the N2b-P3a wave, thus avoiding the difficulty in distinguishing it from the N1 wave at low gap durations. This would be necessary clinically because scalp topographies are typically not available in clinical set-ups, so they could not be used to aid clinicians in differentiating between waveforms. Furthermore, using scalp topographies in clinic may not be feasible even if they were available due to time constraints. Future research would, however, be needed to   32 confirm that jittering the timing of gap onsets would be sufficient to obliterate the N2b-P3a responses in order to use the active-task CAEPs in estimating electrophysiological gap-detection thresholds.  A future goal of this line of research could be to determine if CAEPs during an active task could be used to help distinguish between poor behavioural gap-detection thresholds that are due to CAPD and those that are due to attention deficits. If an N2b-P3a is not seen between 100 and 300 ms after the expected onset of a gap on a 0 ms gap duration trial, this could mean that the participant’s attention is waxing and waning, and therefore that attention may be a confound in behavioural CAPD evaluation for that person. Future studies using populations with attention deficits would be required to test this hypothesis. It is also tempting to consider that if a participant has an N2b-P3a response to a gap to which they also had a passive (i.e. N1) response, this may indicate that they behaviourally perceived the gap but withheld their button press for some reason. However, this conclusion cannot be made with any certainty given the current study’s results. Alternatively, it is possible that N1 waves may occur to gaps without conscious perception, but more research would be required to determine this. Furthermore, only a very small minority of people exaggerate on hearing tests, and to assume that someone is exaggerating without strong objective evidence could have a large negative impact and prevent them from receiving appropriate treatment. Overall, further research is needed to determine if CAEP testing during an active task is useful in distinguishing central auditory processing deficits from attention deficits for those people undergoing CAPD evaluation.    33 Another interesting future research direction could be to explore the utility of objective measures of gap-detection in cases of hidden hearing loss. Hidden hearing loss occurs when hazardous noise causes damage to ribbon synapses between inner hair cells and spiral ganglion neurons (Kujawa & Liberman, 2009). This leads to a loss of low spontaneous rate auditory nerve fibres, often resulting in complaints of difficulty with speech in noise accompanying normal peripheral hearing thresholds (Furman, Kujawa, & Liberman, 2013). Because hidden hearing loss is thought to affect this aspect of temporal processing, it would be interesting to see if future research also found a link to elevated gap-detection thresholds. If so, because clients with histories of noise exposure are often involved in medico-legal cases, objective measures of gap detection such as CAEPs may be indicated.   4.3 Conclusions The present study’s results showed that using an active gap-detection task while recording CAEPs does not significantly affect gap-detection thresholds, although it does increase the amplitudes of the N1-P2 responses. If gaps occur with predictable timing, an additional wave, the N2b, can be seen when participants do not perceive a gap. Because the N2b may overlap with the N1 wave in both latency and morphology, it could lead to errors in the visual judgement of electrophysiological gap detection thresholds. Therefore, while CAEPs to gaps during a passive-listening task may be a useful tool to evaluate gap detection ability in populations who cannot reliably respond in a behavioural task, the use of CAEPs to gaps during an active-listening task using the current paradigm presents some issues. Future research using modified active-task protocols would be needed to validate the utility of using gap-evoked CAEPs in an active task in order to electrophysiologically estimate gap-detection thresholds.   34 Chapter 5: Introduction of Experiment 2  5.1 N2 Background The N2, or N200, is an ERP that can be seen as a negative wave peaking 200 to 350 ms after a visual or auditory stimulus using EEG (Folstein & Van Petten, 2008). It is classified as an endogenous ERP, because it represents the interpretation of incoming sensory information.   In the auditory modality, the N2 can be separated into three subcomponents: the N2a, N2b, and N2c. The N2a, usually called the mismatch negativity (MMN), is evoked by a deviation from a repeating stimulus or repeating pattern of sounds. It is referred to as a pre-attentive ERP because it can be evoked even without conscious attention (Picton, 2010). The MMN has an anterior scalp distribution (Patel & Azzam, 2005). The N2b, while also elicited by deviant stimuli, requires selective attention on the part of the participant (Picton, 2010). It is typically seen when a prepotent, but incorrect, response is primed. The N2b can be distinguished from the MMN by its more central scalp distribution (Patel & Azzam, 2005). Lastly, the N2c is evoked by tasks requiring classification of stimuli and has a frontocentral distribution (Patel & Azzam, 2005).    The P3a wave, just following the N2b at 200 to 300 ms post-stimulus onset, usually occurs to a deviant auditory stimulus. It does not require conscious attention, and typically has a frontal scalp distribution (Picton, 2010). The P3a is evoked in response to an auditory or visual target, and appears to increase in amplitude as the target decreases in probability of being presented over time (Picton, 2010; Folstein & Van Petten, 2008).    35 Another negative wave that can be seen around the same time as the N2 is the error-related negativity (ERN or Ne). This wave is elicited when participants are aware that they have made an error, and it has a similar scalp distribution and waveform morphology to the N2b (Folstein & Van Petten, 2008). Dipole modeling has shown that the source of the ERN is the anterior cingulate cortex (ACC). The ACC is also suspected to be the generator for the N2b, thus suggesting that the ERN and N2b may share a similar underlying neural process (Nieuwenhuis, Yeung, Van Den Wildenburg, & Ridderinkhof, 2003).   The focus of the present study (described in Chapters 5 to 8) was on the N2b component, because N2b waves were generated during an auditory gap-detection task when participants were required to press buttons to gaps in noise bursts but withhold their button presses when no gaps occurred (see Chapters 1 to 4).  5.1.1 N2b Component  5.1.1.1 Go/No-Go and Stop Signal Paradigms The N2b is often evoked using a go/no-go paradigm. In this task, a participant must quickly respond to one type of stimulus (the go stimulus) and withhold a response to a second type of stimulus (the no-go stimulus). In these tasks, an N2b ERP occurs 200 to 300 ms post-stimulus and is typically larger after no-go than go trials (Donkers & Van Boxtel, 2004).   A second task that may be used is the stop-signal paradigm (Logan, Cowan, & Davis, 1984). In this more difficult task, the participant is required to repeatedly respond to a specific stimulus   36 (the S1); however, occasionally the primary stimulus is closely followed by a stop-signal stimulus (the S2) in which case the participant must inhibit their prepotent response to the S1. The longer the delay between the primary stimulus and the stop stimulus, the more difficult it is for the participant to stop their response. This can be explained by the Horse Race Model (Logan et al., 1984). The Horse Race Model postulates that the cognitive processes underlying the go response race against the processes underlying the stop response, and whichever processes are completed first dictate which action is carried out. A measure often used in this paradigm is the stop-signal reaction time (SSRT), which is thought to reflect the latency of the inhibitory response (Logan et al., 1984).  5.1.1.2 Stimulus Modality In the past, there was some debate over whether or not the N2b could be elicited by auditory stimuli. More recently, several studies have shown that the N2b can, in fact, be evoked by auditory stimulation (Nieuwenhuis, Yeung, & Cohen, 2004; Ramautar, Kok, & Ridderinkhof, 2006). For instance, Nieuwenhuis et al. (2004) argued that if the N2b represented a general cognitive control process it should be elicited by various modalities. They suggested that previous auditory studies had not created enough of a bias for participants to respond to the go stimuli, either because the go and no-go stimuli were too similar or because there was no pressure to respond within a certain amount of time. Nieuwenhuis et al. (2004), therefore, created a go/no-go task using letters which included a visually-similar block (using ‘F’ and ‘T’) as well as an auditorily-similar block (using ‘F’ and ‘S’). They found that the resultant N2b amplitude was greater for the more difficult tasks regardless of modality (i.e. when ‘F’ and ‘T’ had to be distinguished in the visual task compared to ‘F’ and ‘S’, and when ‘F’ and ‘S’ had to be   37 distinguished in the auditory task compared to ‘F’ and ‘T’). This led to the conclusion that the N2b could be elicited by either modality (Nieuwenhuis et al., 2004). Furthermore, a study by Ramataur et al. (2006) using a stop-signal paradigm showed that for visual primary stimuli, an auditory stop signal led to faster SSRTs than a visual stop signal. ERPs from this study revealed that the auditory stop signal evoked larger P3a amplitudes whereas the visual stop signal evoked larger N2b amplitudes (Ramautar et al., 2006). The faster reaction time with auditory stop signals may have been due to the fact that peripheral transmission of auditory information is faster than that for visual information. In addition, Ramautar et al. (2006) hypothesized that the smaller auditory N2b could have been due to the fact that the auditory P2 peaks later than the visual P2, thus it may have overlapped the N2b. Finally, Nakata, Arakawa, Suzuki, and Nakayama (2016) found that while there was no difference in the latency of the N2b or P3a between visual and auditory stimuli in go/no-go tasks, the N2b and P3a amplitudes were both larger for the visual modality. They suggested that there may be two separate neural networks for response inhibition based on sensory modality.  5.1.1.3 Inhibition Theory vs. Conflict Monitoring Theory While the N2b is thought to represent some cognitive control process, there is no consensus on whether this process is motor response inhibition or monitoring of response conflict. The N2b has traditionally been associated with inhibition of a motor response. Inhibition is a part of executive control, which is partly mediated by the prefrontal cortex, and encompasses the ability to suppress a prepotent or reflexive response (Falkenstein, 2006). Inhibition is an important function because it allows humans and other animals to adapt their behavior based on both external cues and internal goals (Dimoska, Johnstone, & Barry, 2006). Several studies appear to   38 provide evidence for the inhibition theory of the N2b. Jodo and Kayama (1992) found that the amplitude of the N2b to visual no-go stimuli in a go/no-go task was increased when there was higher pressure on participants to respond quickly. They interpreted this to mean that a stronger bias to respond was created for this group, and thus greater inhibition was needed on no-go trials, resulting in the N2b amplitude increase. Similarly, Falkenstein, Hoormann, and Hohnsbein (1999) found that participants with fewer commission errors or false alarms, interpreted as those participants having stronger inhibitory control, had a larger and earlier N2b. While a larger N2b amplitude is seen to no-go compared to go trials when covert responses in the form of silent counting are used, the N2b amplitude to no-go trials is even larger when overt responses in the form of button pressing are used, suggesting that N2b amplitude increases as the amount of inhibition required increases (Bruin & Wijers, 2002; Pfefferbaum et al., 1985; Folstein & Van Petten, 2008). In addition, in studies of children with ADHD, which is thought to be associated with poor inhibitory control, decreased N2b amplitudes were seen along with slower SSRTs and reduced probability of inhibition (Dimoska, Johnstone, Barry, & Clarke, 2003; Pliszka, Liotti, & Woldorff, 2000).   In contrast, some researchers believe that the underlying cognitive process of the N2b is actually the monitoring of response conflict. Conflict monitoring is an evaluative cognitive control process that takes place in the anterior cingulate cortex (ACC). The theory is that the ACC monitors any conflict in the environment or in performance on a task, and if increased cognitive control is required, the ACC signals for the dorsolateral and ventrolateral prefrontal cortex (dlPFC and vlPFC) to regulate cognitive control by, for instance, reallocating attention or overriding inappropriate responses (Larson, Clayson, & Clawson, 2014). The dlPFC may bias   39 the parietal cortex to reduce conflict on subsequent trials in what is known as the conflict-control loop (Carter & Van Veen, 2007). In the case of go/no-go tasks, the ACC monitors the conflict between the go response, which is prepotent, and the no-go response. ACC activity is greatest when a rare response is required, because there is competition from the bias to perform the more frequent response (Braver, Barch, Gray, Molfese, & Snyder, 2001). Impaired ability to monitor conflict and make appropriate behavioural adjustments is often seen in those with TBIs (Larson et al., 2014). Many studies have provided support for the conflict monitoring theory of the N2b. Nieuwenhuis et al. (2003) conducted a go/no-go study where no-go trials were more frequent than go trials (80% vs. 20%). They found that an N2b was present with greater amplitude to the go trials, which was then localized to the ACC using dipole modeling. This contradicts the inhibition theory because no inhibition is required on go trials. Other studies have found a similar effect to Nieuwenhuis et al. (2003) with a larger N2b to less frequent trials whether or not inhibition was required (Donkers and Van Boxtel, 2004; Enriquez-Geppert et al., 2010). This is likely because highly probable responses are primed, meaning that rare stimuli lead to conflict. Lastly, a recent hybrid go/no-go and Eriksen flanker (Eriksen & Eriksen, 1974) task, which had trials with combinations of low or high conflict and low or high motor inhibition requirements, found greater N2b amplitudes for high conflict than low conflict trials regardless of inhibition requirements (Groom & Cragg, 2015). This suggested that the N2b is a marker of response conflict (whether between prepotent and rare responses or due to incongruent flankers) rather than motor inhibition.    40 5.2 Exogenous vs. Endogenous Stimuli The vast majority of go/no-go studies, including the various studies discussed earlier in this chapter, have utilized exogenous (i.e. externally-generated) stimuli to illustrate the presence of the no-go N2b. One of the first studies to show a that larger N2b is evoked to no-go than go trials used exogenous stimuli in the form of visual words and symbols (Pfefferbaum, Ford, Weller, & Kopell, 1985). Other early studies did the same, including Eimer (1993), who used visual letters with varying locations to show that no-go stimuli evoked a higher amplitude N2b than go stimuli regardless of whether no-go trials were less frequent than go trials or equally as frequent as go trials. Similarly, Jodo and Kayama (1992) showed that lights that differed in location on no-go compared to go trials could evoke a no-go N2b. There are also several examples of stop-signal tasks in which N2b waves were evoked by exogenous stop signals when participants successfully inhibited their responses. For instance, Enriquez-Geppert, Konrad, Pantev, & Huster (2010) found larger N2b amplitudes on less frequent stop trials compared to more frequent go trials using a visual stop signal that differed in colour from the primary stimulus. Furthermore, larger N2b amplitudes on stop trials were also seen in a study that employed an exogenous visual stop signal that differed in shape from the primary stimulus (Kok, Ramataur, De Ruiter, Band, & Ridderinkhof, 2004).   In comparison, there is very little information in the literature on no-go N2b waves evoked by endogenous (i.e. internally-generated) no-go signals. A relevant study using endogenous auditory stimuli, although not in a go/no-go task, recorded EEG while participants pressed a button that produced a sound either 0%, 50%, or 88% of the time. The results showed an N2 wave in the 88% condition when omission was rare (SanMiguel, Widmann, Bendixen, Trujillo-Barreto, &   41 Schroger, 2013). The authors suggested that the N1 wave also seen in the omission trials represented a template of the predicted incoming sensory input, and that the N2 could be attributed to the unexpected omission. Furthermore, in a study on auditory stream segregation performed by Winkler, Takegata, and Sussman (2005), participants were presented with high- and low-pitch tones which could be perceived as either a galloping pattern or as two different streams. Occasionally a low-pitch tone would be omitted. When tones were presented with a short enough duration, an MMN-like ERP was evoked to the omission, regardless of whether the tones were perceived as one or two streams, and an N2b-like ERP was also evoked to the omission, but only when the tones were perceived as one stream. The authors suggested that the N2b was evoked when the tones were perceived as a single stream or pattern because the participants were actively attending to all of the tones rather than switching between attending to the high or low tone stream, and thus actively detected the omissions.   While the previous two studies have shown that N2 waves can be evoked by omission of expected stimuli, there do not appear to be any studies focusing specifically on the omission of gaps in noise and the N2b. However, there is one study conducted by Lange (2008) which briefly mentions a no-go N2 evoked by the omission of gaps. This study investigated how pitch and time expectations would affect auditory processing using a 12-tone induction sequence which was followed by a final tone. The final tone was either continuous or contained a gap, and in one of the experiments, participants were required to perform a go/no-go task in which they responded to the gaps in the final tone with a button press or withheld responses if the final tone was continuous. This study focused primarily on how being able to predict the pitch and timing of the final tone resulted in faster reaction times, more accurate responding, and an attenuated N1   42 wave. However, they also reported a frontal N2 wave which appeared to be larger for the no-go (no gap) than go (gap) trials. No further analyses on this N2 wave were reported. This study also had a relatively small sample size (n=13). Therefore, although an N2b to an endogenous stimulus has been briefly reported before, the present study aimed to further investigate this phenomenon using a larger sample size.  While the effects of gap-detection on the N2b are still relatively unknown, there are several studies which have used the N2a, or MMN, as an indicator of gap detection. Desjardins, Trainor, Hevenor, & Polak (1999) found that the MMN was evoked by gaps as short as 4 ms in duration. In addition, Bertoli, Heimberg, Smurzynski, & Probst (2001) established that MMN gap-detection thresholds were higher than behavioural thresholds. However, in regard to the N2b, the present study appears to be relatively unique in that it used a modified go/no-go paradigm with gaps in noise as go stimuli and the omission of an expected gap in noise as an endogenous no-go stimulus to evoke an N2b ERP.   5.3 Purpose The purpose of the present study was to explore whether an endogenous event could evoke an N2b. More specifically, this study gathered information on the auditory no-go N2b evoked by the absence of a gap in noise during a modified go/no-go paradigm. This study used a purely endogenous auditory no-go signal to evoke an N2b. The no-go signal was the participants’ unfulfilled expectation of perceiving a gap in noise, created by both predictable timing and a high probability of gaps occurring across trials.    43 5.4 Hypotheses The main hypothesis of the present study was that an N2b wave could be evoked by an endogenous auditory signal in the form of the expectancy of a gap in noise with predictable timing. I also hypothesized that there would be a subset of participants that did not have an N2b wave to the 0 ms gap condition because they were not actively attending to the task, meaning that they were either not actively inhibiting their button presses to the 0 ms gap condition or they did not actively detect any conflict between possible responses. I further hypothesized that smaller N2b amplitudes would be correlated with higher variability in reaction times, perhaps along with longer mean reaction times, due to waxing and waning of attention that may have occurred for some participants. Lastly, I hypothesized that higher behavioural gap-detection thresholds would correlate with smaller N2b amplitudes because the probability of perceiving no gap and thus the probability of no-go trials would be increased for these participants, and previous studies have shown that N2b amplitudes increase as no-go trials become less frequent (i.e. Enriquez-Geppert et al., 2010).   44 Chapter 6: Methods of Experiment 2  6.1 Data Acquisition  6.1.1 Paradigm The data used in Experiment 2 in the present study are identical to the data used in Experiment 1. For details on the collection of data used in this study, refer to Chapter 2. Because participants were required to press a button when they detected a gap and to withhold their button press when they did not detect a gap, this procedure can be considered to be a go/no-go paradigm. The current study utilized a go/no-go paradigm with the modification of a warning stimulus. The warning stimulus was the onset of the broadband noise burst at 0 ms, the go stimulus was the perceived gap that always occurred at 500 ms, and the no-go stimulus was the absence of a gap in the noise burst. The noise-burst onset or warning stimulus did not cue the participants as to whether or not they needed to make a response but rather directed their attention to the stimulus for processing. The go stimuli were gaps in noise of 2, 4, 6, 8, 10, 12, 14, 16, 18, or 20 ms in duration which always occurred at 500 ms post-noise burst onset. Participants were instructed to respond to gaps in noise with a button press. The no-go stimulus was considered to be the 0 ms gap (i.e. no gap) condition, or in other words, the continuation of the noise burst. When participants did not perceive a gap in noise at 500 ms post-noise burst onset, they were instructed to withhold their button press. For all gap durations, 100 trials were collected, thus technically go stimuli were more probable than no-go stimuli with a ratio of 10:1. However, for participants with thresholds above 2 ms, this ratio would decrease in that go stimuli would become slightly less likely and no-go stimuli would become slightly more likely.    45  The fact that the gaps always occurred at 500 ms post-noise burst onset was necessary to create a bias for participants to expect or predict a go stimulus at this particular timing in each trial. Furthermore, because the go stimuli were more probable than the no-go stimuli, this also created a response-probability bias in that they would need to respond by pressing the button on the majority of trials. This bias could be modulated by their gap-detection threshold, such that a participant with a higher threshold would have a lower probability of perceiving a go trial than a participant with a lower threshold. Because of this expectancy bias, the no-go stimulus in this task was really the participants’ expectation of perceiving a gap at 500 ms post-noise burst onset (that did not occur), making it an endogenous rather than exogenous no-go signal.  6.2 Data Analysis  6.2.1 CAEP Rating Two raters judged the active-task ERPs for the presence or absence of N2b and P3a waveforms for each of 35 participants. The Hits and Correct Rejections and the Misses and False Alarms were presented on separate screens (see Figure 2.1 and Figure 2.2). For a detailed description of this procedure, see Chapter 2, section 2.2.2.3. After judging the presence or absence of the N1 waves for each gap duration for a given participant, an additional column appeared to the raters at the far right of the screen. In the Hits and Correct Rejections screen, this column included both a percentage of correct behavioural responses (i.e. number of correct responses/total number of trials x 100) as well as a fraction of the number of correct responses (hits for each gap duration of 2 through 20 ms or correct rejections for 0 ms gap duration) over total trials for each gap   46 duration for each participant. In the Misses and False Alarms screen, this column displayed the percentage of incorrect responses (i.e. number of incorrect responses/total number of trials x 100) as well as the corresponding fraction of incorrect responses (misses for gap durations of 2 through 20 ms or false alarms for 0 ms gap duration) over total trials for each gap duration for each participant. The purpose of this column was to inform raters of the waveforms displayed that were visually noisy because of a low number of trials, thus aiding them in rating waves as Could Not Evaluate. The benefit of having this column appear only after rating the N1 waves was to avoid rater bias in deciding if an N1 wave was present or absent based on the participants’ behavioural responses. Once this column appeared, raters could then make a judgement on whether the N2b wave and/or the P3a wave was present for each gap duration for each participant. Because the N2b is known to be larger frontally and the P3a is known to be larger posteriorly, raters were presented with waveforms from Fz, FCz, Cz, and Pz electrodes in order to aid in judgements.   6.2.2 Inter-Rater Reliability Inter-rater reliability between the two raters was determined for judging the presence or absence of the N2b and P3a waves in the active task. Both percent agreement and Cohen’s kappa (Cohen, 1960) were calculated as measures of inter-rater reliability between the two raters for both the N2b and the P3a waves in the same manner as described for the N1 wave in Chapter 2.  6.2.3 Correlation Between N2b-P3a Amplitudes and Behavioural Thresholds N2b-P3a amplitudes were found and compared to behavioural thresholds for 33 participants (excluding participants who were rejected due to high residual noise during waveform rating).    47  Active-task N2b-P3a amplitudes for correct rejections at 0 ms gap duration and misses at 2 ms gap duration were determined through peak-picking in Matlab. A plot of amplitude by time was displayed on a computer screen with the mean waveform for the given gap duration (in black) along with two split buffer replications (in pink and green) for each participant, recorded from a Cz electrode, overlaid over the other gap duration (in grey) for the same participant. The time window of 550 ms to 1000 ms post-stimulus onset was marked by vertical red dotted lines on the plot. Scalp topographies were displayed on the screen below the amplitude-time plot for both the N2b and P3a markers. Topography plots were colour-coded (where blue represents negativity and red represents positivity) and had electrode locations labelled. These plots were used to confirm correct identification of the N2b and P3a. The criteria set to identify the N2b was to choose the most negative peak in the time window that replicated with the waveform morphology of the split buffer replications and the other gap duration for the same participant. N2b identification was always checked by looking for central topography. The criteria set to identify the P3a included choosing the first positive peak following the N2b that replicated with the waveform morphology of the split buffer replications and the other gap duration for the same participant. Again, P3a identification was always checked by looking for frontal topography. In the case of a double-peaked P3, the first peak was chosen in an effort to find the P3a rather than the P3b. If no clear response was present or visual replication was poor for a specific gap duration for a particular participant, the waveform was not marked.  I calculated the Pearson’s correlation coefficient between the N2b-P3a amplitudes for the 0 ms gap duration and the participants’ behavioural gap-detection thresholds using both the Hughson-  48 Westlake procedure as well as the modified GIN procedure during the active CAEP task. If no N2b-P3a amplitude had been determined for a participant, they were excluded from the sample. Overall, 7 participants were excluded, leaving a total of 26.   6.2.4 Correlation Between N2b-P3a Amplitudes and Latencies and Reaction Times Mean reaction time across all trials as well as trial-to-trial reaction time variability (i.e. standard deviation) were found for each participant. I used multiple linear regression to find the relationship between the independent variables (mean N2b-P3a amplitude, mean N2b latency, and mean P3a latency) and the dependent variable (mean reaction time or reaction time standard deviation). However, an assumption of multiple linear regression is that the probability density function of the dependent variable has a normal or Gaussian distribution. Therefore, a Shapiro-Wilk test of normality was performed in Matlab, which showed that the reaction time data were not normally distributed but rather skewed toward the right. The distribution of reaction times was consequently fitted using an exGaussian distribution model in Matlab. The exGaussion function is often used to describe the decision-making and sensory transduction aspects of reaction time, respectively (Lacouture & Cousineau, 2008; Luce, 1986). This involved finding the parameter values that best fit the distribution of reaction times using the Simplex algorithm in Matlab and then evaluating the fit using the likelihood criterion (Lacouture & Cousineau, 2008). Two multiple linear regression analyses were then performed in Matlab to find the relationships between the above independent variables and the exGaussian mean reaction time and between the above independent variables and the exGaussian reaction time standard deviation.    49 6.2.5 Comparison of N2b-P3a Amplitudes Between Correct Rejections and Misses N2b-P3a amplitudes were compared between correct rejections for the 0 ms gap duration and misses at the 2 ms gap duration. This was done using paired samples Student’s t-tests. Thus, only participants with ERPs that were rated as having N2b waves present for both the correct rejections for the 0 ms gap duration condition and the misses for the 2 ms gap duration condition were included in this analysis.   50 Chapter 7: Results of Experiment 2  7.1 Inter-Rater Reliability The percent agreement between the two raters was 89% for judgement of the N2b waves as present or absent, 94% for judgement of the P3a waves as present or absent, and 92% for judgements of CNE. The Cohen’s kappa for the two raters’ judgements was 0.63 for the N2b waves, 0.51 for the P3a waves, and 0.77 for CNEs. This indicated moderate inter-rater reliability for the N2b, weak inter-rater reliability for the P3a waves, and moderate inter-rater reliability for CNEs (McHugh, 2012). The consensus between the two raters is shown in Figure 7.1.     51  Figure 7.1: Consensus Among Raters for the Presence of the N2b and P3a Waves and CNEs Percent of participants rated as having a present N2b wave (top), P3a wave (middle), or rated as CNE (bottom) across gap durations for both correct (left) and incorrect (right) responses. Rater 1’s judgements are shown in blue, Rater 2’s judgements are shown in green, and consensus between the two raters is shown in red.   7.2 Presence of N2b Across Participants The two raters both judged only 29/35 participants as having an N2b present at the 0 ms gap duration condition. However, when comparing the 0 ms gap duration waveforms of the remaining 6 participants to this group after CAEP rating was completed, it became clear that 100% of participants in fact had a present N2b wave for the 0 ms gap condition (see Figure 7.2).   52 The likely reasons that these 6 participants were not judged to have N2b responses present were that their N2b amplitudes were relatively smaller and their signal-to-noise ratios were lower compared to other participants. However, upon overlaying participants’ FCz recordings for the no-gap condition, all participants showed a typical N2b-P3a morphology. This was also confirmed by looking at the scalp topographies across all participants.  Figure 7.2: N2b Presence Across Participants ERPs during the active task split into participants that raters judged as having a present N2b (n=29; black) and participants that raters judged as having an absent N2b (n=6; red), recorded from the FCz electrode. The noise-burst onset P1-N1-P2 complex is labeled as well as the N2b and P3a or P3b waveforms. Note the rising morphology of the N2b waveform in both groups beginning at approximately the same time post-stimulus onset.   53 7.3 Correlation Between N2b-P3a Amplitudes and Behavioural Thresholds The correlations between the mean N2b-P3a amplitudes and the behavioural thresholds using both the Hughson-Westlake and the modified GIN procedure were found for a sample of 26 participants. The Pearson’s correlation coefficients and corresponding p-values are shown in Table 7.1. The results indicate no significant correlation between N2b-P3a amplitude and behavioural thresholds using either procedure.   N2b-P3a Amplitude and Hughson-Westlake Threshold N2b-P3a Amplitude and Modified GIN Threshold (During Active Task) Pearson’s r -0.1771 -0.0716 p-value 0.3867 0.7281 Table 7.1: Correlations Between N2b-P3a Amplitude and Behavioural Thresholds Using Two Procedures  7.4 Correlation Between N2b-P3a Amplitudes and Latencies and Reaction Times Multiple linear regression analyses were performed in order to find the relationships between N2b-P3a amplitude, N2b latency, P3a latency, and exGaussian mean reaction time, and between N2b-P3a amplitude, N2b latency, P3a latency, and exGaussian reaction time standard deviation. No significant correlations were found between N2b-P3a amplitude and exGaussian mean reaction time, N2b-P3a amplitude and exGaussian reaction time standard deviation, N2b latency and exGaussian mean reaction time, N2b latency and exGaussian reaction time standard deviation, or P3a latency and exGaussian mean reaction time (p>0.05 for all, see Table 7.2). A significant positive correlation was found between P3a latency and exGaussian reaction time standard deviation (p = 0.0341), as seen in Figure 7.3.    54  exGaussian Mean Reaction Time exGaussian Reaction Time Standard Deviation  R value p-value R value p-value N2b-P3a Amplitude -0.1155 0.5741 -0.0282 0.8913 N2b Latency 0.0801 0.6972 0.0604 0.7693 P3a Latency 0.3111 0.1219 0.4168 0.0341 Table 7.2: Results of Multiple Linear Regression Analyses   Figure 7.3: Correlation between P3a Latency and exGaussian Reaction Time Standard Deviation P3a latency was significantly correlated with exGaussian reaction time standard deviation (p<0.05).       55 7.5 Comparison of N2b-P3a Amplitudes Between Correct Rejections and MissesThe N2b-P3a amplitudes showed no significant difference between correct rejections for 0 ms gap duration and misses at the 2 ms gap duration (p = 0.544), as seen in Figure 7.4.  Figure 7.4: Comparison of N2b-P3a Amplitudes for 0 ms and 2 ms Gap Durations Mean N2b-P3a amplitudes were not significantly different between correct rejections and misses across participants (p>0.05).   56 Chapter 8: Discussion of Experiment 2  8.1 Interpretation of Results In a study on the use of CAEPs in an active gap-detection task, an unexpected N2b waveform was seen when gaps were either omitted or presented at subthreshold levels. The N2b typically occurs 200 to 300 ms post-stimulus onset when a prepotent response must be withheld, such as when a no-go stimulus is presented in a go/no-go paradigm (Folstein & Van Petten, 2008; Picton, 2010). It was retrospectively observed that the present study could be looked at as a modified auditory go/no-go paradigm in which the no-go signal was not an exogenous stimulus but rather the participants’ unfulfilled expectation of perceiving a gap at 500 ms post-noise burst onset. The purpose of the present study, therefore, was to analyze this endogenously-generated N2b waveform.  8.1.1 Inter-Rater Reliability The inter-rater reliability was found to be moderate for the judgement of N2b waves and CNEs, and weak for the judgement of P3a waves. The moderate inter-rater reliability for the N2b was likely due to the poor consensus between the two raters at high gap durations (shown in Figure 7.1). At high gap durations, occasionally an N2b wave was judged to be present by Rater 1 following the N1 wave when participants correctly identified gaps. The second negativity following the N1 wave is most likely a second N1 to gap offsets (i.e. second noise-burst onset) and Rater 1 mistook it for an N2b. It is possible that the second negativity could be an N2b due to shifting of the participant’s attention to the gap, however, this is less likely due to the fact that gaps were more frequent than no-gap conditions. Despite this minor difference between raters,   57 the consensus between the two raters was very high for gaps of 0 ms or 2 ms in duration, when the N2b waves of interest occurred. The moderate inter-rater reliability for CNEs may have been seen because Rater 2 was more likely than Rater 1 to judge a wave as a CNE for low gap durations, particularly for correct responses. Rater 2 may have been more conservative in judging waveforms that were visually noisy as compared to Rater 1. Lastly, Rater 2 was more likely than Rater 1 to judge the P3a as being present for low gap durations for both correct and incorrect responses, likely leading to the weak inter-rater reliability for judgement of P3a waves. This may have occurred because Rater 2 had greater prior experience in judgement of P3 waves than Rater 1, therefore leading to a higher ability to visually detect their presence.   8.1.2 N2b-P3a Amplitudes and Behavioural Gap-Detection Thresholds The results of this study indicated that there were no significant correlations between participants’ N2b-P3a amplitudes and their behavioural gap-detection thresholds using either the modified Hughson-Westlake or the modified GIN procedure. These results are in opposition to the hypothesis that higher behavioural gap-detection thresholds would effectively increase the probability of no-go trials for participants, creating less of a bias toward responding and therefore decreasing the N2b amplitude. These results also disagree with several studies’ findings that N2b amplitude is increased when no-go trials are rare compared to go trials and decreased as no-go trials become more frequent (Donkers & Van Boxtel, 2004; Enriquez-Geppert et al., 2010; Nieuwenhuis et al., 2003). The lack of a significant correlation between N2b amplitudes and behavioural gap detection thresholds, which in turn affect the probability of no-go trials, was likely due to a restriction of range problem in the present study. Because all of the participants had normal behavioural gap detection thresholds, the difference in probability of   58 no-go trials between participants was very small (e.g. for a participant with a 2 ms gap detection threshold the probability of not perceiving a gap was 1/11, and for a participant with a 6 ms gap duration threshold the probability of not perceiving a gap was only increased to 3/11). It is likely that the small range of probabilities was not enough to create a significant difference in N2b amplitude between participants. It is possible that if this study was repeated using populations with CAPD or attention deficits who have high behavioural gap-detection thresholds, the N2b wave amplitudes would be significantly smaller compared to the normal population.   8.1.3 N2b-P3a Amplitudes and Latencies and Reaction Time The results of a multiple linear regression indicated that there was no significant correlation between N2b-P3a amplitudes or N2b latencies and (exGaussian) mean reaction time or (exGaussian) reaction time standard deviation. This is in contrast to the hypothesis that smaller N2b-P3a amplitudes would be associated with longer mean reaction times and/or higher variability in reaction times, due to some participants’ attention waxing and waning throughout the task. This also disagrees with the results of a combined go/no-go and stop-signal task where participants with faster stop-signal reaction times were found to have larger N2b amplitudes (van Boxtel, van der Molen, Jennings, & Brunia, 2001). However, it is possible that these results are due to the normal population of participants in the present study. Studies of children with ADHD have found both slower stop-signal reaction times and decreased N2b amplitudes (Dimoska et al., 2003; Pliszka et al., 2000). Perhaps if this study were to be repeated with a population with attention deficits, a correlation between longer reaction times and/or greater reaction time variability and smaller N2b-P3a amplitudes might be seen.     59 The results did show a significant correlation between P3a latencies and (exGaussian) reaction time standard deviation (see Figure 7.3), in that earlier latencies were correlated with decreased reaction time standard deviation. This is expected because with a smaller range of reaction times for a given participant (possibly due to more sustained attention), it is more likely that the P3a waves for individual trials will line up and create a peak average P3a that occurs earlier.   8.1.4 N2b-P3a Amplitudes for 0 ms Gaps Compared to 2 ms Gaps The results of the present study showed no significant difference in N2b-P3a amplitudes between correct rejections that occurred to gaps of 0 ms duration and misses that occurred to gaps of 2 ms duration (see Figure 7.4). This indicated that the N2b wave occurred whenever a participant did not perceive a gap when expected, regardless of whether the gap was absent or simply occurred below their gap detection threshold. This is as expected because the N2b is an endogenous ERP, meaning it represents cognitive processing or significance of a signal rather than just the physical presence of the stimulus (Picton, 2010).  8.2 Conclusions on the N2b ERP  8.2.1 Presence of N2b Across Participants Although the two raters judged only 29 out of 35 participants as having a present N2b wave for 0 ms gap duration trials, it was later seen that the remaining 6 participants also had present and replicating N2bs, albeit with smaller amplitudes. In contrast to the hypothesis that a subset of participants would not have N2b waves present to the no-gap conditions due to a lack of sustained attention, these results show that all participants had sufficient attention to produce an   60 N2b wave. These results, from a moderate sample size (n=35), corroborate the results from studies showing that the N2b-P3a can be evoked by auditory stimuli (Nieuwenhuis et al., 2004).    The variability of the N2b amplitudes was fairly high and lead to a small group (n=6) not being initially judged to have N2b-P3a responses. It is possible that this group of participants had higher behavioural gap-detection thresholds, leading to a greater probability of no-go trials. Another possibility is that they had more waxing and waning of attention, leading indirectly to more no-go trials because they were not always paying sufficient attention to perceive and respond to the gaps. Either possibility would theoretically lead to smaller N2b amplitudes (Dimoska et al., 2003; Donkers & Van Boxtel, 2004; Enriquez-Geppert et al., 2010; Nieuwenhuis et al., 2003; Pliszka et al., 2000). However, the results discussed above showed no significant correlations between N2b amplitudes and behavioural gap-detection thresholds or between N2b amplitudes and reaction time standard deviation, which could relate to waxing and waning of attention. Therefore, it seems more likely that the N2bs of these 6 participants were missed by the raters due to higher levels of baseline noise, making visual identification more difficult.  8.2.2 Conflict Monitoring vs. Response Inhibition Theories The results of the present study can be looked at in light of both the conflict monitoring theory and the motor inhibition theory. It is possible that the results lend support for the conflict monitoring theory in the following manner. The onset of the noise burst, which acts as a warning signal, may prime the motor response, but it must be inhibited until the gap occurs. When a gap is detected, inhibition of the motor response is released and the motor command is sent from the   61 motor cortices down the cerebrospinal tract to execute the button press. A gap is likely detected through a decrease in the number of auditory nerve fibres that are firing due to the decrease in intensity of the incoming auditory stimulus during the gap. At some point in the neural pathway, this decrease in neural firing must be translated into an excitatory response through an inhibitory interneuron in order for the release of inhibition to occur in the motor cortex. If this speculation is true, then for trials with no gap at 500 ms (i.e. an endogenous no-go signal), the noise burst would lead to a steady firing of a relatively high number of auditory nerve fibres onto the same theoretical inhibitory interneuron, which would continue the inhibition of the motor response. In other words, there is no change in firing or increase in inhibition that occurs in the no gap condition at 500 ms and thus no exogenous change in stimulus input. Although this explanation is obviously oversimplified, the general principles may suggest that the N2b reflects conflict monitoring rather than motor inhibition when the no-go signal is endogenous. However, it is also possible that the results could be explained by the inhibition theory in that the endogenous signal itself, which is the expectation of a gap at 500 ms post-stimulus onset, leads to an increase in inhibition in the motor cortex.  Because of the limitations in the methodology of this study, namely that the paradigm was not set up to isolate response conflict or motor inhibition but was analyzed retroactively, no explicit conclusions can be made in this regard. It is clear that both response conflict and motor inhibition must occur in go/no-go paradigms when a no-go stimulus is presented. Various researchers have suggested that the N2b may represent response conflict and the P3a may represent subsequent motor inhibition (Bruin, Wijers, & Staveren, 2001; Dimoska et al., 2006; Donkers & van Boxtel, 2004; Enriquez-Geppert et al., 2010; Groom & Cragg, 2015; Smith,   62 Johnstone, & Barry, 2005). However, it is possible and most likely that these two processes cannot be entirely isolated from one another and that the output from a conflict monitoring network might signal a motor inhibition network to either release or strengthen the inhibition onto the pre-programmed motor response.   8.2.3 N2b to Endogenous Signals The main finding of the present study was that the N2b can be evoked by an endogenous auditory stimulus. In this case, an expectation of a gap occurring at 500 ms post-noise burst onset was created through predictable timing (the gaps always occurred at the same time point in each trial) and probability of occurrence across trials (gaps were more common than no gaps with a ratio of 10:1). When a trial deviated from a perceived gap trial template, an N2b was evoked. The noise-burst onset acted as a warning signal, which may have further placed a load on the participants to respond, further priming the generation of an N2b to no-gap trials. Overall, this shows that the auditory system does not need to be physically activated but rather needs to recognize a change in auditory pattern in order for an N2b wave to be evoked. In this study, the N2b was not created by a specific characteristic of an incoming stimulus (as has been previously reported) but was generated by the participants’ decision-making processes on whether or not a gap was perceived.   The use of an entirely endogenous no-go signal in an auditory go/no-go task seems to only occur in one other study. Lange (2008) briefly reported a no-go N2 evoked by the omission of gaps in tones in a study on the effects of expectation on auditory processing. This agrees with the present study’s findings, which replicated these results using a larger sample size. While Lange (2008)   63 did not report on the exact timing of the gaps within the tones, it can be assumed that the gaps occurred at a predictable onset time, as in the present study, setting up a timing expectation for the gap to occur and potentially leading to the N2 wave. However, unlike the present study, Lange (2008) did not have an expectation based on probability of occurrence. Interestingly, the no-go trials were actually more probable (62.5%) than the go trials (37.5%) in their study, opposite of the methodology of the present study. The fact that the N2 was larger to the no-go trials than the go trials even though the no-go trials were more frequent disagrees with several other studies suggesting that the N2b is larger to less frequent trials regardless of the response required (Donkers and Van Boxtel, 2004; Enriquez-Geppert et al., 2010; Nieuwenhuis et al., 2003). Perhaps this suggests that a timing expectation alone is enough to create a template and act as an endogenous signal that can evoke an N2 wave.  In addition, a study that used omitted signals to evoke endogenous ERPs was conducted by SanMiguel et al. (2013). They explored stimulus prediction and had participants press a button that would produce a sound with 0%, 50%, or 88% probability. They found an N2-like event following button presses when participants would expect a sound to occur but no sound was presented (e.g. 12% of trials in the 88% probability condition). Similarly, Winkler et al. (2005) found an N2b wave when an expected tone was omitted from a sequence of high- and low-pitch tones, but only when the sequence was perceived as a single stream or pattern. When the sequence was perceived as two separate streams, and ostensibly participants were not actively attending to both, omissions evoked an MMN instead. Both of these results agree with the present study’s finding that endogenous signals (i.e. participant expectancy or prediction) can lead to an N2 ERP.   64  8.3 Conclusions The present study found that the N2b ERP can be evoked by an endogenous signal in the form of the omission of a gap in noise. This indicates that the N2b-P3a can be evoked in the absence of an object that matches a neural template rather than the absence of or change in auditory stimulation. This provides evidence that the N2b-P3a complex occurs in response to endogenous signals of conflict with an expected event.   65 References Abramson, A. S., & Lisker, L. (1970). Discriminability along the voicing continuum: Cross-language tests. Proceedings of the Sixth International Congress of Phonetic Sciences, 196(7) 569-573.  American Speech-Language-Hearing Association. (2005). (Central) auditory processing disorders. Angel, R. (2016). Cortical Auditory Evoked Potentials to Estimate Gap-Detection Thresholds in Adults. cIRcle: UBC’s Digital Repository: Electronic Theses and Dissertations (ETDs) 2008+. Bergemalm, P., & Lyxell, B. (2005). Appearances are deceptive? Long-term cognitive and central auditory sequelae from closed head injury. International Journal of Audiology, 44(1), 39-49.  Bertoli, S., Heimberg, S., Smurzynski, J., & Probst, R. (2001). Mismatch negativity and psychoacoustic measures of gap detection in normally hearing subjects. Psychophysiology, 38(2), 334-342.  Braver, T. S., Barch, D. M., Gray, J. R., Molfese, D. L., & Snyder, A. (2001). Anterior cingulate cortex and response conflict: Effects of frequency, inhibition and errors. Cerebral Cortex, 11(9), 825-836.    66 Bruin, K., & Wijers, A. (2002). Inhibition, response mode, and stimulus probability: A comparative event-related potential study. Clinical Neurophysiology, 113(7), 1172-1182.  Bruin, K. J., Wijers, A. A., & van Staveren, A. S. J. (2001). Response priming in a go/nogo task: Do we have to explain the go/nogo N2 effect in terms of response activation instead of inhibition?. Clinical Neurophysiology, 112(9), 1660-1671. Carter, C. S., & Van Veen, V. (2007). Anterior cingulate cortex and conflict detection: An update of theory and data. Cognitive, Affective, & Behavioral Neuroscience, 7(4), 367-379.  Cestnick, L., & Jerger, J. (2000). Auditory temporal processing and lexical/nonlexical reading in developmental dyslexics. Journal of the American Academy of Audiology, 11(9), 491-503.  Chermak, G. D. (1997). Central auditory processing disorders. Chermak, G. D., Hall, J. W., & Musiekl, F. E. (1999). Differential diagnosis and management of central auditory processing disorder and. Journal of the American Academy of Audiology, 10(6), 289-303.  Chermak, G. D., Somers, E. K., & Seikel, J. A. (1998). Behavioral signs of central auditory processing disorder and attention deficit hyperactivity disorder. Journal-American Academy of Audiology, 9, 78-84.  Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20(1), 37-46.    67 Cooper Jr, J., & Gates, G. A. (1991). Hearing in the elderly-the framingham cohort, 1983-1985: Part II. prevalence of central auditory processing disorders. Ear and Hearing, 12(5), 304-311.  Desjardins, R. N., Trainor, L. J., Hevenor, S. J., & Polak, C. P. (1999). Using mismatch negativity to measure auditory temporal resolution thresholds. Neuroreport, 10(10), 2079-2082.  Dimoska, A., Johnstone, S. J., & Barry, R. J. (2006). The auditory-evoked N2 and P3 components in the stop-signal task: Indices of inhibition, response-conflict or error-detection? Brain and Cognition, 62(2), 98-112.  Dimoska, A., Johnstone, S. J., Barry, R. J., & Clarke, A. R. (2003). Inhibitory motor control in children with attention-deficit/hyperactivity disorder: Event-related potentials in the stop-signal paradigm. Biological Psychiatry, 54(12), 1345-1354.  Donkers, F. C., & Van Boxtel, G. J. (2004). The N2 in go/no-go tasks reflects conflict monitoring not response inhibition. Brain and Cognition, 56(2), 165-176.  Eimer, M. (1993). Effects of attention and stimulus probability on ERPs in a Go/Nogo task. Biological Psychology, 35(2), 123-138.  Enriquez-Geppert, S., Konrad, C., Pantev, C., & Huster, R. J. (2010). Conflict and inhibition differentially affect the N200/P300 complex in a combined go/nogo and stop-signal task. Neuroimage, 51(2), 877-887.    68 Eriksen, B. A., & Eriksen, C. W. (1974). Effects of noise letters upon the identification of a target letter in a nonsearch task. Attention, Perception, & Psychophysics, 16(1), 143-149.  Falkenstein, M., Hoormann, J., & Hohnsbein, J. (1999). ERP components in Go/Nogo tasks and their relation to inhibition. Acta Psychologica, 101(2), 267-291.  Falkenstein, M. (2006). Inhibition, conflict and the nogo-N2. Clinical Neurophysiology : Official Journal of the International Federation of Clinical Neurophysiology, 117(8), 1638-1640.  Fleiss, J. L., & Cohen, J. (1973). The equivalence of weighted kappa and the intraclass correlation coefficient as measures of reliability. Educational and Psychological Measurement, 33(3), 613-619.  Folstein, J. R., & Van Petten, C. (2008). Influence of cognitive control and mismatch on the N2 component of the ERP: A review. Psychophysiology, 45(1), 152-170.  Furman, A.C., Kujawa, S.G., & Liberman, M.C. (2013). Noise-induced cochlear neuropathy is selective for fibres with low spontaneous rates. Journal of neurophysiology, 110(3), 577-586. Groom, M. J., & Cragg, L. (2015). Differential modulation of the N2 and P3 event-related potentials by response conflict and inhibition. Brain and Cognition, 97, 1-9.  Hillyard, S. A., Hink, R. F., Schwent, V. L., & Picton, T. W. (1973). Electrical signs of selective attention in the human brain. Science (New York, N.Y.), 182(4108), 177-180.    69 Jerger, J., & Musiek, F. (2000). Report of the consensus conference on the diagnosis of auditory processing. Journal of the American Academy of Audiology, 11(9), 467-474.  Jodo, E., & Kayama, Y. (1992). Relation of a negative ERP component to response inhibition in a Go/No-go task. Electroencephalography and Clinical Neurophysiology, 82(6), 477-482.  Keith, R. (2000). Random gap detection test. St.Louis, MO: Auditec, 13. Kok, A., Ramautar, J. R., De Ruiter, M. B., Band, G. P., & Ridderinkhof, K. R. (2004). ERP components associated with successful and unsuccessful stopping in a stop-signal task. Psychophysiology, 41(1), 9-20.  Kujawa, S.G., & Liberman, M.C. (2009). Adding insult to injury: cochlear nerve degeneration after “temporary” noise-induced hearing loss. Journal of Neuroscience, 29(45), 14077-14085). Lacouture, Y., & Cousineau, D. (2008). How to use MATLAB to fit the ex-gaussian and other probability functions to a distribution of response times. Tutorials in Quantitative Methods for Psychology, 4(1), 35-45.  Lange, K. (2009). Brain correlates of early auditory processing are attenuated by expectations for time and pitch. Brain and cognition, 69(1), 127-137. Larson, M. J., Clayson, P. E., & Clawson, A. (2014). Making sense of all the conflict: A theoretical review and critique of conflict-related ERPs. International Journal of Psychophysiology, 93(3), 283-297.    70 Logan, G. D., Cowan, W. B., & Davis, K. A. (1984). On the ability to inhibit simple and choice reaction time responses: A model and a method. Journal of Experimental Psychology: Human Perception and Performance, 10(2), 276.  Luce, R. D. (1986). Response times: Their role in inferring elementary mental organization. Oxford University Press on Demand. McHugh, M. L. (2012). Interrater reliability: The kappa statistic. Biochemia Medica, 22(3), 276-282.  Michalewski, H. J., Starr, A., Nguyen, T. T., Kong, Y., & Zeng, F. (2005). Auditory temporal processes in normal-hearing individuals and in patients with auditory neuropathy. Clinical Neurophysiology, 116(3), 669-680.  Moore, D. R., Hartley, D. E., & Hogan, S. C. (2003). Effects of otitis media with effusion (OME) on central auditory function. International Journal of Pediatric Otorhinolaryngology, 67, S63-S67.  Musiek, F. E., Shinn, J. B., Jirsa, R., Bamiou, D., Baran, J. A., & Zaida, E. (2005). GIN (gaps-in-noise) test performance in subjects with confirmed central auditory nervous system involvement. Ear and Hearing, 26(6), 608-618.  Nakata, H., Arakawa, N., Suzuki, C., & Nakayama, M. (2016). Modality difference in N2 and P3 components between visual and auditory Go/No-go paradigms. Journal of Psychophysiology, 30(4), 131-140.    71 Nieuwenhuis, S., Yeung, N., & Cohen, J. D. (2004). Stimulus modality, perceptual overlap, and the go/no-go N2. Psychophysiology, 41(1), 157-160.  Nieuwenhuis, S., Yeung, N., Van Den Wildenberg, W., & Ridderinkhof, K. R. (2003). Electrophysiological correlates of anterior cingulate function in a go/no-go task: Effects of response conflict and trial type frequency. Cognitive, Affective, & Behavioral Neuroscience, 3(1), 17-26.  Palmer, S. B., & Musiek, F. E. (2014). Electrophysiological gap detection thresholds: Effects of age and comparison with a behavioral measure. Journal of the American Academy of Audiology, 25(10), 999-1007.  Patel, S. H., & Azzam, P. N. (2005). Characterization of N200 and P300: Selected studies of the event-related potential. International Journal of Medical Sciences, 2(4), 147-154.  Pfefferbaum, A., Ford, J. M., Weller, B. J., & Kopell, B. S. (1985). ERPs to response production and inhibition. Electroencephalography and Clinical Neurophysiology, 60(5), 423-434.  Phillips, D. P. (1999). Auditory gap detection, perceptual channels, and temporal resolution in speech perception. Journal of the American Academy of Audiology, 10(6), 343-354.  Picton, T. W. (2010). Endogenous auditory evoked potentials: Attention must be paid. Human auditory evoked potentials. Plural Publishing. Picton, T., & Hillyard, S. (1974). Human auditory evoked potentials. II: Effects of attention. Electroencephalography and Clinical Neurophysiology, 36, 191-200.    72 Pliszka, S. R., Liotti, M., & Woldorff, M. G. (2000). Inhibitory control in children with attention-deficit/hyperactivity disorder: Event-related potentials identify the processing component and timing of an impaired right-frontal response-inhibition mechanism. Biological Psychiatry, 48(3), 238-246.  Plomp, R. (1964). Rate of decay of auditory sensation. The Journal of the Acoustical Society of America, 36(2), 277-282.  Polich, J. (2007). Updating P300: An integrative theory of P3a and P3b. Clinical Neurophysiology, 118(10), 2128-2148.  Ramautar, J., Kok, A., & Ridderinkhof, K. (2006). Effects of stop-signal modality on the N2/P3 complex elicited in the stop-signal paradigm. Biological Psychology, 72(1), 96-109.  SanMiguel, I., Widmann, A., Bendixen, A., Trujillo-Barreto, N., & Schroger, E. (2013). Hearing silences: Human auditory processing relies on preactivation of sound-specific brain activity patterns. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 33(20), 8633-8639.  Shen, Y., & Richards, V.M. (2013). Temporal modulation transfer function for efficient assessment of auditory temporal resolution. The Journal of the Acoustical Society of America, 133(2), 1031-1042. Smith, J. L., Johnstone, S. J., & Barry, R. J. (2006). Effects of pre-stimulus processing on subsequent events in a warned Go/NoGo paradigm: Response preparation, execution and inhibition. International Journal of Psychophysiology, 61(2), 121-133.    73 Stach, B. A., Spretnjak, M. L., & Jerger, J. (1990). The prevalence of central presbyacusis in a clinical population. Journal of the American Academy of Audiology, 1(2), 109-115.  Stapells, D. R. (2002). Cortical event-related potentials to auditory stimuli. Handbook of Clinical Audiology, 5, 378-406.  van Boxtel, G. J., van der Molen, M.W., Jennings, J. R., & Brunia, C. H. (2001). A psychophysiological analysis of inhibitory motor control in the stop-signal paradigm. Biological Psychology, 58(3), 229-262.  Viemester, N.F. (1979). Temporal modulation transfer functions based upon modulation thresholds. The Journal of the Acoustical Society of America, 66(5), 1364-1380. Winkler, I., Takegata, R., & Sussman, E. (2005). Event-related brain potentials reveal multiple stages in the perceptual organization of sound. Cognitive Brain Research, 25(1), 291-299. 

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:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0360659/manifest

Comment

Related Items