Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An event-related potential investigation of inhibition of return Prime, David J. 2004

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

Item Metadata

Download

Media
831-ubc_2005-995348.pdf [ 4.07MB ]
Metadata
JSON: 831-1.0092368.json
JSON-LD: 831-1.0092368-ld.json
RDF/XML (Pretty): 831-1.0092368-rdf.xml
RDF/JSON: 831-1.0092368-rdf.json
Turtle: 831-1.0092368-turtle.txt
N-Triples: 831-1.0092368-rdf-ntriples.txt
Original Record: 831-1.0092368-source.json
Full Text
831-1.0092368-fulltext.txt
Citation
831-1.0092368.ris

Full Text

AN EVENT-RELATED POTENTIAL INVESTIGATION OF INHIBITION OF RETURN by DAVID J. PRIME B.A. (First Class Honors), Simon Fraser University, 1996 M.A., The University of British Columbia, 2000 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (PSYCHOLOGY) •  THE UNIVERSITY OF BRITISH COLUMBIA July 29, 2004  © David J. Prime, 2004  11  Abstract In visual spatial-cueing experiments with uninformative peripheral cues, reaction time to cued-location targets is facilitated (relative to that for targets at uncued-locations) at short cue-target intervals and inhibited at longer intervals. Posner, Rafal, Choate and Vaughan (1985) labeled this inhibitory effect Inhibition o f Return (IOR) and proposed that it serves to bias the visual system to orient towards new locations. This proposal remains a popular topic of debate and no consensus has yet been reached regarding the mechanisms responsible for producing IOR. The present project utilized event-related potential ( E R P ) measures o f brain activity to investigate the processing changes underlying IOR. I examined the effect o f uninformative visual cues both on the amplitude of early visual E R P peaks and on the motorrelated lateralized readiness potential (LRP). The L R P is a measure o f motor cortex activity that is related to the preparation of responses. The onset o f the L R P relative to the target and response provide relative measures of the duration o f premotor and motor processes, respectively. Chapters 2 and 3 present the results o f four experiments, in which observers were required to make manual responses to visual targets. In each experiment the target was preceded by an uninformative peripheral cue followed by a central re-orienting event. In separate experiments, subjects were required to make identity-based discrimination responses, localization responses, detection responses, or identity-based go-nogo responses. In each o f these experiments I O R was associated with a slowing o f premotor processes. The duration o f motor processes were not affected, however, indicating that, when measured with manual keypresses, I O R does not arise from inhibition o f motor processes. Consistent with a perceptual locus o f IOR, the amplitudes o f the occipital E R P peaks were reduced for cuedlocation targets relative to uncued-location targets. In a fifth experiment (Chapter 5) the reorienting event was omitted and subjects were required to discriminate the target identity. Compared to when a re-orienting event was present, the magnitude o f both I O R and the E R P amplitude effects were reduced. This co-variation of behavioural and E R P effects suggests that that changes in the perceptual processes are, at least in part, responsible for generating IOR.  Ill  TABLE OF CONTENTS Abstract  ii  TABLE OF CONTENTS  iii  List of Tables  v  List of Figures  '.  vi  List of Nomenclature  vii  Acknowledgements  viii  CHAPTER I: Introduction and Overview 1.1 Introduction 1.2 Background: Spatial Orienting and Inhibition of Return 1.2.1 The Spatial-cueing Paradigm 1.2.2 Stimulus Cues and Inhibition of Return 1.3 Proposed Mechanisms of IOR 1.3.1 Attentional Accounts 1.3.2 Response-related Accounts 1.3.3 Summary 1.4 Goals and Approach of the Present Project References CHAPTER II: Event-related Potential Correlates of Inhibition of Return in a Discrimination Task 2.1 Chapter Preface 2.2 Introduction 2.3 Method 2.4 Results and Discussion 2.5 Conclusion Footnotes References CHAPTER III: Event-related Potential Correlates of Inhibition of Return Across Tasks 3.1 Chapter Preface 3.2 Introduction 3.2.1 Inhibition of Return 3.2.2 What Processes are Inhibited? 3.2.4 Event-related Potential Investigations IOR 3.2.5 Present Study 3.3 Results 3.3.1 Behavioural Measures 3.3.2 LRP Measures 3.3.3 ERP Measures 3.4 Discussion 3.4.1 What processes are inhibited? 3.5 Methods 3.5.1 General Procedure 3.5.2 Experiment 1: Localization Task 3.5.3 Experiment 2: Detection Task  1 1 2 2 3 4 4 6 8 8 10 14 14 14 19 21 24 26 27 30 30 30 31 32 35 37 40 40 40 45 52 53 63 63 63 64  IV  3.5.4 Experiment 3: Go-NoGo Task 3.5.5 Electrophysiological Recording and Data Processing 3.5.6 Data Analysis 3.5.7 Subjects References  64 64 65 68 69  Chapter IV: Examining the Relationship Between Event-related Potential and Reaction Time Effects 73 4.1 Chapter Preface 4.2 Introduction 4.3 Method 4.4 Results and Discussion 4.5 Conclusions References CHAPTER V: Summary and Conclusions References  73 73 74 75 76 79 80 85  V  List of Tables Chapter 2: Table 2.1 Mean event-related potential amplitudes and lateralized readiness potential onset latencies  22  Chapter 3: Table 3.1 Mean reaction times in milliseconds as a function o f cue validity.  41 Table 3.2 Lateralized readiness potential (LRP) onset latencies in milliseconds.  41 Table 3.3 E R P average magnitudes (p,v, windows as described in text)  45  Table 3.4 Summary of Event-related potential effects found in studies investigating the effect o f non-predictive stimulus at long cue-target intervals. For the P I and N I effects 'Reduction' refers to a smaller peak on valid trials and 'Enhancement' refers to a larger peak on valid trials. IOR: valid minus invalid reaction time difference; S O A : stimulus onset asynchrony between cues and targets; N d : a negative difference observed between valid and invalid trials within the latency range o f 200-300 ms; Nde; an early N d effect observed within the latency range o f 100-200 ms; n.s.: measured effect was not significant; ?: effect was observed but was not subjected to a statistical test; *: the indicated effect overlaps in time and space with the Nde effect; N A : information about the effect is not available 55  List of Figures Chapter 2 Figure 2.1 Relationship between lateralized readiness potential ( L R P ) onsets and information processing stages: Target- and response-locked lateralized readiness potentials (LRPs) indicating time o f L R P onset (a) and model of target-related information processing between stimulus and response (b). See text for details 18 Figure 2.2 Electrophysiological data, (a) Target- and response-locked lateralized readiness potentials (LRPs) for valid and invalid trials, (b) Event-related potentials (ERPs) to valid and invalid targets at posterior electrode sites. Vertical bars indicate target onset for target-locked waves and time of response for response-locked L R P s . Nd=negative difference 23  Chapter 3 Figure 3.1 A n example o f the stimulus display and trial sequence used in all three experiments. A valid-cue trial is shown. S O A : stimulus onset asynchrony 39 Figure 3.2 Lateralized-readiness potentials (LRPs) from Experiments 1,2, and 3  42 Figure 3.3 Event-related potentials (ERPs) from Experiment 1  47  Figure 3.4 Event-related potentials (ERPs) from Experiment 2  49  Figure 3.5 Event-related potentials (ERPs) from Experiment 3  51  Figure 3.6 Graphical representation o f the 'real neural architecture" model (Shipp, 2004) o f visual spatial attention and inhibition o f return (IOR). The focus o f attention is jointly determined by stimulus salience and top-down control signals. I O R is implemented by a mechanism that reduces the salience o f stimuli at recently explored spatial locations. We expand upon this model by assuming that (1) allocation of attention to spatial location improves the signal-to-noise o f stimuli at this location relative to stimuli at unattended locations and (2) the input from the target stimulus to an evidence-accumulator decision mechanism depends on the signal-to-noise ratio. See text for details. This figure is an elaboration o f Figure 2(g) in Shipp (2004) ...62  Chapter 4 Figure 4.1 Electrophysiological data, (a) Target- and response-locked lateralized readiness potentials (LRPs) for valid and invalid trials, (b) Event-related potentials (ERPs) to valid and invalid targets at posterior electrode sites. Vertical bars indicate target onset for target-locked waves and time o f response for response-locked L R P s . Nd=negative difference 78  Vll  List of Nomenclature  P1: An early occipital positive peak (peak latency approximately 120 ms) in the event-related potential elicited by the onset of visual stimuli. N1: An early occipital-parietal negative peak (peak latency approximately 180 ms) in the event-related potential elicited by the onset of visual stimuli. N2: A later central negative peak (peak latency approximately 300 ms) in the eventrelated potential elicited by the onset of visual stimuli. Nd: A negative deflection in the event-related potential difference wave calculated by subtracting the event-related potentials elicited by invalidly cued targets from those elicited by validly cued targets.  vm  Acknowledgements The research on which this thesis is based was conducted under the guidance o f my supervisor Lawrence Ward. Lawrence, I can't thank you enough for your support over the years. Y o u are one o f a very small number of people who have encouraged me or given me an opportunity to pursue my goals and dreams. Thank you for allowing me to pursue my own interests. Thank you for respecting my abilities. Thank you for all your advice and guidance. Thank you for being a good friend.  I would also like to thank Vince D i Lollo, John McDonald, Matthew Tata, and Christian Richard. Y o u have all helped me out in too many ways to mention. Without your support and friendship my time in graduate school would have been much less rewarding.  Finally, I would like to acknowledge my family for a lifetime o f support. I would like to thank my mother, for her dedication to her children. M y father, who did not live long enough to see me complete my studies, for teaching me many things about the world. M y brother Steven, as good a friend as anyone could ask for.  1  CHAPTER I: Introduction and Overview 1.1 Introduction In everyday life the amount of information provided by our visual system far exceeds the brain's processing capacity. For this reason, the human brain has many selective mechanisms that isolate important information from the sensory input for detailed analysis. These mechanisms are usually grouped together under the general label o f attention (for a review see, Wright & Ward, 1998). Attentional mechanisms allow people both to search efficiently for and select specific information (goal-driven attention) and respond to unexpected but potentially important information (attentional capture). Attention can be directed to a spatial location either by overt changes in eye position that direct the high acuity fovea to the location o f interest or by covert orienting o f attention that allocates limitedcapacity mental processes to the selected location in the absence of changes in eye position. Attentional orienting can be elicited either voluntarily in response to information regarding the likely location o f relevant information, or involuntarily in response to salient stimulus changes (e.g., luminance changes). Regardless of how attention is oriented, the speed and accuracy of stimulus processing are increased at attended locations relative to unattended locations (Wright & Ward, 1998). Numerous lines o f evidence indicate that these changes in performance arise from changes in perceptual stages of stimulus processing. For example, studies utilizing signal detection methodology have found that spatial attention increases perceptual sensitivity (Hawkins et al., 1990; Luck et al., 1994) and electrophysiological studies have found that attention can affect visual processing in extrastriate visual areas as early as 80 msec after stimulus onset (Luck et al., 1994; Mangun, 1995).  2  1.2 Background: Spatial Orienting and Inhibition of Return 1.2.1 The Spatial-cueing Paradigm The spatial orienting o f attention has often been studied within the context o f the spatial-cueing paradigm (also known as the cue-target paradigm). In this paradigm, an initial cue stimulus is used to direct attention to a specific location prior to the presentation of a target stimulus to which a response is required. A cue may take the form o f a symbol, such as an arrow, presented at fixation, or a salient stimulus change at a peripheral location. Trials in which the target was presented at the cued location are referred to as valid-cue trials, whereas trials in which the target was presented at an uncued location are referred to as invalid-cue trials. When the probability o f a valid-cue trial is greater than that o f an invalid-cue trial, the cues are said to be informative. When valid-cue and invalid-cue trials are equally likely the cues are said to be uninformative. When cues are informative, attention orienting is assumed to be at least partly under voluntary control. When cues are uninformative, attention orienting is assumed to be involuntary. Facilitatory effects of cue validity have been found for both symbolic and stimulus cues, regardless of the predictiveness of the cue. The general result o f these experiments is that individuals respond to the target more quickly and accurately when it is presented at a validly cued location than when it is presented at an invalidly cued location. The facilitatory effects o f spatial cues have been demonstrated across a wide range o f tasks involving simple detection responses, identity-based choice responses, and localization responses, indicating that shifting spatial attention has general effects on visual performance (e.g., Jonides, 1981; Posner, 1980). However, several differences have been found between the effects o f the various cue types and it has not yet been established whether these differences reflect different ways in which the same attentional processes are brought to bear on the target stimuli at selected spatial locations (e.g., Jonides, 1981; Miiller & Rabbitt,  3 1989; Posner, 1980) or whether they reflect different attentional processes (e.g., Briand & K l e i n , 1987, see K l e i n & Shore, for a review).  1.2.2 Stimulus Cues and Inhibition of Return Peripherally presented stimulus cues have been demonstrated to be particularly effective in eliciting involuntary shifts of spatial attention. When the interval between cue and target is short (~<300 msec), stimulus cues can facilitate performance even when the participants are given explicit instructions to ignore them (e.g., Jonides, 1981). Moreover, the facilitatory effects o f stimulus cues are less susceptible to interference from concurrent memory tasks relative to the effects o f symbolic cues (Jonides, 1981). However, Posner and Cohen (1984) demonstrated that this early facilitation is followed by an inhibitory effect at longer cue-target intervals in which subjects respond more slowly on valid trials than on invalid trials. Posner, Rafal, Choate and Vaughan (1985) labeled this inhibitory effect Inhibition o f Return (IOR) and proposed that it serves to bias the visual system to acquire novel information at new locations. This provocative proposal remains a popular topic o f debate and has generated a substantial body o f research (for a review see, K l e i n , 2000). In their seminal study, Posner and Cohen found that IOR only occurred when attention was oriented involuntarily by a stimulus cue and not when attention was oriented voluntarily in response to a symbolic cue. Further research has revealed many other properties o f visual IOR. The inhibitory effect has been found to last for several seconds after cue onset (Tassinari & Berlucchi, 1995) and to affect simple detection responses (e.g., Posner & Cohen, 1984), localization responses (e.g., Maylor, 1985) and nonspatial discrimination responses (e.g., Pratt, 1995; Pratt, Kingstone & Khoe, 1997). Like attention, I O R has been shown to affect target detection accuracy (Handy, Jha, & Mangun, 1999) and nonspeeded target identification responses (Klein & Dick, 2002). In addition to its effect on manual  responses, IOR can affect the direction (Posner et al, 1985) and latency (Abrams & Dobkin, 1994) of saccadic eye movements. IOR is not an exclusively visual phenomenon. It has been observed in other spatial modalities, including hearing (e.g., McDonald & Ward, 1999; Reuter-Lorenz & Rosenquist, 1996; Schmidt, 1996a) and touch (e.g., Tassinari & Campara, 1996). In addition to these within-modality effects, IOR has also been found in cross-modal studies in which the cues and targets are presented in different sensory modalities (e.g., McDonald & Ward, 2001; Spence & Driver, 1998a). The ubiquitous nature of IOR across many tasks and sensory modalities indicates that the processes underlying IOR are important and general mechanisms in the spatial selection of information.  1.3 Proposed Mechanisms of IOR A variety of accounts have been proposed to explain how IOR arises. These accounts have been extensively reviewed and evaluated previously (e.g.; Reuter-Lorenz, Jha, & Rosenquist, 1996; Taylor & Klein, 1998; Taylor & Klein, 2000). For this reason, only a brief review will be provided here. 1.3.1 Attentional Accounts The most influential account of IOR is based on Posner et al.'s (1985) proposal that IOR arises from a mechanism that inhibits orienting towards previously attended locations. This account is based on a model of attention that likens spatial attention to a spotlight that enhances the efficiency of perceptual processing of stimuli within its beam (Posner, Snyder, & Davidson, 1980). First, attention is involuntarily oriented to the cued location resulting in faster responses on valid-cue trials relative to invalid-cue trials at short cue-target intervals. If a target does not occur after a short delay, typically less than 300 msec, attention is reoriented  5  to fixation and an inhibitory mechanism is activated that impairs covert attention from orienting to the previously attended location. At long cue-target intervals, therefore, IOR is obtained because of a relative perceptual deficit arising from attention being impaired from returning to the previously cued location relative to orienting to novel locations. Other attentional accounts of IOR have been also been proposed. For example, Cheal (1997) has put forward an account of attentional facilitation and IOR based on the operation of hypothetical filters that can either facilitate or inhibit the flow of information into the perceptual system. By contrast, Lupianez and colleagues (Lupianez & Milliken, 1999; Lupianez, Milliken, Solano, Weaver, & Tipper, 2001) have proposed that the facilitatory and inhibitory effects of uninformative cues can be accounted for in terms of opening, closing and integrating information into object files. Finally, Pratt, Spalek and Bradshaw (1999) have proposed that IOR does not arise from an inhibitory system that affects reorienting to the previously attended location but instead arises from "attentional momentum" that biases attention to continue moving in the direction it most recently traveled (away from the cued location). Despite their differences, two predictions can be derived from these attentional accounts of IOR. First, IOR should be sensitive to the same factors that influence attention orienting. Second, IOR and attentional facilitation should arise from changes in the same processes; however, the differences in target processing between valid and invalid trials should be in opposite directions in the two cases. Given the considerable evidence that spatial attention enhances perceptual processing, this account suggests that IOR should be associated with a relative inhibition of perceptual processing at the cued location relative to other locations. These two predictions have been subjected to several tests, although the results to date have been mixed (Taylor & Klein, 1998; Taylor & Klein, 2000). Consistent with an  6 attentional account of IOR, attention and I O R are similarly affected by target modality, target intensity, and response type (e.g.; Reuter-Lorenz et al., 1996). In addition, evidence that I O R arises from an inhibition of perceptual processing has come from experiments showing that I O R affects the accuracy of unspeeded target discrimination responses (Handy et al., 1999; K l e i n & Dick, 2002). Unlike attention, however, IOR apparently does not affect the speed o f transmission o f sensory information (e.g.; Maylor, 1985; Schmidt, 1996b; K l e i n , Schmidt, & Miiller, 1998). Although this result has sometimes been taken as evidence against attentional accounts o f IOR, it is important to note that attention is not a unitary phenomenon (e.g.; Pashler, 1998) and it is possible that some aspects of attention may be inhibited (i.e. perceptual enhancement) without affecting other aspects (i.e. transmission speed).  1.3.2 Response-related Accounts In contrast to attentional accounts o f IOR that propose that I O R arises from changes in perceptual processes, several authors have proposed that IOR arises from changes in response-related processes (e.g.; Tassinari, Aglioti, Chelazzi, M a r z i , Berlucchi, 1987; K l e i n & Taylor, 1994; Taylor & K l e i n , 1998; Godijn & Theeuwes, 2002). A response-related effect could conceivably arise at several stages of processing. I O R may arise from inhibition that directly affects motor processes such as motor programming or execution. Alternatively, I O R may arise from changes in decisional processes responsible for response selection and initiation. Numerous findings have implicated the oculomotor system in generating I O R (e.g.; Posner et al., 1985; Rafal, Calabresi, Brennen, & Sciolto, 1989; Sapir, Soroker, Berger, & Henik, 1999; Dorris, K l e i n , Everling, & Munoz, 2002; R o , Fame, & Chang, 2002). Furthermore, recent research has demonstrated that IOR can affect oculomotor programming as well as saccadic reaction time. R o , Pratt & Rafal (2000) found that when saccade  7 amplitude could not be pre-programmed IOR affected the amplitudes of saccades. Similarly, Godijin and Theeuwes (2002) found that the trajectories of saccades towards colour-defined targets deviated away from the location of uninformative stimulus cues. It has been proposed that inhibition generated in the oculomotor system may also give rise to IOR when manual responses are required by inhibiting manual motor processes (e.g.; Tassinari et al., 1987), possibly mediated through motor control areas in the posterior parietal cortex (Godijn & Theeuwes, 2002). Unlike when saccadic responses are required, however, IOR has not been found to inhibit response execution or to affect the movement path of manual responses (Howard, Lupianez, & Tipper, 1999; Fischer, Pratt, & Neggers, 2003). If present, any inhibition of manual motor processes may be limited to the duration of response programming. A decisional locus of IOR has also received support. Klein and Taylor (1994) proposed that IOR is the result of a "reluctance to respond" to an event at a previously attended location. Consistent with this account, Ivanoff and Klein (2001) found that in a go/nogo task IOR was associated with fewer false alarms to validly-cued nogo targets than to invalidly-cued nogo targets. This result suggests that IOR may arise, at least in part, from a more conservative response criterion on valid-cue trials than on invalid-cue trials. However, it is important to note that IOR is sometimes associated with increased errors for validly-cued targets (e.g.; Cheal & Chastain, 1999). Similarly, Handy et al. (1999) found that IOR was associated with smaller d' values for valid targets than for invalid targets in the absence of any change in beta. These results indicate that not every IOR effect can be attributed to changes in response criterion. An alternative decisional mechanism that could account for both the reaction time and accuracy results has been proposed by Taylor and Klein (1998). According to these authors, IOR may arise from a deficit in linking perceptual information  8 with the required motor response (visuomotor integration). However, it is difficult to distinguish this mechanism from a perceptual deficit resulting from an inhibition of attention orienting using exclusively behavioural measures.  1.3.3 Summary At the current time there is no consensus regarding either the mechanisms responsible for producing IOR or the processing stages affected by these mechanisms. As indicated above, there is evidence that IOR may arise from processing changes occurring at either perceptual or response-related stages of processing. In addition, the characteristics of IOR vary depending on experimental task and response type (e.g.; Taylor & Klein, 2000). These results have prompted several researchers to propose that IOR arises from multiple mechanisms (e.g.; Rafal et al., 1989; Abrams & Dobkin, 1994; Kingstone & Pratt, 1999; Taylor & Klein, 2000; Hunt & Kingstone, 2003). 1.4 Goals and Approach of the Present Project One reason for the current theoretical uncertainty about the mechanism(s) of IOR is that behavioral measures reflect contributions from multiple stages of processing, and it is therefore difficult to attribute a change in performance unambiguously to a specific stage. One technique that can contribute to the resolution of such level-of-processing questions is the recording and measuring of event-related potentials (ERPs). Event-related potentials are scalp-recorded electrical voltage changes that are time locked to stimulus or response events. The ERP signals are produced by the synchronous activity of large groups of neurons during information processing. If many parallel neurons are simultaneously active their postsynaptic potentials sum together and conduct through the brain and skull to the scalp surface. Unlike neuroimaging techniques that rely on measures of hemodynamic responses, ERPs provide a  measure o f brain activity with millisecond temporal accuracy. The temporally continuous nature o f E R P waveforms allows some aspects of the neural activity that occurs between targets and responses to be observed. The aim o f the present project is to use E R P measures of brain activity to help identify the stages o f information processing affected by IOR. The studies that follow examine the E R P correlates o f visual IOR in spatial-cueing experiments that require manual responses. The relative duration o f motor processes (motor planning and execution) on valid cue and invalid-cue trials w i l l be evaluated by determining the onset o f the motor-related lateralized readiness potential (LRP). The L R P is an electrophysiological measure o f brain activity generated in motor cortex that is related to the selection and preparation o f motor responses (Coles, 1989; Eimer, 1998). If IOR arises from inhibition o f motor processes then the interval between L R P onset the response should be longer on valid-cue trials than on invalid-cue trials. Conversely, i f IOR arises from inhibition o f processing occurring prior to the onset o f motor planning and execution, then the interval between target onset and L R P onset should be longer on valid-cue trials than on invalid-cue trials. Similarly, the effect o f IOR on perceptual processing w i l l be assessed by examining short-latency E R P components generated in extrastriate visual cortex. If IOR arises from changes in perceptual processes, the amplitudes o f the early visual components w i l l be reduced on valid-cue trials relative invalid-cue trials. More detailed discussions o f goals and methods w i l l be presented in the following chapters.  10 References Abrams, R. A., & Dobkin, R. S. (1994). Inhibition of return: Effects of Attentional Cuing on Eye Movement Latencies. Journal of Experimental Psychology, 20, 467-477. Briand, K. A., & Klein, R. M. (1987). Is Posner's "Beam" the Same as Treisman's "Glue"?: On the Relation Between Visual Orienting and Feature Integration Theory. Journal of Experimental Psychology, 13, 228-241. Cheal, M. (1997). Understanding diverse effects of visual attention with the VAP-filters metaphor. Consciousness and Cognition, 6, 348-362. Cheal, M., & Chastain, G. (1999). Inhibition of return: Support for generality of the phenomenon. Journal of General Psychology, 126, 375-390. Coles, M. G. H. (1989). Modern mind-brain reading: Psychophysiology, physiology and cognition. Psychophysiology, 26, 251-269. Dorris, M. C , Klein, R. M., Everling, S., & Munoz, D. P. (2002). Contribution of the primate superior colliculus to inhibition of return. Journal of Cognitive Neuroscience, 14, 1256-1263. Eimer, M. (1998). The lateralized readiness potential as an on-line measure of central response activation processes. Behavioral Research Methods. Instruments, & Computers. 30, 146-156. Fischer, M. H., Pratt, J., & Neggers, S. F. W. (2003). Inhibition of return in manual pointing movements. Perception & Psychophysics, 63, 379-387. Godijn, R., & Theeuwes, J. (2002). Oculomotor capture and inhibition of return: Evidence for an oculomotor suppression acoount of IOR. Psychological Research/Psychologishce Forschung, 66, 234-246. Handy, T. C , Jha, A. P., & Mangun, G. R. (1999). Promoting novelty in vision: Inhibition of return modulates perceptual-level processing. Psychological Science. 10. 157-161. Hawkins, H. L., Hillyard, S. A., Luck, S. J., Mouloua, M., Downing, C. J., Woodward, D. P. (1990). Visual Attention Modulates Signal Detectability. Journal of Experimental Psychology: Human Perception and Performance. 16, 802-811. Howard, L.A., Lupianez, J., & Tipper, S. P. (1999). Inhibition of Return in a selective reaching task: an investigation of reference frames. The Journal of General Psychology. 126. 421 -442.  11 Hunt, A. R., & Kingstone, A. (2003). Inhibition of return: Dissociating attentional and oculomotor components. Journal of Experimental Psychology: Human Perception and Performance. 29. 1068-1074. Ivanoff, J., & Klein, R. M. (2001). The presence of a nonresponding effector increases inhibition of return. Psychonomic Bulletin & Review. 8, 307-314. Jonides, J. (1981). Voluntary versus automatic control over the mind's eye's movement. In J.B. Long & A.D. Baddeley (Eds.), Attention & Performance IX (pp. 187-203). Hillsdale, NJ: Erlbaum. Kingstone, A., Pratt, J. (1999). Inhibition of return is composed of attentional and oculomotor processes. Perception & Psychophysics. 61, 1046-1054. Klein, R. (2000). Inhibition of return. Trends in Cognitive Sciences. 4(4). 138-147. Klein, R. M., & Dick, B. (2002). Temporal Dynamics of reflexive attention shifts: A dualstream rapid serial visual presentation exploration. Psychological Science. 13. 176179. Klein, R. M., Schmidt, W. C , & Muller, H. J. (1998). Disinhibition of return: Unnecessary and unlikely. Perception & Psychophysics, 60. 862-872. Klein, R. M., & Shore, D.I. (2000). Relations among modes of visual orienting. In S. Monsell & J. Driver (Eds) Attention & Performance XVIII: Control of cognitive processes. Cambridge: MIT Press: pp 195-208. Klein, R. M., & Taylor, T. L. (1994). Categories of cognitive inhibition, with reference to attention. In D. Dagenbach & T. H. Carr (Eds.), Inhibitory processes in attention, memory, and language (pp. 113-150). San Diego, CA: Academic Press. Luck, S. J., Hillyard, S. A., Mouloua, M, Woldorff, M. G., Clark, V. P., & Hawkins, H. L. (1994). Effects of spatial cuing on luminance detectability: Psychophysical and electrophysiological evidence for early selection. Journal of Experimental Psychology: Human Perception and Performance. 20. 887-904. Lupianez, J., & Milliken, B. (1999). Inhibition of return and the attentional set for integrating versus differentiaating information. The Journalf of General Psychology. 126,392-418. Lupianez, J., Milliken, B., Solano, C , Weaver, B., & Tipper, S. P. (2001). On the strategic modulation of the time course of facilitation and inhibition of return. The Quarterly Journal of Experimental Psychology, 54A. 753-773. Mangun, G. R. (1995). Neural mechanisms of visual selective attention. Psychophysiology, 32, 4-18. Maylor, E. A. (1985). Facilitatory and inhibitory components of orienting in visual space. In M. I. Posner & O. S. M. Marin (Eds.), Mechanisms of Attention: Attention and Performance XL Hillsdale, New Jersey: Lawrence Erlbaum Associates.  McDonald, J. J., & Ward, L. M. (1999). Spatial relevance determines facilitatory and inhibitory effects of auditory covert spatial orienting. Journal of Experimental Psychology: Human Perception and Performance. 25. 1234-1252.  12  Miiller, H. J., & Rabbitt, P. M. A. (1989). Reflexive and voluntary orienting of visual attention: Time course of activation and resistance to interruption. Journal of Experimental Psychology: Human Perception and Performance. 15(2). 315-330. Posner, M. I. (1980). Orienting of attention. Quarterly Journal of Experimental Psychology, 32, 3-25. Posner, M. I., & Cohen, Y. (1984). Components of visual orienting. In H. Bouma & D. G. Bouwhuis (Eds.), Attention and performance. X (pp. 531-556). Hillsdale, NJ: Erlbaum. Posner, M. I., Rafal, R. D., Choate, L. S., & Vaughan, J. (1985). Inhibition of return: Neural Basis and Function. Cognitive Neuropsychology, 2. 211-228. Pratt, J. (1995). Inhibition of return in a discrimination task. Psychonomic Bulletin and Review. 2, 117-120. Pratt, J., Kingstone, A., & Khoe, W. (1997). Inhibition of return in location- and identitybased choice decision tasks. Perception & Psychophysics, 59, 964-971. Pratt, J., Spalek, T. M., & Bradshaw, F. (1999). The time to detect targets at inhibited and noninhibited locations: preliminary evidence for attentional momentum. Journal of Experimental Psychology, 25, 730-746. Rafal, R. D., Calabresi, P. A., Brennen, C. W., & Sciolto, T. K. (1989). Saccade preparation inhibits reorienting to recently attended locations. Journal of Experimental Psychology: Human Perception and Performance, 15, 673-685. Reuter-Lorenz, P. A., Jha, A. P., & Rosenquist, J. N. (1996). What is inhibited in inhibition of return? Journal of Experimental Psychology: Human Perception and Performance, 22, 367-378. Reuter-Lorenz, P. A., & Rosenquist, J. N. (1996). Auditory cues and inhibition of return: The importance of oculomotor activation. Experimental Brain Research. Ro, T., Fame, A., & Chang, E. (2003). Inhibition of return and the human frontal eye fields. Experimental Brain Research. 150, 290-296. Ro, T., Pratt, J., & Rafal, R. D. (2000). Inhibition of return in saccadic eye movements. Experimental Brain Research, 130, 264-268. Sapir, A., Soroker, N., Berger, A., & Henik, A. (1999). Inhibition of return in spatial attention: Direct evidence for collicular generation. Nature Neuroscience, 2, 10531054. Schmidt, W. C. (1996a). 'Inhibition of return' without visual input. Neuropsychologia, 34. 943-952.  13 Schmidt, W. C. (1996b). Inhibition of return is not detected using illusionary line motion. Perception & Psychophysics, 58, 883-898. Spence, C. J., & Driver, J. (1998). Auditory and audiovisual inhibition of return. Perception & Psychophysics. 60, 125-139. Tassinari, G., Aglioti, S., Chelazzi, L., Marzi, C. A., & Berlucchi, G. (1987). Distribution in the visual field of the costs of voluntarily associated attention and of the inhibitory after-effects of covert orienting. Neuropsychologia, 25. 55-71. Tassinari, G., & Campara, D. (1996). Consequences of covert orienting to non-informative stimuli of different modalities: A unitary mechanism? Neuropsychologia. 34, 235245. Tassinari, G., & Berlucchi, G. (1995). Covert orienting to non-informative cues: Reaction time studies. Behavioural Brain Research. 71. 101-112. Taylor, T. L., & Klein, R. M. (1998). On the causes and effects of inhibition of return. Psychonomic Bulletin & Review, 5, 625-643. Taylor, T. L., & Klein, R. M. (2000). Visual and motor effects in inhibition of return. Journal of Experimental Psychology: Human Perception and Performance, 26, 1639-1656. Wright, R. D., & Ward, L. M. (1998). The control of visual attention. In R. D. Wright (Ed.), Visual Attention (pp. 132-186). New York, NY: Oxford University Press.  14  CHAPTER II: Event-related Potential Correlates of Inhibition of Return in a Discrimination Task 2.1 Chapter Preface This chapter presents the results of my initial study of the ERP correlates of inhibition of return (IOR). A version of this chapter has been published in Psychological Science under the title "Inhibition of Return from Stimulus to Response." * This is the first published study to utilize measures of lateralized readiness potential (LRP) onsets to investigate IOR. A two alternative forced choice discrimination task was chosen because this type of task produces large LRP waves with clear onsets.  2.2 Introduction In a typical visual spatial-cueing experiment with uninformative peripheral cues, reaction time to cued-location targets is facilitated (relative to that for targets at uncued locations) at short cue-target stimulus-onset-asynchronies (SOAs) and inhibited at longer SOAs. This latter inhibitory effect is called inhibition-of-return (IOR). Since it was first reported by Posner and Cohen (1984), IOR has attracted a great deal of research interest (e.g., Klein, 2000). This interest is primarily motivated by the belief that IOR is closely associated with both covert and overt orienting of attention. Posner, Rafal, Choate, and Vaughan (1985) proposed that IOR serves to bias the visual system to acquire novel information at new locations. Expanding on this proposal, Klein (1988) suggested that IOR serves to facilitate visual search by biasing orienting responses away from recently inspected locations. This provocative proposal remains a popular topic of debate and has inspired numerous empirical studies (e.g., Klein, 2000). Given that compelling evidence of inhibitory * Prime, D. J., & Ward, L. M. (2004). Inhibition of returnfromstimulus to response. Psychological Science. 15, 272-276.  effects in visual search has been obtained (Klein & Machines, 1999; Miiller & von  15  Muhlenen, 2000; Takeda & Yagi, 2000), an understanding of the mechanisms underlying IOR may provide valuable insights into how visual information is selected. Despite a significant accumulation of empirical knowledge, no consensus has yet been reached regarding the stages of information processing affected by IOR. Explanations of IOR can be broadly classified into two groups, those that involve perception and attention and those that involve response-selection and motor processes (e.g., Taylor & Klein, 1998). These types of explanations are not mutually exclusive, and IOR may arise from multiple mechanisms, or from a single mechanism that affects multiple stages of processing. Posner et al. (1985) proposed that IOR arises from attention being inhibited from returning to previously-attended locations. Given the considerable evidence that spatial attention enhances perceptual processing (e.g., Wright & Ward, 1998), any mechanism that inhibits attention from being oriented to a spatial location should result in a relative inhibition of perceptual processing at that location. It has also been suggested that IOR may inhibit perceptual processing independently of attention (e.g., Posner & Cohen, 1984; ReuterLorenz, Jha, & Rosenquist, 1996). Evidence that IOR arises from an inhibition of perceptual processes has come from experiments showing that IOR affects the accuracy of unspeeded target-discrimination responses (Handy, Jha, & Mangun, 1999; Klein & Dick, 2002). In addition, Reuter-Lorenz et al. (1996) demonstrated that the magnitude of IOR and attentional facilitation were similarly affected by changes in target intensity and modality, providing support for the attentional orienting explanation of IOR. Other researchers have suggested that IOR may be associated with changes in response-related processes. Klein and Taylor (1994) proposed that IOR may arise from a reluctance to respond to events at the cued location and that the inhibition may arise in a  16 spatial motor map that directs action. Alternatively, Tassinari, Aglioti, Chelazzi, Marzi, and Berlucchi, (1987) proposed that IOR arises from inhibition in the motor system that results from suppression of overt orienting towards the cue. A response-related effect could conceivably arise at either decisional or motor stages of processing. Response-level explanations have received support from evidence that IOR is associated with a more conservative response criterion on valid trials than on invalid trials (Ivanoff & Klein, 2001) and that IOR can affect oculomotor programming (Ro, Pratt & Rafal, 2000). Given that the currently available evidence provides support for both perceptual and response-related mechanisms, it has been proposed that IOR may arise from multiple mechanisms (Kingstone & Pratt, 1999; Taylor & Klein, 2000). One technique that has proven to be especially useful in investigating the information-processing stages generating experimental effects is the recording of event-related brain potentials (ERPs) of the electroencephalogram (EEG). Relatively few studies have examined the ERP consequences of IOR, however, and the results have, so far, been somewhat inconsistent. McDonald, Ward and Kiehl (1999) found that IOR was associated with a reduction in the amplitude of the visual PI component (the first positive peak) and a negative difference (Nd) between ERPs on valid-cue and invalidcue trials in the P2 latency range. These results indicate that IOR is associated with a suppression of perceptual processing in extrastriate visual areas. However, two other studies challenged this conclusion. Hopfinger and Mangun (1998) found a significant PI reduction accompanied by a non-significant 2-ms IOR effect. Conversely, Hopfinger and Mangun (2001) found a significant 13-ms IOR effect but a non-significant PI reduction. Although both of these studies found behavioral and PI amplitude effects in the same direction as McDonald et al. (1999), Hopfinger and Mangun (2001) concluded that IOR was dissociable from the PI reduction. Because the available data seem to allow no firm conclusions about  17 the ERP correlates of IOR that would clarify its mechanism, additional study of these correlates seems warranted. In the present study, we investigated the electrophysiological correlates of IOR in a visual form discrimination task. Unlike previous studies, we examined the effect of uninformative visual cues both on the amplitude of early sensory ERP components and on the motor-related lateralized readiness potential (LRP). The LRP is an electrophysiological measure of brain activity generated in motor cortex that is related to the selection and preparation of motor responses (see Coles, 1989 and Figure 1). The latency of LRP onset is dependent on response selection so that the interval between target onset and the onset of the target-locked LRP provides a relative measure of the duration of processes involved in stimulus evaluation and response selection. Similarly, the interval between the onset of the response-locked LRP and the response provides a relative measure of the duration of motor processes (e.g., motor planning and execution). If IOR arises from inhibition of motor processes then the interval between the onset of the response-locked LRP and the response should be longer on valid-cue trials than on invalid-cue trials. Conversely, if IOR arises from inhibition of processing occurring prior to the onset of motor planning and execution, then the target-locked LRP should begin sooner on invalid-cue trials. The presence of both effects would implicate both perceptual/attentional and motor mechanisms of IOR. In addition, if a pre-motor effect of IOR due to an inhibition of perceptual processing is present, IOR should also be accompanied by a reduction in the amplitude of the short-latency ERP components at occipital sites.  a.  18  Target-locked LRP  Target onset  Response-locked LRP  t LRP onset  L }P onset  Respc nse  b. Perception ... Response Selection ... Response Programming ... Response Execution ... Target < Target-locked LRP onset latency  LRP  •  Response t  Response-locked LRP onset latency  •  Figure 1.1. Relationship between lateralized readiness potential ( L R P ) onsets and information processing stages: Target- and response-locked lateralized readiness potentials (LRPs) indicating time o f L R P onset (a) and model of target-related information processing between stimulus and response (b). See text for details.  19  2.3 Method Participants viewed a computer monitor from a distance o f 42 cm and were instructed to maintain fixation on a centrally-located fixation cross during the experimental blocks. The screen background was black and, two grey square outline boxes (1.5°x 1.5°) were centered 5° above and below fixation. After a 700-ms inter-trial interval, each trial began with a 133ms offset o f the fixation cross, followed 800 ms later by the cue. The cue consisted o f a 200ms brightening o f one o f the two square boxes and was equally likely to occur at either location. T w o hundred fifty ms after cue offset, a white circle 0.75° in diameter was presented at fixation for 160 ms. This re-orienting event was intended to redirect the participants' attention back to fixation from the cued location. After a variable delay o f 290 to 590 ms, the target was presented for 1000 ms. The target, either a "+" or an "x" (.75 °x .75°), was presented with equal probability within one o f the two peripheral boxes (chance coincidence o f cue and target locations). Participants pressed the "z" key on a standard keyboard i f the "x" was presented and the "/" key i f the "+" was presented. Both speed and accuracy were stressed in the instructions. The total cue-target S O A ranged between 900 and 1200 ms. This entire interval is well within the range o f S O A s that produce a reliable I O R effect and also ensures that overlap of high-frequency components of E R P s to cues, reorienting events, and targets is minimized. Trials on which cue and target occurred at the same location were classified as valid-cue trials, whereas trials on which cue and target occurred at the opposite locations were classified as invalid-cue trials. Twenty participants performed one practice and eight experimental blocks of 72 trials. Two participants were excluded from the analysis due to excessive eye movement artifacts.  20 Scalp potentials were recorded from the following scalp sites: C I , C 2 , C 3 , C 4 , Pz, P 0 7 , P 0 8 . These electrodes were referenced to the right mastoid and subsequently rereferenced to averaged mastoids. Eye position was monitored by both the horizontal and vertical electrooculogram (EOG). The E E G and E O G were sampled at 250 H z . Trials containing eye movement, muscle, and blocking artifacts were removed prior to averaging by applying automated artifact detection routines. In addition, trials with errors and those with reaction times outside o f 100-1000 ms were excluded from the analysis. E R P s were calculated separately for valid-cue and invalid-cue trials at electrodes sites Pz, P 0 7 and P 0 8 . Valid-cue and invalid-cue L R P s were calculated at C 1 - C 2 and C 3 - C 4 using the averaging method (Coles, 1989). To reduce the smearing o f the L R P slope due to time jitter, trials with reaction times greater than 1.5 standard deviations from the mean in each condition were excluded from the L R P averages. After averaging, the E R P s and L R P s were low-pass filtered (12-Hz and 4-Hz, respectively) to eliminate high-frequency artifacts in the waveforms. Target- and response-locked L R P onsets were determined using the one-degree-offreedom ( I D F ) regression technique (Mordkoff and Gianaros, 2000)'. Differences in L R P onsets were statistically assessed with a jackknifing procedure (Miller, 1974; M i l l e r , Patterson, and Ulrich, 1998) . For the E R P data, amplitude differences were analyzed by 2  separate repeated measures M A N O V A s for each effect. The mean amplitudes corresponding to the latency o f the positive P I (100-140 ms) and negative N I (156-196 ms) components were analyzed at sites P 0 7 and P 0 8 . These 40 ms latency windows are centered on the peaks of the P I and N I components in the grand-average waveforms. The mean amplitude o f the N d was measured in a latency range (240-280 ms) corresponding to the component's maximal amplitude and analyzed separately at parietal (Pz) and occipital sites ( P 0 7 , P 0 8 ) .  21  2.4 Results and Discussion A s expected, participants were significantly faster in responding to targets on invalidcue trials (564 ms) than on valid-cue trials (585 ms; F(l,18)=29.8,/?<0.0001), a typical I O R effect. There was no significant difference between the error rates on valid (4.2%) and invalid trials (5.4%). The target-locked L R P began at a shorter latency for invalid-cue than for valid-cue trials at both the C 1 - C 2 (t(\8)=2.1,^=0.05)  and C3-C4 (/(18)=2.2 p<0.05) 5j  electrode pairs (Figure 2a, Table 1). B y contrast, no onset differences were found between valid-cue and invalid-cue response-locked L R P s at either electrode pair (both ^-values<l; Figure 2a, Table 1). These results strongly indicate that I O R in this standard paradigm arises from delays in perceptual and decisional processes that occur before L R P onset, and not from later-occurring motor processes. The E R P results were consistent with a perceptual locus o f IOR. The amplitudes o f the occipital P I and N I components were both reduced on valid-cue relative to invalid-cue trials, F(l,18)=4.8;^<0.05, and F(l,18)=4.8;/?<0.05, respectively (Figure 2b). Dipole modeling techniques have localized both the occipital PI and N I components to areas o f extrastriate cortex ( D i Russo, Martinez, Sereno, Pitzalis & Hillyard, 2001). These effects indicate that I O R is associated with an inhibition o f early sensory processing o f validly-cued targets relative to invalidly-cued targets, an effect opposite that of spatial attention (e.g., Mangun & Hillyard, 1991). In addition, an N d was present at both occipital (F(l,18)=6.1, ^<0.03) and parietal sites (F(l,18)=4.2,/?<0.06; Figure 2b, Table 1). This effect is similar to the N d found by M c D o n a l d et al. (1999) and may reflect either perceptual or decision processes.  22  Table 1. Mean event-related potential amplitudes and lateralized readiness potential onset latencies. Component and Electrode  Cue Condition Valid  Invalid  Cue Effect (Valid-Invalid)  Event-related Potential Amplitude Measures in uVolts. PI  P07, P08  0.80  1.23  -0.43*  Nl  P07, P08  -1.75  -2.3  0.55*  Nd  P07, P08  1.00  1.63  -0.63*  Pz 4.25 4.97 -0.72t Lateralized Readiness Potential Latency Measures in milliseconds (relative to time of stimulus or response). LRP  LRP  Stimulus-locked C1-C2  264  216  48*  C3-C4  256  212  44*  C1-C2  -260  -272  12  C3-C4  -252  -252  0  Response-locked  *p<.05,tp<.06  a.  23  Target-locked LRP C1-C2  C3-C4 \—I—I  k  H  1 1  Response-locked LRP C3-C4  /  \ C1-C2  Valid  r -° ^ 2  Invnlirl " I n v a  a  I—I— — — —I—I +0 600 1  1  1  Figure 1.2. Electrophysiological data, (a) Target- and response-locked lateralized readiness potentials (LRPs) for valid and invalid trials, (b) Event-related potentials (ERPs) to valid and invalid targets at posterior electrode sites. Vertical bars indicate target onset for target-locked waves and time of response for response-locked LRPs. Nd=negative difference.  24  2.5 Conclusion The results presented here clearly show that IOR is associated with a delay in premotor processes. The target-locked L R P began at a longer latency on valid-cue than on invalid-cue trials. This effect is consistent with either a suppression o f perceptual processing or a delay in response selection. The onset o f the response-locked L R P was not affected, suggesting that, when measured with manual keypresses, I O R does not arise from inhibition o f motor processes. Although it is not possible to rule out an effect o f I O R on motor processes on the basis o f null results, these results do indicate that premotor effects were primarily responsible for the observed differences in reaction time in this experiment. However, these results do not exclude the possibility that an inhibition o f motor processes may contribute to the I O R effect when oculomotor and, perhaps, spatially-directed pointing or reaching movements are used. Although a change in the latency of the target-locked L R P does not distinguish between perceptual and decisional effects, the modulations o f sensory E R P components indicate that I O R arises at least in part from changes in sensory processes. The amplitudes o f the occipital P I and N l components were smaller on valid-cues trials than on invalid-cue trials. These results are consistent with the results o f attention orienting studies that have found that enhanced P I and N l amplitudes are accompanied by reaction time facilitation (e.g., Mangun, 1995). This parallel between the effects of attention orienting and I O R provides support for the-inhibition-of-attention explanation of IOR. Unlike the current study, M c D o n a l d et al. (1999) found that IOR in a simple detection task was associated only with a reduction o f the P I component. Although there are several differences between these experiments, it is possible that this difference in results arises from the nature o f the response tasks. Previous research has shown that the visual N l is associated with discrimination  25 processes (Vogel & Luck, 2000) and that attentional enhancement o f the N l is typically observed only for discrimination tasks (Mangun & Hillyard, 1991). Thus, our observation o f a reduction in N l magnitude in the presence of IOR is consistent with an IOR-related reversal o f the effects of attentional enhancement on the N l in our discrimination task. Taken together, the L R P and E R P effects observed in this experiment indicate that I O R must arise at least in part from changes in perceptual processes and that, at least when measured with manual keypresses, I O R does not arise from inhibition of motor processes.  26  Footnotes 1. Regression-based techniques define the onset o f the L R P as the intersection o f two straight lines that are fitted to segments o f the L R P wave-form (Schwarzenau, Faulkenstein, Hoormann and Hohnsbein, 1998). The I D F technique sets the pre-onset line to have a height and slope o f zero. The terminus of the post-onset line is locked to the peak o f the L R P so that only the time intersection can vary. The post-onset line is then found using the least-squares technique to find the best fit to the L R P . 2. The jackknifing procedure involves determining the L R P onset latency for N different subsample grand-averages, with each participant omitted from one o f the subsamples. The values o f these onset latencies across subsamples are then used to calculate a standard error for pairwise comparisons between conditions.  27  References Coles, M . G . H . (1989). Modern mind-brain reading: Psychophysiology, physiology and cognition. Psychophysiology, 26, 251-269. D i Russo, F., Martinez, A . , Sereno, M . I., Pitzalis, S., & Hillyard, S. A . (2001). Cortical sources o f the early components of the visual evoked potential. Human Brain Mapping, 15, 95-111. Handy, T. C , Jha, A . P., & Mangun, G . R. (1999). Promoting novelty in vision: Inhibition o f return modulates perceptual-level processing. Psychological Science, 10, 157-161. Hopfinger, J. B . , & Mangun, G . R. (1998). Reflexive attention modulates processing o f visual stimuli in human extrastriate cortex. Psychological Science, 9, 441-447. Hopfinger, J. B . , & Mangun, G . R. (2001). Tracking the influence o f reflexive attention on sensory and cognitive processing. Cognitive, Affective & Behavioral Neuroscience, L 56-65. Ivanoff, J., & K l e i n , R. M . (2001). The presence o f a nonresponding effector increases inhibition o f return. Psychonomic Bulletin & Review, 8, 307-314. Kingstone, A . , Pratt, J. (1999). Inhibition of return is composed of attentional and oculomotor processes. Perception & Psychophysics, 61, 1046-1054. K l e i n , R. M . (2000). Inhibition of return. Trends in Cognitive Sciences, 4, 138-147. _  K l e i n , R. M . (1988). Inhibitory tagging system facilitates visual search. Nature, 334(4), 430431. K l e i n , R. M . & Taylor, T. L . (1994). Categories of cognitive inhibition with reference to attention. In D . Dagenbach & T. H . Carr (Eds.), Inhibitory processes in attention, memory, and language, (pp. 113-150). San Diego, C A : Academic Press. K l e i n , R. M . , & Machines, W . J. (1999). Inhibition o f return is a foraging facilitator in visual search. Psychological Science, 10, 346-352. K l e i n , R. M . , & Dick, B . (2002). Temporal Dynamics o f reflexive attention shifts: A dualstream rapid serial visual presentation exploration. Psychological Science, 13, 176179. Mangun, G . R., & Hillyard, S. A . (1991).Modulations o f sensory-evoked brain potentials indicate changes in perceptual processing during visual spatial priming. Journal o f Experimental Psychology: Human Perception and Performance, 17, 1057-1074. Mangun, G . R. (1995). Neural mechanisms o f visual selective attention. Psychophysiology, 32, 4-18.  28 M c D o n a l d , J. J., Ward, L . M . , & K i e h l , K . A . (1999). A n event-related brain potential study of inhibition o f return. Perception & Psychophysics, 61, 1411-1423. Miller, R. G . (1974). The jackknife-a review. Biometrika, 61, 1-15. Miller, J., Patterson, T., & Ulrich, R. (1998). Jackknife-based method for measuring L R P onset latency differences. Psychophysiology, 35, 99-115. Mordkoff, J. T., & Gianaros, P. J. (2000). Detecting the onset o f the lateralized readiness potential:A comparison o f available methods and procedures. Psychophysiology, 37, 347-360. M M e r , H . J., & von Miihlenen, A . (2000). Probing distractor inhibition in visual search: Inhibition o f return. Journal o f Experimental Psychology: Human Perception and Performance, 26, 1591-1605. Posner, M . I., & Cohen, Y . (1984). Components of visual orienting. In H . Bouma & D . G . Bouwhuis (Eds.), Attention and performance, X (pp. 531-556). Hillsdale, N J : Erlbaum. Posner, M . I., Rafal, R. D . , Choate, L . S., & Vaughan, J. (1985). Inhibition o f return: Neural Basis and Function. Cognitive Neuropsychology, 2, 211-228. Reuter-Lorenz, P. A . , Jha, A . P., & Rosenquist, J. N . (1996). What is inhibited in inhibition of return? Journal of Experimental Psychology: Human Perception and Performance, 22, 367-378. R o , T., Pratt, J., & Rafal, R. D . (2000). Inhibition o f return in saccadic eye movements. Experimental Brain Research, 130, 264-268. Schwarzenau, P., Faulkenstein, M . , Hoormann, J., & Hohnsbein, J. (1998). A new method for the estimation of the onset of the lateralized readiness potential ( L R P ) . Behavior Research Methods, Instruments, & Computers, 30, 100-117. Takeda, Y . , & Y a g i , A . (2000). Inhibitory tagging in visual search can be found i f search stimuli remain visible. Perception & Psychophysics, 62, 927-934. Tassinari, G . , Aglioti, S., Chelazzi, L . , M a r z i , C . A . , & Berlucchi, G . (1987). Distribution in the visual field of the costs o f voluntarily allocated attention and the inhibitory aftereffects o f covert orienting. Neuropsychologia, 25, 55-71. Taylor, T. L . , & K l e i n , R. M . (1998). O n the causes and effects o f inhibition o f return. Psychonomic Bulletin & Review, 5(4), 625-643. Taylor, T. L . , & K l e i n , R. M . (2000). Visual and motor effects in inhibition o f return. Journal of Experimental Psychology: Human Perception and Performance, 26, 1639-1656.  29 Vogel, E. K., & Luck, S. J. (2000). The visual NI component as an index of a discrimination process. Psychophysiology, 37, 190-203. Wright, R. D., & Ward, L. M. (1998). The control of visual attention. In R. D. Wright (Ed.), Visual Attention (pp. 132-186). New York, NY: Oxford University Press.  30  CHAPTER III: Event-related Potential Correlates of Inhibition of Return Across Tasks 3.1 Chapter Preface The results presented in Chapter 1 indicate that I O R arises from a delay in pre-motor processes. It can be argued, however, that this study does not provide a definitive test o f motor-based accounts o f IOR. Although the results presented in Chapter 1 demonstrate that IOR can occur in the absence o f an inhibition of motor processes, it is possible that motor inhibition may contribute to I O R in some circumstances. In order to test this possibility, three further experiments were conducted. The rationale for these experiments is provided i n the introduction. A version o f this chapter has been submitted to the Journal o f Cognitive Neuroscience. The chapter is organized in the format required by this journal.  3.2 Introduction In natural environments people are continuously confronted with a wide range o f sensory events. Our ability to processes this vast flow o f information, however, is limited. For this reason, the human brain has many selective mechanisms that isolate important information from the sensory input for detailed analysis. These mechanisms are usually grouped together under the general label o f attention (for a review see, Wright & Ward, 1998). In vision, attention can be directed to spatial locations either by overt orienting o f the eyes so as to foveate objects o f interest or by covert shifts o f attention that allocate limitedcapacity mental resources to non-foveated stimuli. Regardless o f how attention is oriented, the speed and accuracy o f stimulus processing are increased at attended locations relative to unattended locations (Wright & Ward, 1998). In order to function in dynamic environments, people must be able to search efficiently for specific information and respond to unexpected but potentially important  31 events. Consequently, attention can be oriented to spatial locations either voluntarily in accordance with an observer's goals or involuntarily in response to salient stimulus events. Regardless o f how attention is oriented, the sampling o f the visual environment would be more efficient i f attention was preferentially directed towards novel locations. It has been suggested that the mechanisms responsible for the inhibition of return (IOR) effect serve such a novelty seeking function (Posner, Rafal, Choate, and Vaughan, 1985; K l e i n , 1988, 2000).  3.2.1 Inhibition of Return "Inhibition o f return" is the label given to the empirical finding that observers are often slower to respond to stimuli that appear at spatial locations where other stimuli have recently appeared than to stimuli at locations that have not been recently stimulated. This effect was first discovered by Posner and Cohen (1984) in the context o f the cue-target paradigm introduced to study attention orienting. In a typical experiment o f this type, target stimuli are preceded by spatially non-predictive peripheral cues (e.g., luminance transients) and reaction times to cued-location targets (here called validly-cued targets) are facilitated relative to those for targets at uncued locations (here called invalidly-cued targets) at short cue-target stimulus-onset-asynchronies (SOAs) and inhibited at longer S O A s . It is the latter inhibitory effect that has been labeled IOR. Subsequent research has demonstrated that I O R occurs in a wide variety of experimental situations and affects both manual and oculomotor responses (for reviews see K l e i n , 2000; Taylor & K l e i n , 1998). Inhibition o f return has attracted a great deal o f research interest since its discovery. This interest has primarily been motivated by the belief that an understanding o f IOR w i l l illuminate basic mechanisms o f spatial attention. Posner et al. (1985) proposed that I O R serves to bias the visual system to acquire novel information at new locations. Expanding on  32 this proposal, K l e i n (1988) suggested that I O R serves to facilitate visual search by biasing orienting responses away from recently inspected locations. According to this account, attention is first involuntarily oriented to the cued location and, i f a target does not occur after a short delay, attention is then reoriented to fixation and an inhibitory mechanism is activated that impairs covert attention and eye movements from orienting to the previously inspected location. This proposal has recently received support from evidence that inhibitory after-effects do occur in visual search tasks (Klein & Machines, 1999; Miiller & von Muhlenen, 2000; Takeda & Y a g i , 2000; Machines & K l e i n , 2003).  3.2.2 What Processes are Inhibited? M u c h o f the research investigating I O R has focused on determining what processing changes give rise to the reaction time effect. According to the inhibition-of-attention account, IOR is observed in cue-target experiments because of a relative perceptual deficit arising from attention being inhibited from re-orienting to the previously-cued location relative to orienting to a novel location. Despite a significant accumulation o f empirical knowledge, however, no consensus has yet been reached regarding the stages o f information processing affected by IOR. In addition to the inhibition-of-attention account, several other possible mechanisms have been proposed (Taylor & K l e i n , 1998) and virtually every stage o f processing leading to the eventual behavioural response has been implicated. It has been suggested that I O R may arise from an inhibition o f basic sensory processes (Posner & Cohen, 1984), an inhibition of perceptual processes unrelated to attention orienting (ReuterLorenz, Jha, & Rosenquist, 1996), an inhibition of response selection processes (Ivanoff & Klein, 2001; Taylor & K l e i n , 2000), a disconnection between stimulus and response selection processing (Fuentes, Vivas, & Humphreys, 1999) or an inhibition o f motor processes  33 (Tassinari, Aglioti, Chelazzi, M a r z i , Berlucchi, 1987; K l e i n & Taylor, 1994; Taylor & K l e i n , 1998; Godijn & Theeuwes, 2002). Consistent with an attentional account o f IOR, attention and I O R are similarly affected by target modality, target intensity, and response type (e.g.; Reuter-Lorenz et al., 1996). In addition, evidence that I O R arises from an inhibition o f perceptual processing has come from experiments showing that I O R affects the accuracy of unspeeded target discrimination responses (Handy et al., 1999; K l e i n & Dick, 2002). Unlike attention, however, I O R apparently does not affect the speed of transmission of sensory information (e.g.; Maylor, 1985; Schmidt, 1996; K l e i n , Schmidt, & Muller, 1998). Although this result has sometimes been taken as evidence against attentional accounts o f IOR, it is important to note that attention is not a unitary phenomenon (e.g.; Pashler, 1998) and it is possible that some consequences o f attention may be inhibited (i.e. perceptual enhancement) without affecting other consequences (i.e. transmission speed). A decisional locus o f IOR has also received support. K l e i n and Taylor (1994) proposed that I O R is the result of a "reluctance to respond" to an event at a previouslyattended location. Consistent with this notion, Ivanoff and K l e i n (2001) found that in a go/nogo task I O R was associated with fewer false alarms to validly-cued nogo targets than to invalidly-cued nogo targets. This result suggests that I O R may arise, at least in part, from a more conservative response criterion on valid-cue trials than on invalid-cue trials. However, it is important to note that I O R is sometimes associated with increased errors for validly-cued targets (e.g.; Cheal & Chastain, 1999). Similarly, Handy et al. (1999) found that I O R was associated with smaller d' values for validly-cued targets than for invalidly-cued targets in the absence o f any change in / J (the criterion location). These results indicate that I O R does not arise solely from changes in response criterion. A n alternative decisional mechanism that  could account for both the reaction time and accuracy results has been proposed by Taylor and Klein (1998). According to these authors, IOR could arise from inhibition of the visuomotor integration process linking perceptual information with the required motor response. Using exclusively behavioural measures, however, it is difficult to distinguish this mechanism from a perceptual deficit resulting from an inhibition of attention re-orienting. Finally, it has been proposed that IOR arises from inhibition of motor processes such as motor programming or execution. Numerous findings have implicated the oculomotor system in generating IOR (e.g.; Posner et al., 1985; Rafal, Calabresi, Brennen, & Sciolto, 1989; Sapir, Soroker, Berger, & Henik, 1999; Dorris, Klein, Everling, & Munoz, 2002; Ro, Fame, & Chang, 2002). Furthermore, recent research has demonstrated that IOR can affect oculomotor programming as well as saccadic reaction time. Ro, Pratt & Rafal (2000) found that, when saccade amplitude could not be pre-programmed, IOR affected the amplitudes of saccades. It has been proposed that inhibition generated in the oculomotor system may also give rise to IOR, when manual responses are required, by inhibiting manual motor processes (e.g.; Tassinari, et al., 1987; Klein & Taylor, 1994; Taylor & Klein, 1998), possibly mediated through motor control areas in the posterior parietal cortex (Godijn & Theeuwes, 2002). Alternatively, it has been proposed that a portion of the IOR effect on reaction time observed in speeded-response cue-target tasks arises from observers inhibiting an automatic tendency to respond to the cue stimuli (Spence & Driver, 1998; Poliakoff, Spence, O'Boyle, McGlone & Cody, 2002; Coward, Poliakoff, O'Boyle & Lowe, 2004). This proposal is based on Harvey's (1980) suggestion that inhibiting a response to stimulus will slow responding to a following stimulus, and that the amount of slowing is greater when the two stimuli share common features. Consequently, responses to targets appearing at the cued location will be inhibited more than responses to targets appearing at uncued locations because cues and  35 targets presented at cued locations have location in common. Because IOR is also present when observers respond to a sequence of targets (target-target paradigm), this mechanism cannot be solely responsible for generating IOR. However, IOR magnitude has been found to be significantly greater in cue-target experiments than in equivalent target-target experiments in vision (Coward et al., 2004), audition (Tassinari, Campara, Benedetti, & Berlucchi, 2002), and touch (Poliakoff et al., 2002). These results suggest that response inhibition to the cue may contribute to the IOR in cue-target experiments. At present, however, direct evidence for inhibition of motor processes has not been obtained. Unlike when saccadic responses are required, IOR has not been found to inhibit response execution time or to affect the movement path of manual responses (Howard, Lupianez, & Tipper, 1999; Fischer, Pratt, & Neggers, 2003). If present, any inhibition of manual motor processes may be limited to the duration of response programming.  3.2.4 Event-related Potential Investigations IOR One technique that has proven to be especially useful in investigating the information-processing stages generating experimental effects is the recording of eventrelated brain potentials (ERPs) of the electroencephalogram (EEG). ERPs provide a continuous record of cortical activity time-locked to stimulus or response events and can allow researchers to observe the effects of experimental manipulations on neural activity associated with either perceptual or motor processes. Relatively few studies have examined the effect of spatially non-predictive peripheral cues on target-elicited ERPs at long cuetarget SOAs (Eimer, 1994; Hopfinger & Mangun, 1998; Hopfinger & Mangun, 2001; McDonald, Ward, Kiehl, 1999; Prime & Ward, 2004; Wascher & Tipper, in press). Furthermore, two of these studies failed to obtain behavioural IOR (Eimer, 1994; Hopfinger & Mangun, 1998). The results of these studies are somewhat inconsistent and they do not  36  definitively identify the changes in neural activity underlying IOR. Consistent with attentional and perceptual accounts of IOR, however, several of these studies provide evidence for a relative inhibition of perceptual processing at the cued location, as indexed by a significant reduction in the amplitude of the occipital PI component (Eimer, 1994; Hopfinger & Mangun, 1998; McDonald, Ward, Kiehl, 1999; Prime & Ward, 2004; Wascher & Tipper, in press). It has also been found that peripheral cues are associated with negative differences (Nds) at several latencies between ERPs elicited by validly-cued targets and those elicited by invalidly-cued targets (Eimer, 1994; McDonald, Ward, Kiehl, 1999; Prime & Ward, 2004; Wascher & Tipper, in press). Each of the PI and Nd effects, however, have been observed both when IOR was obtained and when it was not. Consequently, the available data seem to allow no firm conclusions about the relationship between IOR and the observed modulations of the target-elicited ERPs. A more detailed examination of the existing ERP data and their relationship to the present results will be provided in the discussion. At the present time, only one study has used ERPs to examine the relationship between IOR and motor processes. Prime and Ward (2004) examined the effect of nonpredictive cues both on the amplitude of early sensory ERP components and on the motorrelated lateralized readiness potential (LRP). The LRP is an electrophysiological measure of brain activity generated in motor cortex that is related to the selection and preparation of motor responses (see Coles, 1989; Eimer, 1998). The latency of LRP onset is dependent on response selection so that the interval between target onset and the onset of the target-locked LRP provides a relative measure of the duration of processes involved in stimulus evaluation and response selection. Similarly, the interval between the onset of the response-locked  LRP  and response initiation provides a relative measure of the duration of motor processes (e.g., motor planning and execution). In the visual form discrimination task studied by Prime and  37  Ward, IOR was associated with a delay in pre-motor processes. The target-locked LRP began at a longer latency for validly-cued targets relative to invalidly-cued targets. In contrast, the onset of the response-locked LRP was not affected by the spatial relationship between cues and targets. Consistent with a perceptual locus of IOR, the amplitudes of the occipital PI and NI components were smaller for validly-cued targets relative to invalidlycued targets. Furthermore, an Nd effect was observed in the P2 latency range. 3.2.5 Present Study Although, the processing changes that give rise to IOR have not yet been identified, Prime and Ward's (2004) results indicate that IOR arises from a delay in pre-motor processes. It can be argued, however, that this study does not provide a definitive test of motor-based accounts of IOR. Although the lack of a response-locked LRP onset effect indicates that IOR can occur in the absence of an inhibition of motor processes, it is possible that motor inhibition may contribute to IOR in some circumstances. For example, Klein and Taylor (1994) proposed that IOR arises from an inhibition of responses to a specific spatial location and made the ad hoc assumption that detection responses are implicitly made "to a spatial location." Similarly, if IOR arises in part from the inhibition of an automatic tendency to respond to the cue, then a motor effect may only be observed when the response hand is known before the trial starts (e.g.; simple detection or Go-NoGo tasks) or the cue location is associated with a particular response (e.g.; target localization tasks). According to these accounts, an effect of IOR on motor processing may not have been observed by Prime and Ward because 1) the target location was unrelated to the correct response hand and 2) the appropriate response hand could not be selected until the identity of the target had been determined.  38 The present experiments were intended to serve two purposes. First, the possibility that inhibition o f motor processes may contribute to I O R under some circumstances was examined by measuring target- and response-locked L R P onsets in three tasks (Localization, Detection, and Identity-based Go-NoGo) that may be more susceptible to motor effects. Second, the effect of non-predictive peripheral cues on target processing was further explored by examining the target-elicited E R P s in the same tasks. If IOR arises from an inhibition o f perceptual processes then a reduction in the amplitude of the early occipital components should be observed for all three tasks. Three separate experiments were conducted. A n example of the stimulus display and trial sequence used in these experiments is shown in Figure 1. Target stimuli (white squares) were preceded by spatially non-predictive cues (brightening o f placeholder boxes) and by central reorienting events that followed the cues. The overall cue-target S O A varied between 900 and 1200 ms. In Experiment 1 (Localization Task), subjects were required to localize the target by pressing a spatially-corresponding button. In Experiment 2 (Detection Task) subjects were required to press a response button as soon as they detected the onset o f the target and to withhold their response on no-target catch trials. In Experiment 3 (Go-NoGo Task), subjects were required to respond to targets (white squares) and withhold responses from non-target nogo stimuli (white X shapes). Because responses from both hands are required to calculate uncontaminated L R P waves, in Experiments 2 and 3 subjects responded to half of the targets with each hand in separate blocks of trials.  39  Reorienting Event: 160ms  Cue: 200ms  S O A : 450-750ms SOA: 450ms  Figure 3.1. An example of the stimulus display and trial sequence used in all three experiments. A valid-cue trial is shown. SOA: stimulus onset asynchrony  40  3.3 Results 3.3.1 Behavioural Measures Reaction times. Mean reaction times on valid-cue and invalid-cue trials for all three experiments are presented in Table 1. The expected I O R effect, faster responses on invalidcue trials than on valid-cue trials, was obtained in all three experiments (all ^-values < 0.001). Error rates. In addition to the reaction time I O R effects, trial type also affected error rates in Experiments 1 and 3. More localization errors were made on invalid-cue trials (1.13%) than on valid-cue trials (0.45%) in Experiment 1 (F(l,26) = 14.60,p < 0.001). Similarly, subjects made more false alarm errors on invalid-cue N o G o trials (6.60%) than on valid-cue N o G o trials (3.43%) in Experiment 3 (F(l,27) = 16.55,p < 0.001). The fact that more errors were made on the faster invalid-cue trials than on the slower valid-cue trials in these experiments indicates that the I O R effect observed in the reaction time measure may arise in part from a speed-accuracy tradeoff. Misses and anticipations were rare (~1%) and the differences between error rates on valid- and invalid-cue trials did not approach significance i n either Experiment 2 (F(\,25) = 0.05,/? > 0.82) or on G o target trials in Experiment 3 (F(l,25) = 0.12,/? > 0.74).  3.3.2 LRP Measures The L R P s obtained in the present experiments are displayed in Figure 2 and the calculated L R P onset latencies are presented in Table 2. The effect o f Trial-type was considered separately for cue-locked L R P s (Experiment 1 only), response-locked L R P s , and target-locked L R P s . The onset latency effects were found to be consistent across experiments and the results o f all three experiments w i l l be presented together.  41  Table 1. Mean reaction times in milliseconds as a function o f cue validity. Experiment & Task  Reaction Time Valid  Invalid  IOR Effect  Significance  Valid-Invalid  Level  Experiment 1: Localization  444  426  18  p< 0.001  Experiment 2: Detection  446  410  36  p< 0.001  Experiment 3: Go/NoGo  449  416  33  p< 0.001  Table 2. Lateralized readiness potential ( L R P ) onset latencies in milliseconds. Experiment & Task  Component  Cue Condition  Effect  Significance  Valid  Invalid  Valid-Invalid  Level  Experiment 1  Response-locked LRP  -309  -302  -7  p>0.25  Localization  Target-locked LRP  180  168  12  p < 0.025  Experiment 2  Response-locked LRP  -292  -300  8  NA  Detection  Target-locked LRP  216  164  52  p< 0.001  Experiment 3  Response-locked LRP  -276  -324  48  NA  Go/NoGo  Target-locked LRP  224  164  60  p< 0.001  42  LRPs ( C 3 - C 4 ) Experiment 1: Localization »f "n  1  J  J  -1  i — i — - 1 i  Cue-locked  Target-locked  Response-locked  Experiment 2: Detection  Target-locked  Response-locked  Experiment 3: G o - N o G o  Target-locked Valid Invalid  Response-locked i.o  n\i - i — i — i — i  +0  600  Figure 3.2 Lateralized-readiness potentials (LRPs) from Experiments 1,2, and 3.  43 Cue-locked LRPs. The purpose o f the cue-locked L R P analysis in Experiment 1 was to examine the possibility that I O R may arise in part from an inhibition o f motor activity elicited by the cue. Coward et al. (2004) proposed that in cue-target experiments a proportion of the I O R effect is due to a spatially-specific motor inhibition that arises when subjects inhibit an automatic tendency to respond to the cue. To test this hypothesis, a cue-locked L R P wave was calculated with respect to the response hand associated with the location o f the cue such that negative (upward) deflections indicate response preparation in the direction of the cue and positive (downward) deflections indicate response preparation in the direction opposite to the cue. Previous research has shown that cues that predict the location o f to-belocalized targets give rise to L R P waves in the cue-target interval (Eimer, 1993). If subjects first prepared and then inhibited responses in the direction o f the cue in Experiment 1, we would expect the cue-locked L R P to show an initial upward deflection followed by a later return to baseline and possibly a downward deflection indicating that the response associated with the cued location is now inhibited. A s can be seen in Figure 2, this pattern of results was not obtained. The cue-locked L R P s did not significantly deviate from zero (Bonferroni ttests) in the direction o f the cue at short latencies (101-200 ms; 201-300 ms) nor did they did they deviate away from the cue before target onset (701-800 ms; 801-900 ms). This result suggests that either response inhibition to the cue was not contributing to the I O R effect observed in Experiment 1, or that response inhibition does not directly affect the handspecific motor preparation indexed by the L R P . Given the response-locked L R P results presented below, the response inhibition proposed by Coward et al. (2004) and others may instead arise from premotor stages of processing. Response-locked  LRPs. Motor accounts o f IOR propose that subjects respond more  slowly on valid-cue trials than on invalid-cue trials because on valid-cue trials, when the  44 subject must respond to targets presented at previously-cued locations, motor processes must overcome location-specific motor inhibition generated in response to the cue (e.g.; Tassinari, et a l , 1987; Godijn & Theeuwes, 2002). According to this account the responselocked L R P should onset earlier on valid-cue than on invalid-cue trials reflecting the longer time, because o f the need to overcome cue-related motor inhibition, required to generate the level of motor activation required to respond on valid-cue trials. A n examination o f the response-locked L R P waves presented in Figure 2 reveals virtually identical L R P waves on valid-cue and invalid-cue trials. This impression is supported by the onset latencies reported in Table 2. Despite the use o f powerful one-tailed tests the effect o f trial-type did not approach significance in Experiment 1 (7(26) = 0.45, p > 0.25). In Experiments 2 and 3 the latency difference between valid-cue and invalid-cue trials was in the opposite direction from that predicted by the motor account. These results indicate that motor inhibition did not contribute significantly, i f at all, to the reaction time effect. Target-locked LRPs. Target- and response-locked L R P s partition the reaction time interval into two mutually exclusive and exhaustive sub-intervals. A n effect of Trial-type was not observed on the onset latency o f the response-locked L R P and, therefore, an effect should be found on onset latency o f the target-locked L R P . Specifically, the target-locked L R P should onset later on the slower valid-cue trials than on the faster invalid-cue trials. A s can be seen in Figure 2, the target-locked L R P s are somewhat stretched and distorted due to the latency 'jitter' o f the response relative to the target. However, the L R P s rise faster and peak sooner on invalid-cue trials than on valid-cue trials. The presence o f a target-locked latency effect is supported by the onset latency measures (Table 2). The target-locked L R P s onset at a shorter latency on invalid-cue trials than on valid-cue trials in all three experiments (Experiment 1: (r(26) = 2.27,/? < 0.025; Experiment 2: (f(25) = 3.85,/? < 0.001; Experiment  45 3: (7(27) = 5.28, p < 0.001). Taken together with the lack o f response-locked effects, these results indicate that the I O R effect observed in the current experiments arises from changes in premotor processes.  3.3.3 ERP Measures Although the L R P results obtained in the current experiments and those obtained by i  Prime and Ward (2004) indicate that I O R does not arise from a slowing o f motor processes, they do not specify which premotor processes are involved. Target-locked L R P latency effects are consistent with changes in the duration of either or both perceptual or responseselection processes. To further characterize the processes underlying I O R we examined the visual E R P s elicited by the targets. The following analyses primarily focus on the effect of Trial-type on the amplitude o f the early occipital components (PI and N I ) and on the later parietal N d (see Methods). Unlike the L R P effects, the observed E R P effects varied somewhat across the experiments. We therefore present the E R P results from each experiment separately. A summary of the E R P amplitude effects is provided in Table 3.  Table 3. E R P average magnitudes (u.v, windows as described in text) Experiment & Task  Component and Electrode  Cue Condition  Effect  Significance  Valid  Invalid  Valid-Invalid  Level  Experiment 1  PI  P07, P08  0.55  0.66  -0.11  /?> 0.11  Localization  NI  P07, P08  -1.96  -2.37  0.41  p < 0.001  Nd  Pz  0.39  1.07  -0.68  pO.001  Experiment 2  PI  P07, P08  0.56  0.62  -0.06  p>0.25  Detection  NI  P07, P08  -1.59  -2.20  0.61  p< 0.001  Nd  Pz  0.26  1.80  -1.54  p< 0.001  Experiment 3  PI  P07, P08  1.00  1.20  -0.20  /?<0.05  Go/NoGo  NI  P07, P08  -1.27  -1.49  0.22  jD<0.05  Go Trials  Nd  Pz  0.29  1.33  -1.04  p< 0.001  46  Experiment 1: Localization  Task. The target-elicited E R P s observed in Experiment 1  are shown in Figure 3. In the analysis of the PI amplitude, neither the main effect o f Trialtype (F(l,26) = 1.56, p > 0.11) nor the main effect o f Electrode (F(l,26) = 0.07, p > 0.79) reached significance. The Trial-type x Electrode interaction, however, was significant (F(\,26)  = 6.88,  p < 0.02). Subsequent Bonferroni t-tests at each electrode site (family-wise  a = 0.05) revealed that the PI was reduced on valid-cue trials relative to invalid-cue trials over right occipital cortex (P08) but not over the left (P07). The analysis o f N l amplitude revealed a significant main effect o f Trial-type (F(l,26) = 12.15,/? < 0.001) and a significant Trial-type x Electrode interaction (F(l,26) - 7.98,p < 0.01). A s can be seen in Figure 3, the effect of Trial-type on N l amplitude was larger over right occipital cortex. The main effect o f Electrode approached significance (F(l,26) = 4.02,p < 0.06). Subsequent Bonferroni t-tests at each electrode site (family-wise a = 0.05) revealed that the N l was reduced on valid-cue trials relative to invalid-cue trials at both electrode sites. In addition to the P I and N l effects, a significant parietal N d effect was also observed (F(l,26) = 12.21,/? < 0.001). With the exception o f the lateralization o f the P I effect, these results replicate those o f Prime and Ward's (2004) discrimination experiment.  Experiment 1 Target-locked ERPs C2_  Valid Invalid  +0  400  Figure 3.3 Event-related potentials (ERPs) from Experiment 1.  48 Experiment 2: Detection Task. The target-elicited E R P s observed in Experiment 2 are shown in Figure 4. The analysis of P I amplitude failed to reveal significant main effects or interaction (all /?-values > 0.25). The small difference in P I amplitude between valid-cue and invalid-cue trials, about -0.06 pv averaged across electrodes (Table 3), is, however, in the same direction as that for Experiment 1 and that reported by Prime and Ward (2004). B y contrast, the analysis o f N l amplitude revealed a significant amplitude reduction on valid-cue trials relative to invalid-cue trials (F(l,25) = 21.50,p < 0.001). Neither the main effect of Electrode nor the Trial-type x Electrode interaction approached significance (both /^-values > 0.25). The N d effect measured over the parietal scalp was again observed (F(\,25) = 25.28,p < 0.001).  Experiment 2 Target-locked ERPs  --  PI  Valid Invalid  2.0 yLiV  +0  H  h  400  Figure 3.4 Event-related potentials (ERPs) from Experiment 2.  50 Experiment 3: GoNo Task. The E R P s elicited by both G o targets and N o G o stimuli in Experiment 3 are shown in Figure 5. A n examination o f Figure 5 reveals that the effect o f Trial-type on E R P amplitude in the P I , N l , and N d latency ranges is qualitatively similar for Go and N o G o trials. However, in this experiment G o trials were presented three times more often than N o G o trials and consequently Go-target-elicited E R P s have a higher signal-tonoise ratio. For this reason, the P I , N l , and N d effects were only analyzed for Go-target trials. The analysis o f the amplitudes o f the early occipital components revealed that both the P I (F(\,27) = 3.17,/? < 0.05) and N l (F(l,27) - 2.98,/? < 0.05) were reduced on valid-cue trials relative to invalid-cue trials. Neither the Electrode main.effects nor the Trial-type x Electrode interactions approached significance (all /?-values > 0.25). The N d effect measured over the parietal scalp was again observed (F(l,27) = 14.72,/? < 0.001). A n examination o f Figure 5 also reveals a central N 2 wave observed on N o G o trials but not on G o trials. Its amplitude was significantly larger on invalid-cue trials than on valid-cue trials (F(\, 27) = 21.16,/? < 0.001). There are a variety o f N 2 waves and it is not clear exactly what processes are giving rise to this effect. One possibility is that this effect represents a modulation o f the frontocentral " N o G o " N 2 wave that is generally considered to reflect a premotor response inhibition process (e.g.; Jodo & Kayama, 1992; Falkenstein, Hoorman, & Hohnsbein, 1999). If so, this finding may indicate that more inhibition was required on invalid-cue N o G o trials than on valid-cue N o G o trials. Alternatively, this effect may represent a modulation of the attention sensitive N 2 b component that is considered to represent late stages o f stimulus processing (e.g.; Mulder, Wijers, Brookhuis, Smid, & Mulder, 1994). Unfortunately, the origin of this effect cannot be unambiguously determined on the basis o f the current results.  51  Experiment 3 G o Target-locked ERPs  NoGo Target-locked ERPs  C3^ A  J/ V Pz  POL  NI  P08.  PI Valid Invalid  2.0 /zV .  +0  — i — i — i — i  400  Figure 3.5 Event-related potentials (ERPs) from Experiment 3.  52 Overall, the analysis o f target-elicited E R P s revealed effects of Trial-type that are consistent with those reported in previous studies. In particular, the occipital N l component was reduced on valid-cue trials relative to that on invalid-cue trials in all three experiments, and a parietal N d also appeared in all three. The IOR-related reduction in the occipital PI component was more erratic, appearing only on the right side in Experiment 1, not reaching significance in Experiment 2, and appearing on both sides only in Experiment 3. In general, these E R P effects support the idea that IOR affects early sensory/perceptual processing. We next discuss the implications o f these data and their relationship to previous results.  3.4 Discussion The results o f the present experiments provide no support for the purported contribution o f the inhibition o f motor processes to IOR. In all three experiments, cue validity had no effect on the duration o f motor processes (response-locked L R P onset latencies). B y contrast, the duration o f premotor processes was longer on valid-cue than on invalid-cue trials (target-locked L R P onset latencies). Although it is not possible to rule out an effect o f I O R on motor processes on the basis o f null results, these results do indicate that premotor effects were primarily responsible for the observed differences in reaction time in these experiments. Furthermore, no effect o f cue-validity was found on the onset latency o f the response-locked L R P s even though a large sample o f subjects was tested and powerful 1tailed significance tests were used. In addition, significant effects o f cue-validity were found on the onset latency o f the target-locked L R P despite the fact that L R P s are better timelocked to the motor response and tests of response-locked effects have more statistical power than tests of target-locked effects (Miller, Patterson, & Ulrich, 1998). Taken together with the results of Prime and Ward (2004), these results indicate that, when manual keypress responses are required, uninformative peripheral cues give rise to I O R by affecting  53 perceptual and possibly response-selection processes. Although we feel that it is likely that this conclusion applies to cue-target I O R generally, it remains a possibility that an inhibition of motor processes may contribute to I O R in some circumstances.  3.4.1 What processes are inhibited? Although the L R P onset results indicate that I O R arises from inhibition o f premotor processes, they do not specify exactly what processes are inhibited. One possibility is that IOR arises from changes in decisional processes. Consistent with the results o f Ivanoff and K l e i n (2001), subjects made significantly fewer errors on valid-cue trials than on invalid-cue trials in Experiments l-and 3. These results suggest that the reaction time differences between valid-cue trials and invalid-cue trials may arise, at least in part, from speed-accuracy tradeoffs in these experiments. Ivanoff and K l e i n (2001) suggested that such speed-accuracy tradeoffs result from subjects adopting a more conservative response criterion on valid-cue trials than on invalid-cue trials. Unfortunately, the exact nature o f speed-accuracy tradeoffs is still poorly understood (e.g.; Rinkenauer, Osman, Ulrich, Muller-Gethmann, & Mattes, 2004). It is not possible to state exactly what processing changes are giving rise to the tradeoff, nor is it possible to determine exactly how much these processes are contributing to the reaction time effect. Furthermore, speed-accuracy tradeoffs cannot be the sole cause of IOR because I O R is often observed either without any changes in accuracy (e.g., Prime & Ward, 2004) or accompanied by an increase in errors on valid-cue trials (e.g., Cheal & Chastain, 1999). A n alternative possibility is that I O R arises because o f changes in perceptual processes, possibly because attention was inhibited from reorienting to the cued location. Consistent with a perceptual locus o f IOR, the amplitudes o f the early occipital P I and N I E R P peaks were reduced on valid-cue trials relative to invalid-cue trials. These E R P  54 components are considered to reflect perceptual processing in extrastriate visual areas (e.g.; Mangun & Hillyard, 1995; D i Russo, Martinez, Sereno, Pitzalis, & Hillyard, 2001). Unlike Prime and Ward (2004), the P1 effects were relatively weak in the present experiments and only reached significance in Experiments 1 and 3. Although the reason for this across-task variation is not clear, it does suggest that the P I effect is not simply the result o f sensory refractoriness arising from repeated stimulation o f the same location. B y contrast, reliable N l reductions and parietal N d effects were found in all three experiments. If the changes in perceptual processes observed in the present experiments are involved in producing I O R then these changes should be associated with I O R across a variety o f experimental contexts. A t present, however, relatively few studies have examined the E R P correlates o f IOR and their results have been somewhat inconsistent. The reasons for this variability are unknown and cross-experiment comparisons are hampered by the many differences in stimuli, tasks, and procedures between studies. In order to facilitate our discussion o f the E R P correlates of IOR, we provide a summary of the relevant studies in Table 4. The observed effects of cue validity on reaction time (IOR) and on the amplitude of the P I and N l peaks are presented. In addition, the presence or absence o f two valid-cue minus invalid-cue N d waves is reported. One wave, labeled N d , corresponds to the parietal N d observed in the present experiments. The other, labeled Nde, is maximal at parietal and occipital sites ipsilateral to the target location and peaks at a latency around 150 ms, considerably earlier than the N d . The table provides a summary o f not only the statistically significant results (presented in bold), but also results that were present but failed to satisfy the statistical criteria (indicated by n.s.) and results that were present in the E R P waveforms but were not tested for statistical significance (presented in italics followed by a ' ? ' ) .  55 Table 4. Summary of Event-related potential effects found in studies investigating the effect of non-predictive stimulus at long cue-target intervals. For the PI and NI effects 'Reduction' refers to a smaller peak on valid trials and 'Enhancement' refers to a larger peak on valid trials. IOR: valid minus invalid reaction time difference; SOA: stimulus onset asynchrony between cues and targets; Nd: a negative difference observed between valid and invalid trials within the latency range of 200-300 ms; Nde; an early Nd effect observed within the latency range of 100-200 ms; n.s.: measured effect was not significant; ?: effect was observed but was not subjected to a statistical test; *: the indicated effect overlaps in time and space with the Nde effect; NA: information about the effect is not available  Behavioural IOR Observed: Task IOR (ms) P1 Effect Current Results: SOA 900-1200 ms Localization 18 Reduction Detection 36 Reduction (n.s.) Go/NoGo 33 Reduction  N1 Effect  Nde  Nd  Reduction Reduction Reduction  No No No  Yes Yes Yes  Prime & Ward (2004): SOA 900-1200 ms Discrimination 21 Reduction  Reduction  No  Yes  Yes  Yes  Enhancement (n.s.)*  Yes?  Yes (n.s.)  Enhancement (n.s.)*  Yes?  Yes  None  No  No  Enhancement*^.s.)  Yes  Yes  Enhancement*(n.s.)  Yes  Yes  Enhancement*(n.s.)  Yes  Yes  None  No  Yes  Enhancement Enhancement  Yes Yes  Yes Yes  No  NA  McDonald, Ward, & Kiehl (1999): Experiment 1: SOA 500-700 ms Detection 13 Reduction?* Experiment 1: SOA 900-1100 ms Detection 17 Reduction Experiment 2: SOA 500-700 ms Detection 13 Reduction Hopfinger & Mangun (2001): SOA 600-800 ms Detection 13 Reduction(n.s.) Wascher & Tipper (in press) Experiment 1: Transient Cue SOA 900 ms Detection 31 Reduction* Experiment 1: Sustained Cue SOA 900ms Detection 12 (n.s.) Reduction* Experiment 2: Cued Location SOA 900 ms Detection 24 Reduction* Experiment 2: Adjacent Location SOA 900 ms Detection 6 Reduction  Enhancement?*  No Behavioural Effect Observed: Eimer (1994): SOA 700 ms Discrimination 2 (n.s.) Localization 0  Reduction* Reduction*  Hopfinger & Mangun (1998): SOA 600-800 ms Discrimination 2 (n.s.) Reduction  Enhancement?  56 The Nde. One problem that has arisen in attempting to determine the effect o f cue validity on the early PI and N l peaks is that, in some experiments, these peaks are overlapped by the Nde, a relative negative deflection on valid trials (Eimer, 1994, M c D o n a l d et al., 1999; Wascher & Tipper, in press). Because o f the spatio-temporal overlap between the Nde and the P I and N l peaks, the PI peak appears to be reduced and the N l appears to be enhanced on valid trials. Peak amplitude measures that overlap with the Nde are marked with an asterisk in Table 4. M c D o n a l d et al. (1999) attributed the Nde to sensory refractoriness. Although speculative, this proposition is supported by evidence that the Nde is reduced or eliminated under all o f the following conditions: at longer S O A s (McDonald et al., 1999), when cues and targets are presented to different eyes (McDonald et al., 1999), when cues and targets are spatially separated (Wascher & Tipper, in press), and when 'subtle' cues are used (Hopfinger & Mangun, 1998, 2001). What effect the neural processes underlying the Nde may have on reaction time is unknown. We speculate, however, that consistent with the appellation "sensory refractoriness" the Nde slows neural processing and could thus slow reaction times to targets whose representations were affected by it. The PI Peak. Despite the problem posed by the Nde i n measuring the effect o f cuevalidity on P I and N l amplitude, several studies have reported reduced P I amplitudes on valid-cue trials that appear to be uncontaminated by overlap from the Nde. Unfortunately, P I reductions have been found both when IOR was observed (Mcdonald et al., 1999; Prime & Ward, 2004; Wascher & Tipper, in press; current results) and when it was not (Hopfinger & Mangun, 1998). Hopfinger and Mangun (1998) found a significant P I reduction accompanied by a non-significant 2-ms IOR effect. Conversely, Hopfinger and Mangun (2001) found a significant 13-ms IOR effect but a non-significant P I reduction. O n the basis of these results, Hopfinger and Mangun (2001) concluded that I O R was dissociable from the  57 P I reduction. The available data, however, suggest an alternative explanation. First, the PI reduction found by Hopfinger and Mangun (1998) was accompanied by a large but unmeasured N I enhancement. These two effects may have had opposite effects on reaction time and cancelled each other out. This speculation is supported by the results o f Eimer (1994) who also found opposite PI and N I effects in the absence of a reaction time effect. Second, although not statistically significant, a PI reduction was observed by Hopfinger & Mangun (2001). The small sample size (n=8) employed by Hopfinger & Mangun may not have yielded sufficient power to reliably detect the P I reduction. We note that our Experiment 2, also, in spite of large I O R found only a non-significant PI reduction, even though we ran enough observers to yield adequate power. Nonetheless, no study has ever found I O R to be associated with an enhancement o f the PI component, so the overall likelihood o f a P I reduction being associated with I O R is substantially greater than 1. Taken together, the currently available evidence indicates that I O R is usually associated with PI amplitude reductions. Furthermore, a substantial body o f attention research has found that PI amplitude is usually directly related to behavioural performance (Luck, Woodman, & Vogel, 2000; Mangun, 1995). This suggests that changes in early perceptual processes indexed by the occipital P I component could play a causal role in IOR. The NI Peak. In all o f the present experiments, and consistent with the results o f Prime and Ward (2004), the amplitude o f the N I peak was reduced on valid-cue trials relative to invalid-cue trials. A n inspection of Table 4 reveals that these findings are atypical. Most studies to date have found a non-significant trend in the opposite direction. It is likely, however, that these enhancement effects result from the overlap o f the Nde, and possibly the early portion o f the N d , with the N I peak. Furthermore, the two studies that show N I enhancement effects that seem not to result from Nde overlap did not find I O R (Eimer, 1994;  58 Hopfinger & Mangun, 1998). The existing experiments differ in far too many respects to allow any clear conclusions to be drawn regarding the effect of cue-validity on the amplitude of the N l peak. Three major differences between the present experiments and all o f the others are: (1) the presence of reorienting events between cues and targets, (2) the vertical distribution o f cue and target locations, and (3) the somewhat longer S O A s . One possibility is that the N l reduction may only arise at very long S O A s , such as those used in the present experiments and in Prime and Ward (2004). This could happen because o f persistence of facilitatory attentional effects at the cued location which would counteract the IOR. Another possibility is that the re-orienting event and/or the vertical distribution o f cue/target sites could have activated a different N l generator than that engaged by the task demands o f the other studies listed in Table 4. The Parietal Nd. Inspection of Table 4 also reveals that an N d effect was found in most experiments. The neural origin and functional significance o f this component is unknown. Wascher and Tipper (in press) found a parietal N d component, which they labeled the Nd250, that was smaller when the I O R effect was large (their Experiment 1, transient cue condition) than when the I O R effect was small (sustained cue condition). O n the basis o f this result, Wascher and Tipper proposed that this component reflects the activity o f a facilitatory process that is compensating for earlier perceptual inhibition on valid-cue trials. This proposal is supported by the results of several studies that have found that similar N d effects are observed when attention is oriented by informative symbolic or peripheral cues ( e.g., Eimer, 1993, 1994, 1997). Furthermore, Eimer (1994) found a central-parietal N d that was larger in magnitude when peripheral cues were informative and facilitated reaction time on valid-cue trials than when the cues were uninformative and no reaction time effects were observed. If this explanation is correct then the N d is not related to the generation o f IOR.  59 To sum up, it appears from the current results and those o f several other studies that early sensory and/or perceptual processes, as indexed by changes in the PI and N I E R P components, are involved in the generation of the I O R effect. It is possible that some IORlike effects might also be generated by processes indexed by the Nde, although not all studies find this effect. While the exact roles that these changes play remain to be discovered, it is possible to offer a plausible account o f the present data within the framework o f a recent account of the brain circuitry of visual attention orienting (Shipp, 2004). Figure 3.6 summarizes this account, which was arrived at by integrating what is known about the anatomy and physiology of the relevant brain areas with five psychological process models of the operation of attention orienting in vision. Each o f the five psychological models is consistent with a large body of empirical behavioral data, so that this account integrates a very large body o f physiological and psychological data. In this "real neural architecture" model of visual attention, the ventral pulvinar nucleus o f the thalamus is supposed to contain a saliency map that pools information from several brain areas, including various visual cortical processing areas, the superior colliculus, the frontal and parietal eye fields (top-down control, strategy), and the prefrontal cortex (memory for goals, e.g., target features). Focal states o f attention emerge from the interaction of all of this information in the ventral pulvinar and take the form o f a localized "beam" of neural activity in the ventral pulvinar that spreads along reciprocal connections to the various visual and other processing areas. In this model, I O R is explicitly represented by a loop from the frontal and/or parietal eye fields through the superior colliculus to the ventral pulvinar that reduces the salience o f a recently explored spatial location so that other locations in the visual field can out compete that location for focal attention. This mechanism is consistent with evidence that both the frontal eye fields (Ro et al., 2002) and the superior colliculus (e.g., Sapir et al., 1999) are  60 involved in generating IOR. In order to make contact with behavioral data such as reaction time, two additional, reasonable, assumptions are needed. First, the effect o f focal attention is assumed to be to increase the signal-to-noise ratio of the attended stimulus relative to unattended stimuli (indexed by amplitude modulations of the P I and N l peaks). Second, the input from the target stimulus to an evidence-accumulator decision mechanism is assumed to depend on the signal-to-noise ratio. In such a decision mechanism (e.g., Ratcliff & Rouder, 1998) evidence is accumulated in favor o f a particular response until some threshold is reached, at which time response execution begins (indexed by L R P onset). Reaction time thus depends directly on the rate o f evidence accumulation, among other factors. In the present experiments and most others using the cue-target paradigm, a reduction o f the salience of a cued location through IOR, as illustrated in Figure 3.6, would have the effect o f depriving this location of focal attention for a longer time after the target had appeared there than for unaffected locations. Although focal attention would eventually affect processing at any target location, since the target is highly salient based on inputs from prefrontal cortex and from the visual response to an abrupt onset stimulus, focal attention would affect processing at uncued locations sooner, since salience buildup at those locations would not have to overcome the negative salience applied by the I O R mechanism to the cued location. Thus, for targets at so-called valid-cue locations the initial target processing would be less effective and the signal-to-noise ratio of the signal sent to decision mechanisms would be lower, resulting in a longer reaction time to targets at those locations. Although this account should be useful as a framework for future studies, direct evidence as to its adequacy is lacking. What is needed is more systematic study o f the E R P and other brain imaging correlates o f IOR and other mechanisms o f attention. The existing studies are too fragmentary and exploit idiosyncratic designs in single or few-study series.  61 Parametric manipulation o f several important design factors, such as the presence and type of reorienting events, distribution o f cue and target locations, S O A , task type and difficulty, and even sensory modality of cue and target, need to be performed during E R P recording before solid conclusions can be drawn about exactly how the brain implements IOR, and how these processes are related to those responsible for attention and visual search.  62  Prefrontal  IOR  Frontal Eye Fields Parietal Eye Fields  Attention Orienting & Salience  Signals^ Ventral Pulvinar Saliency Map  'i to i&  Top-down Commands  v  Salient Locations  Decision Mechanism Attention Modulated Perceptual Signals  \ VN  Figure 3.6 Graphical representation of the 'real neural architecture" model (Shipp, 2004) of visual spatial attention and inhibition of return (IOR). The focus of attention is jointly determined by stimulus salience and top-down control signals. IOR is implemented by a mechanism that reduces the salience of stimuli at recently explored spatial locations. We expand upon this model by assuming that (1) allocation of attention to spatial location improves the signal-to-noise of stimuli at this location relative to stimuli at unattended locations and (2) the input from the target stimulus to an evidence-accumulator decision mechanism depends on the signal-to-noise ratio. See text for details. This figure is an elaboration of Figure 2(g) in Shipp (2004).  63  3.5 Methods 3.5.1 General Procedure A n example o f the stimulus display and trial sequence used i n all three experiments is shown in Figure 1. Subjects viewed a computer monitor from a distance o f 42 cm and were instructed to maintain fixation on a centrally-located fixation cross during the experimental blocks and to blink between trials. The screen background was black and displayed at all times two grey square outline boxes (1.5°x 1.5°) centered 5° above and below fixation. After a 700-ms inter-trial interval, each trial began with a 133-ms offset o f the fixation cross, followed 800 ms later by the cue. The cue consisted o f a 200-ms brightening of one o f the two square boxes and was equally likely to occur at either location. Two hundred fifty ms after cue offset, a white circle 0.75° in diameter was presented at fixation for 160 ms. This re-orienting event was intended to redirect the participants' attention back to fixation from the cued location. O n most trials (see below), after a variable delay o f 290 to 590 ms, the target was presented for 1000 ms. The total cue-target S O A ranged between 900 and 1200 ms. The target, a white square (0.75°), was presented with equal probability within one of the two peripheral boxes (chance coincidence of cue and target locations). Trials on which cue and target occurred at the same location were classified as valid-cue trials, whereas trials on which cue and target occurred at opposite locations were classified as invalid-cue trials. Trials were presented in short blocks and subjects were allowed to rest as long as they wished between blocks.  3.5.2 Experiment 1: Localization Task In this experiment subjects responded based each target's spatial location. T w o response buttons were positioned one above the other (60° from horizontal) beneath the display. Subjects were required to press the upper button with the index finger o f their right  64 hand for targets that appeared above fixation and to press the lower button with the index finger o f their left hand for targets that appeared below fixation. Both speed and accuracy were stressed i n the instructions. This arrangement of response and target locations has been shown to elicit stimulus-response compatibility effects such as the Simon effect (e.g.; De Jong, Liang & Lauber, 1994). Each subject completed 24 blocks o f 28 trials.  3.5.3 Experiment 2: Detection Task Subjects were required to press a response key when they detected the onset o f the target. In order to discourage anticipatory responding, 14% o f trials were no-target catch trials. The subjects alternated across blocks between responding with their left and right hands. O n half o f the experimental blocks the subject responded by pressing the 7 ' key on a standard computer keyboard with the index finger of their right hand and on the other half they pressed the ' Z ' key with the index finger o f their left hand. During each block the nonresponding hand rested on the subject's lap. Each subject completed 24 blocks o f 28 trials.  3.5.4 Experiment 3: Go-NoGo Task In this experiment subjects were required to press a response key whenever they detected the onset o f the square ' G o ' target stimulus and to refrain from responding when a ' N o G o ' stimulus (an X shape, 0.75° x 0.75°) was presented. The G o stimulus was presented on 75% of trials and the N o G o stimulus was presented on the remaining 2 5 % o f trials. Subjects alternated response hand across blocks in the same manner as Experiment 2. Each subject completed 20 blocks o f 32 trials.  3.5.5 Electrophysiological Recording and Data Processing Scalp potentials were recorded from central ( C I , C 2 , C 3 , C4), parietal (Pz), and occipital ( P 0 7 , P 0 8 ) scalp sites. These electrodes were referenced to the right mastoid and  65 subsequently re-referenced to averaged mastoids. Eye position was monitored by both the horizontal and vertical electro-oculogram (EOG). The E E G and E O G channels were amplified with a bandpass of 0.01-100 H z (half-amplitude cutoffs) and sampled at 250 H z . Trials containing eye movement, muscle, and blocking artifacts were removed prior to averaging by applying automated artifact detection routines. In addition, trials with errors and those with reaction times outside the range of 100-1000 ms were excluded from the analysis. E R P s were calculated separately for valid-cue and invalid-cue trials at all electrodes sites. After averaging, the E R P s were digitally low-pass filtered (16 H z half-amplitude cutoff) to . eliminate high frequency noise and high-pass filtered (1.5 H z half-amplitude cutoff) to remove low frequency overlap from the cue, re-orienting event, and motor potentials. Target and response-locked L R P s were calculated at C 3 - C 4 for valid-cue and invalid-cue trials using the averaging method (Coles, 1989). The L R P s were low-pass filtered (6 H z halfamplitude cutoff). Target-locked L R P s were baseline corrected to the mean voltage o f the 100 ms interval preceding the target. Response-locked L R P s were baseline corrected to the mean voltage between 1000 and 1200 ms preceding the response.  3.5.6 Data Analysis Behavioural  Analysis  For all experiments, trials with reaction times smaller than 100 ms (anticipations) or larger than 1000 ms (misses) were counted as errors and excluded from the reaction time analysis. Mean reaction times for each subject were then calculated separately for correctlyresponded-to valid-cue and invalid-cue trials and entered into repeated measures Analyses o f Variance ( A N O V A s ) . Mean error rates (percent errors) were also calculated for valid-cue and invalid-cue trials and analyzed by repeated measures A N O V A s . In Experiment 1, errors  66 consisted o f anticipations, misses, and incorrect localizations. In Experiment 3, separate error rates were calculated for ' G o ' trials (anticipations and misses), and N o G o trials (false alarms). Electrophysiological  Analyses  Response-locked L R P onsets were determined using the one-degree-of-freedom ( I D F ) regression technique (Mordkoff and Gianaros, 2000). Regression-based techniques define the onset o f the L R P as the intersection o f two straight lines that are fitted to segments of the L R P waveform (Schwarzenau, Faulkenstein, Hoormann and Hohnsbein, 1998). The I D F technique sets the pre-onset line to have a height and slope o f zero. The terminus o f the post-onset line is locked to the peak o f the L R P so that only the time intersection can vary. The post-onset line is then found using the least-squares technique to find the best fit to the L R P . The target-locked L R P s observed in the present study deviated significantly from the straight lines assumed by the regression technique and poor fits were obtained (especially in Experiments 2 and 3). For this reason, target-locked L R P latencies were determined using the threshold technique recommended by Miller, Patterson, and U l r i c h (1998) with the threshold set to 50 percent o f the peak amplitude. This technique provides high power and low bias i n detecting target-locked effects in the absence o f response-locked effects, as in the present study (Miller, et al., 1998). Both response- and target-locked L R P onsets were determined and statistically assessed with a jackknifing procedure (Miller, 1974; Miller, et al., 1998). This jackknifing procedure involves determining the L R P onset latency for N different subsample grand averages, with each participant omitted from one o f the sub-samples. The values o f these onset latencies across sub-samples are then used to calculate a jackknife standard error for pairwise comparisons between conditions. Onset differences between valid- and invalid-cue trials were then tested for statistical significance by repeated measures  67 t-test using this jackknifed standard error. In addition, cue-locked L R P amplitudes were measured i n Experiment 1. Mean amplitudes were measured i n 100 ms latency windows separately for valid-cue and invalid-cue trials shortly after cue presentation (101-200 ms; 201-300 ms) and immediately before target .presentation (701-800 ms; 801-900 ms). These amplitude measures were tested against zero with one-sample Bonferroni t-tests (family-wise a =0.10). Amplitude differences between E R P waveforms from valid- and invalid-cue trials were measured for the occipital PI (peak latency 120 ms), occipital N I (peak latency 172 ms), and parietal N d components in all three experiments. The amplitudes o f the occipital P I and N I components were measured as the mean amplitude in 40-ms latency windows centered on the peaks o f these components in the grand-average waveforms at electrode sites P 0 7 and P08.. These amplitude measures were analyzed by a 2x2 repeated measures A N O V A with factors o f Trial-type (valid-cue, invalid-cue) and Electrode ( P 0 7 , P 0 8 ) . The parietal N d component was assessed by measuring the mean amplitude at Pz i n a 40-ms window centered on the peak o f the valid-cue minus invalid-cue difference wave between 200-300 ms. The measurement windows were 204-244 ms in Experiment 1, 224-264 ms i n Experiment 2, and 204-244 ms in Experiment 3. Differences between mean amplitudes on valid- and invalid-cue trials in these measurement windows were analyzed with repeated measures A N O V A s . In addition to these three components, the amplitude o f the central N 2 elicited on N o G o trials in Experiment 3 was measured for valid- and invalid-cue trials as the mean amplitude between 280-330 ms at sites C I and C 2 . These amplitudes were submitted to a 2x2 repeated measures A N O V A with factors o f Trial-type (valid-cue, invalid-cue) and Electrode ( C 1 , C 2 ) .  68 With the exception of the central N2 observed in Experiment 3, the effect of Trialtype (valid-cue vs. invalid-cue) was tested with 1-tailed tests for all ERP and LRP measures. The use of 1-tailed tests is justified (or even preferred) in the present experiments for two reasons. First, strong a priori directional hypotheses exist based on both theoretical and empirical grounds (see the Introduction). Second, 1-tailed tests increase power and minimize the chance of Type-II errors. This is particularly important with respect to the LRP onset measures. Falsely accepting the null hypothesis would have important theoretical consequences in this case. Inspection of the p-values reported in the Results section indicate that the results would be largely the same had 2-tailed tests been used. In addition, in no case would an effect opposite to that predicted have approached significance had 2-tailed tests been used. 3.5.7 Subjects All subjects were naive volunteers and were paid $15 (Canadian) dollars for participation in a single 1.5-hour session. All subjects reported normal or corrected-to-normal vision. Separate groups of subjects participated in each of the three experiments. Some subjects participated in more than one experiment. Thirty-one subjects participated in Experiment 1 (Localization task). Data from four of these subjects were excluded from analysis because of excessive eye movements (3 subjects) or equipment malfunction (1 subject). The remaining 27 subjects (17 female) had a mean age of 19.8 years. Twenty-eight subjects participated in Experiment 2 (Detection task). Two subjects were excluded from analysis due to excessive eye movements. The remaining 26 subjects (15 female) had a mean age of 21.0 years. Thirty-one subjects participated in Experiment 3 (Go-NoGo task). Three subjects were excluded from analysis due to excessive eye movements. The remaining 28 subjects (17 female) had a mean age of 19.7 years.  69  References Cheal, M . , & Chastain, G . (1999). Inhibition of return: Support for generality o f the phenomenon. Journal o f General Psychology, 126, 375-390. Coles, M . G . H . (1989). Modern mind-brain reading: Psychophysiology, physiology and cognition. Psychophysiology, 26, 251-269. Coward, R. S., Poliakoff, E . , O'Boyle, D . J., & Lowe, C. (2004). The contribution o f nonocular response inhibition to visual inhibition of return. Experimental Brain Research, 155, 124-128. Dorris, M . C , K l e i n , R. M . , Everling, S., & Munoz, D . P. (2002). Contribution o f the primate superior colliculus to inhibition of return. Journal o f Cognitive Neuroscience, 14, 1256-1263. Eimer, M . (1994). Spatial cueing, sensory gating and selective response preparation: an E R P study on visuo-spatial orienting, Electroencephalography and Clinical Neurophysiology, 88, 408-420. Eimer, M . (1994). A n E R P study on visual spatial priming with peripheral onsets. Psychophysiology, 31, 154-163. Eimer, M . (1998). The lateralized readiness potential as an on-line measure o f central response activation processes. Behavioral Research Methods, Instruments, & Computers, 30, 146-156. Falkenstein, M . , Hoorman, J., & Hohnsbein, J. (1999). E R P components in Go/Nogo tasks and their relation to inhibition. Acta Psychologica, 101, 267-291. Fischer, M . H . , Pratt, J., & Neggers, S. F. W . (2003). Inhibition o f return in manual pointing movements. Perception & Psychophysics, 63, 379-387. Fuentes, L . J., Vivas, A . B . , & Humphreys, G . W . (1999). Inhibitory mechanisms o f attentional networks: Spatial and semantic inhibitory processing. Journal o f Experimental Psychology: Human Perception & Performance, 25, 1114-1126. Godijn, R., & Theeuwes, J. (2002). Oculomotor capture and inhibition o f return: Evidence for an oculomotor suppression acoount of IOR. Psychological Research/Psychologishce Forschung, 66, 234-246. Handy, T. C , Jha, A . P., & Mangun, G . R. (1999). Promoting novelty in vision: Inhibition o f return modulates perceptual-level processing. Psychological Science, 10, 157-161. Harvey, N . (1980). Non-informative effects of stimuli functioning as cues. Quarterly Journal o f Experimental Psychology, 32, 413-425.  70 Hopfinger, J. B . , & Mangun, G . R. (1998). Reflexive attention modulates processing o f visual stimuli in human extrastriate cortex. Psychological Science, 9(6), 441-447. Hopfinger, J. B . , & Mangun, G . R. (2001). Tracking the influence of reflexive attention on sensory and cognitive processing. Cognitive, Affective & Behavioral Neuroscience, L 56-65. Howard, L . A . , Lupianez, J., & Tipper, S. P. (1999). Inhibition o f Return in a selective reaching task: an investigation o f reference frames. The Journal o f General Psychology, 126, 421 -442. Ivanoff, J., & K l e i n , R. M . (2001). The presence o f a nonresponding effector increases inhibition o f return. Psychonomic Bulletin & Review, 8, 307-314. Jodo, E . , & Kayama, Y . (1992). Relation o f a negative E R P component to response inhibition in a Go/Nogo task. Electoencephalography and Clinical Neurophysiology, 82, 477482. K l e i n , R. (1988). Inhibitory tagging system facilitates visual search. Nature, 334(4) 430-431. K l e i n , R. (2000). Inhibition of return. Trends in Cognitive Sciences, 4(4), 138-147. K l e i n , R. M . , & Dick, B . (2002). Temporal Dynamics of reflexive attention shifts: A dualstream rapid serial visual presentation exploration. Psychological Science, 13, 176179. K l e i n , R. M . , & Maclnnes, W . J. (1999). Inhibition of return is a foraging facilitator in visual search. Psychological Science, 10(4), 346-352. K l i e n , R. M . , Schmidt, W . C , & Miiller, H . J. (1998). Disinhibition o f return: Unnecessary and unlikely. Perception & Psychophysics, 60, 862-872. K l e i n , R. M . , & Taylor, T. L . (1994). Categories of cognitive inhibition, with reference to attention. In D . Dagenbach & T. H . Carr (Eds.), Inhibitory processes in attention, memory, and language (pp. 113-150). San Diego, C A : Academic Press. Maclnnes, W . J., & K l e i n , R. M . (2003). Inhibition o f return biases orienting during the search o f complex scenes. The Scientific World, 3, 75-86. Maylor, E . A . (1985). Facilitatory and inhibitory components of orienting in visual space. In M . I. Posner & O. S. M . M a r i n (Eds.), Mechanisms o f Attention: Attention and Performance X I . Hillsdale, N e w Jersey: Lawrence Erlbaum Associates. M c D o n a l d , J. J., Ward, L . M . , & K i e h l , K . A . (1999). A n event-related brain potential study of inhibition o f return. Perception & Psychophysics, 61(7), 1411-1423. Miller, R. G . (1974). The jackknife-a review. Biometrika, 61, 1-15.  71 Miller, J., Patterson, T., & Ulrich, R. (1998). Jackknife-based method for measuring L R P onset latency differences. Psychophysiology. 35, 99-115. Mordkoff, J. T., & Gianaros, P. J. (2000). Detecting the onset of the lateralized readiness potential: A comparison o f available methods and procedures. Psychophysiology, 37. 347-360. Mulder, G . , Wijers, A . A . , Brookhuis, K . A . , Smid, H . G . O. M . , & Mulder, L . J. M . (1994). Selective visual attention: Selective cuing, selective cognitive processing, and selective response processing. In H . J. Heinze, T. F. Miinte & G . R. Mangun (Eds.), Cognitive Electrophysiology (pp. 26-80). Boston, M A : Birkauser. Muller, H . J., & von Miihlenen, A . (2000). Probing distractor inhibition in visual search: Inhibition o f return. Journal o f Experimental Psychology: Human Perception and Performance, 26, 1591-1605. Poliakoff, E . , Spence, C , O'Boyle, D . J., M c G l o n e , F. P., & Cody, F. W . J. (2002). Tactile inhibition of return: non-ocular response inhibition and mode of response. Experimental Brain Research, 146, 54-59. Posner, M . I., & Cohen, Y . (1984). Components o f visual orienting. In H . Bouma & D . G . Bouwhuis (Eds.), Attention and performance, X (pp. 531-556). Hillsdale, N J : Erlbaum. Posner, M . I., Rafal, R. D . , Choate, L . S., & Vaughan, J. (1985). Inhibition o f return: Neural Basis and Function. Cognitive Neuropsychology. 2. 211-228. Prime, D . J., & Ward, L . M . (2004). Inhibition o f return from stimulus to response. Psychological Science, 15, 272-276. Rafal, R. D . , Calabresi, P. A . , Brennen, C. W . , & Sciolto, T. K . (1989). Saccade preparation inhibits reorienting to recently attended locations. Journal o f Experimental Psychology: Human Perception and Performance, 15, 673-685. Rinkenauer, G . , Osman, A . , Ulrich, R., Muller-Gethmann, H . , & Mattes, S. (2004). O n the locus o f speed-accuracy trade-off in reaction-time: Inferences from the lateralized readiness potential. Journal of Experimental Psychology: General, 133, 261-282. Reuter-Lorenz, P. A . , Jha, A . P., & Rosenquist, J. N . (1996). What is inhibited in inhibition of return? Journal o f Experimental Psychology: Human Perception and Performance, 22, 367-378. R o , T., Fame, A . , & Chang, E . (2003). Inhibition of return and the human frontal eye fields. Experimental Brain Research, 150, 290-296. R o , T., Pratt, J., & Rafal, R. D . (2000). Inhibition of return in saccadic eye movements. Experimental Brain Research, 130, 264-268.  72 Sapir, A . , Soroker, N . , Berger, A . , & Henik, A . (1999). Inhibition o f return in spatial attention: Direct evidence for collicular generation. Nature Neuroscience, 2, 10531054. Schmidt, W . C . (1996). Inhibition o f return is not detected using illusionary line motion. Perception & Psychophysics, 58, 883-898. Schwarzenau, P., Faulkenstein, M . , Hoormann, J., & Hohnsbein, J. (1998). A new method for the estimation o f the onset o f the lateralized readiness potential ( L R P ) . Behavior Research Methods, Instruments, & Computers, 30, 100-117. Shipp, S. (2004). The brain circuitry o f attention. Trends in Cognitive Science, 8, 223-230. Spence, C. J., & Driver, J. (1998). Auditory and audiovisual inhibition o f return. Perception & Psychophysics, 60, 125-139. Takeda, Y . , & Y a g i , A . (2000). Inhibitory tagging in visual search can be found i f search stimuli remain visible. Perception & Psychophysics, 62, 927-934. Tassinari, G . , Aglioti, S., Chelazzi, L . , M a r z i , C. A . , & Berlucchi, G . (1987). Distribution in the visual field o f the costs o f voluntarily associated attention and o f the inhibitory after-effects o f covert orienting. Neuropsychologia, 25, 55-71. Tassinari, G . , & Campara, D . , Benedetti, C , & Berlucchi, G . (2002). The contribution of general and specific motor inhibitory sets to the so-called auditory inhibition o f return. Experimental Brain Research, 146, 523-530. Taylor, T. L . , & K l e i n , R. M . (1998). O n the causes and effects o f inhibition o f return. Psychonomic Bulletin & Review, 5, 625-643. Taylor, T. L . , & K l e i n , R. M . (2000). Visual and motor effects in inhibition o f return. Journal of Experimental Psychology: Human Perception and Performance, 26, 1639-1656. Wascher, E . , & Tipper, S. P. (in press). Revealing effects of noninformative spatial cues: A n E E G study o f inhibition of return. Psychophysiology Wright, R. D . , & Ward, L . M . (1998). The control o f visual attention. In R. D . Wright (Ed.), Visual Attention (pp. 132-186). N e w York, N Y : Oxford University Press.  73  Chapter IV: Examining the Relationship Between Event-related Potential and Reaction Time Effects 4.1 Chapter Preface The results presented in the previous two chapters demonstrate that target-elicited activity recorded over occipital cortex is suppressed on valid-cue trials relative to invalid-cue trials. Although these results are consistent with a perceptual locus o f inhibition o f return (IOR), it is possible that the observed processing changes do not substantially contribute to observed reaction time effects. The experiment presented in this chapter provides further evidence for a perceptual locus of IOR by demonstrating covariation between event-related potential ( E R P ) and reaction time effects. The experiment presented in this chapter has not yet been submitted for publication and w i l l be included in an upcoming manuscript.  4.2 Introduction Although the results presented in the previous two chapters clearly demonstrate that IOR is associated with changes in early visual processing, these data alone cannot establish that the reductions o f the early E R P components are responsible for the behavioral effect. Such a claim would be strengthened i f it could be shown that factors that affect the magnitude o f behavioral I O R have a corresponding effect on the magnitude o f the observed E R P modulations. However, care must be taken to ensure that experimental conditions do not differ in regard to other factors that affect the amplitude or latency o f the E R P components of interest. For example, varying target luminance not only modulates the magnitude o f the behavioural I O R effect but also modulates the magnitude and latency o f the early sensoryrelated components o f the visual E R P in non-IOR paradigms. Such differences in E R P morphology arising from confounded factors make comparisons o f E R P effects across  74 experiments difficult to interpret. Furthermore, E R P s have low a signal-to-noise ratio, necessitating experimental manipulations that w i l l have large effects on E R P amplitude. In order to avoid such difficulties, I examined the effect of eliminating the reorienting event between cue and target on both E R P s and behaviour. Recently, Prime, Visser and Ward (2004) demonstrated that, at a cue-target S O A of 800 ms, I O R was only observed in a form discrimination task when a re-orienting event was presented in the interval between the cue and the target. Consequently, it seems plausible that eliminating the re-orienting event from the very similar experimental design used by Prime and Ward (Chapter 2) w i l l eliminate the behavioural I O R effect they found. This manipulation is ideal for present purposes because (1) it is anticipated to have a large effect on the reaction time I O R effect and (2) it would not create either stimulus-dependent or task-dependent differences i n the target E R P s between conditions. The current experiment is identical to the experiment o f Prime and Ward (2004) (Chapter 2), with the exception that a re-orienting event was not presented between the cue and the target. It was anticipated that little or no I O R would be observed in this experiment. A causal link between the amplitude modulations of the P I and N I effects observed by Prime and Ward (2004) and I O R would be supported i f these modulations were also reduced or eliminated in the current experiment.  4.3 Method A l l aspects o f this experiment were identical to those o f Prime and Ward (2004) (Chapter 2) except that a re-orienting event was not presented between the cue and target. The cue-target S O A remained 900-1200 ms. Data were collected from twenty-seven participants. T w o participants were excluded from the analysis due to excessive eye movement artifacts. Cross-  75 experiment comparisons o f the effect of the presence or absence o f a re-orienting event on reaction time and E R P effects were performed. The behavioural and E R P measures were submitted to the same analyses as those performed by Prime and Ward (2004). In addition, cue validity effects (valid-cue minus invalid-cue), for the present experiment and that o f Prime and Ward (2004) were calculated for reaction time and for occipital P I , occipital N l , parietal N d , and occipital N d amplitudes. These effects were submitted to between groups analyses-of-variance ( A N O V A s ) . In addition to the between group factor o f Experiment, the analysis of the occipital E R P amplitude effects included the factor o f recording electrode (P07, P08).  4.4 Results and Discussion A s in Prime and Ward (2004), participants in the current experiment responded significantly faster on invalid-cue trials (548 ms) than on valid-cue trials (553 ms; F ( l , 2 4 ) = 4.3, p < 0.05). There was no significant difference between the error rates on valid (5.0%) and invalid trials (5.7%). Although the I O R effect was not eliminated by removing the reorienting event, the magnitude o f the effect was significantly smaller in the present experiment (5 ms) than in Prime and Ward (2004) (21 ms; F ( l , 4 2 ) = 13.6, p < 0.001). Unlike Prime and Ward, no onset differences were found between valid-cue and invalid-cue trials for either target- or response-locked L R P s (Figure 4.1, all /-values <1). Because the targetand response-locked L R P s partition the reaction time interval into two mutually exclusive and exhaustive sub-intervals, an effect on reaction time must be present in one or the other or both. The lack o f an effect in the present experiment must be due to a lack o f sufficient power to detect L R P onset differences associated with such a small reaction time effect.  76 The reduced I O R effect observed in this experiment was also associated with an elimination o f the P I (F(l,24) = 0.1,/? > 0.70) a n d N l (F(l,24) = 0.0,/? = 1) amplitude modulations (Figure 4.1). The cross-experiment A N O V A revealed that the main effect o f cue validity on P1 amplitude differed reliably between the current no re-orienting event experiment (0.05 pV) and the with re-orienting event experiment (-0.43 juV) o f Prime and Ward (F(l, 42) = 2.77,/? = 0.05; 1-tailed). Similarly, the effect o f cue validity on the N l amplitude was reliably different between the current no re-orienting event experiment (0.00 juV) and the previous with re-orienting event experiment (0.55 juV; F(\, 42) = 3.15,/? = 0.04; 1-tailed). This co-variation between the magnitude o f the reaction time effect and the P I and N l modulations suggests that these E R P effects may reflect the operation o f brain processes that contribute to the observed differences in reaction time. In contrast to the P I and N l effects, the N d effect was present at both parietal (F(l,24) = 9.1,/? < 0.01) and occipital (F(l,24) = 17.0,/? < 0.001) sites. Furthermore, the cross-experiment analyses revealed that the magnitudes o f these effects were not significantly affected by the presence or absence o f the re-orienting event (both p's>  0.30).  This result is consistent with the conclusion o f Chapter 3 that this component is not directly involved in the generation of the I O R effect.  4.5 Conclusions The present experiment and analyses demonstrate that elimination o f the re-orienting event both reduced the magnitude o f IOR and eliminated the P I and N l amplitude effects reported by Prime and Ward (2004) (Chapter 2). This co-variation of behavioural and E R P effects provides support for the view that I O R arises, at least in part, from changes in early perceptual processes. These findings are consistent with spatial attention research  demonstrating that changes in occipital P I and N l amplitude are reliably associated with changes in behavioural performance across a wide variety o f experimental paradigms (e.g. Mangun, 1995). These findings are also consistent with the model presented in Chapter 3.  78  No Re-orienting Event Experiment Target-locked LRP C3-C4  C1-C2 |  y.^rr  1  1  H——I—h  1 (_  Response-locked LRP C3-C4 I  / \  1 .y.*..4* \ r  1  Valid Invalid  C1-C2 r—*u  K~H—"t-L^  1.0 M V  H—i—h  +0  H  1 1 600  Valid  r -°  Invalid  I————  2  1  ^ 1  1  1  1  —  1  invalid Figure 4.1 Electrophysiological data, (a) Target- and response-locked lateralized readiness potentials (LRPs) for valid and invalid trials, (b) Event-related potentials (ERPs) to valid and invalid targets at posterior electrode sites. Vertical bars indicate target onset for target-locked waves and time of response for response-locked LRPs. Nd=negative difference. +  Q  6  0  Q  79  References Mangun, G . R. (1995). Neural mechanisms of visual selective attention. Psychophysiology, 32,4-18. Prime, D a v i d J., Visser, Troy A . W . , Ward, Lawrence M . (2004). Reorienting attention and inhibition of return. Article submitted for publication. Prime, D . J., & Ward, L . M . (2004). Inhibition of return from stimulus to response. Psychological Science, 15, 272-276.  80  CHAPTER V: Summary and Conclusions Posner, Rafal, Choate, and Vaughan (1985) proposed that inhibition o f return (IOR) serves to bias the visual system to acquire novel information at new locations. B y providing a possible function for IOR, this proposal has inspired both empirical research and theoretical debate (e.g.; K l e i n , 2000). However, the functional significance o f IOR cannot be definitively established until the mechanisms responsible for the effect are understood. For this reason, numerous empirical studies have investigated the factors responsible for causing I O R and the effect that these mechanisms have on target processing. The present results contribute to our understanding o f the mechanisms responsible for IOR by examining brain activity associated with both response generation and perceptual processing. The results presented in Chapters 2 and 3 provide strong evidence that, when measured with manual responses, I O R does not arise from an inhibition o f motor processes. Although caution must be exercised when interpreting null results, the present results demonstrate that changes in premotor processes are primarily, i f not solely, responsible for producing manual I O R in the most commonly used experimental tasks. This result contrasts with research indicating that, when measured with eye movement responses, I O R is associated with changes in oculomotor programming as well as saccadic reaction time (Ro, Pratt & Rafal, 2000). This difference provides additional support for the proposition that at least partly different mechanisms are responsible for manual and saccadic I O R (e.g.; Taylor & K l e i n , 2000; Hunt & Kingstone, 2003). Consistent with previous research, the current results also provide evidence that uninformative peripheral cues affect the amplitudes o f the early visual event-related potential (ERP) peaks. These E R P peaks are generally considered to represent neural activity associated with visual perceptual processing. The occipital PI and N I peaks reflect the  81 activity o f multiple extrastriate visual areas. Source localization techniques indicate that the occipital P I peak reflects the activity o f multiple areas of the ventral visual processing pathway responsible for the analysis of stimulus features (e.g.; D i Russo, Martinez, Sereno, Pitzalis & Hillyard, 2001; Clarke, Fan, & Hillyard; 1995). Although the anatomical loci o f the generators o f the occipital and parietal N l peaks have proven difficult to determine, both ventral occipital and parietal generators have been implicated (Di Russo et al., 2001). Further insight into the nature o f the brain processes underlying these E R P peaks can be gained by considering their latency. Categorization experiments have revealed that that E R P differences at latencies below 150 ms are related to differences in the visual properties o f the stimuli while E R P differences related to the task relevance of the stimuli emerge at longer latencies (e.g.; V a n Rullen & Thorpe, 2001). This indicates that the PI occurs during the time interval when early perceptual analysis of stimulus features is being performed. Whereas the N l occurs during a later stage of perceptual processing in which stimulus identity and task relevance begins to affect the neural response. In the present experiments the amplitudes o f the early occipital visual event-related potential peaks were smaller on valid-cue trials than on invalid-cue trials. These effects indicate that perceptual processing was inhibited on valid-cue trials relative to invalid-cue trials. A considerable amount o f spatial attention research has demonstrated that changes in occipital P I and N l amplitude are reliably associated with changes in behavioural performance across a wide variety o f experimental paradigms (e.g.; Mangun, 1995). This relationship suggests that changes in the perceptual processes indexed by these E R P peaks may be, at least in part, responsible for generating IOR. Providing additional support for a causal connection, the magnitude o f the E R P effects were found to co-vary with the magnitude o f I O R (Chapter 5).  82 Although the observed modulation o f the early occipital E R P peaks is consistent with an inhibition-of-attention account o f IOR, it is possible that these effects arise from changes in perceptual processing unrelated to attention orienting. A t the present time, neither the mechanisms o f attention nor the mechanisms of IOR are sufficiently well understood to definitively establish a link between attention orienting and IOR. Further E R P experiments examining both the facilitatory effect observed at short stimulus-onset-asynchronies (SOAs) and the later-occurring I O R effect are needed. In order to provide new information and address inconsistencies between previous studies, future E R P studies must systematically vary a variety o f stimulus and task variables. Inhibition-of-attention accounts o f I O R w i l l be supported i f future research demonstrates a correspondence between the E R P correlates of attentional facilitation and the E R P correlates o f IOR across a variety o f situations. A s discussed in Chapter 3, uninformative peripheral cues also give rise to unanticipated E R P modulations. These effects take the form o f negative displacements o f the E R P waveforms on valid-cue trials relative to those on invalid-cue trials. The neural origins and information processing functions represented by these negative difference (Nd) waves are unknown. Consequently, the relationships between these N d waves and behavioural performance are also unknown. Insight into the nature of these relationships, i f any, may be gained i f the neural origins of these N d waves could be determined. Unfortunately, the scalp distribution o f an E R P effect does not directly indicate the neural source o f the effect. Estimating the neural origins o f the N d waves observed in peripheral cueing studies w i l l require the use o f high-density electrode arrays and source analysis techniques. A variety o f different source analysis techniques exist (e.g.; Scherg, Vajsar, & Picton, 1989; Kineses et al., 2003) and each o f these techniques relies on simplified electrical models o f the human brain, skull and scalp. Regardless o f the particular technique used, E R P source analysis  83 involves constructing and testing hypothetical neural source models. These models are evaluated by calculating the degree to which they account for the observed scalp distribution. When the observed E R P effect is generated by a limited number of neural sources, source analysis techniques can be highly accurate. If the neural sources o f the N d effects can be accurately modeled, knowledge o f functional neuroanatomy obtained from functional imaging and intra-cranial electrophysiological recording could then be used to obtain a better understanding o f the information processing functions represented by these E R P effects.  Although the current results provide evidence against a motor locus o f I O R and in favor o f a perceptual locus, these results do not rule out the contribution o f post-perceptual decision and response selection processes. In fact, some o f the results o f the current study could be interpreted as supporting a decisional locus o f IOR. First, speed-accuracy tradeoffs were observed (Chapter 3). These effects may have arisen from differences in the decision criteria used on valid-cue trials and invalid-cue trials. Second, only the amplitudes and not the latencies of the early visual E R P peaks were affected by cue validity. This result is consistent with behavioural evidence that I O R does not affect the speed of transmission o f sensory information (e.g.; Maylor, 1985; Schmidt, 1996; Klein, Schmidt, & Miiller, 1998). The lack o f latency effects indicates that the speed o f perceptual processing represented by these peaks does not differ between cue conditions. This suggests that the actual differences in response times arise at later processing stages. A s proposed in the model presented in Chapter 3, changes in perceptual processing may affect the time necessary to select a response (or choose to initiate a pre-selected response) by changing the quality or strength o f the perceptual information available to the decision processes. In a random-walk type decision model (e.g.; Ratcliff & Rouder, 1998), for example, the time required to make a  84  response decision is determined, in part, by the rate of information accumulation available to the decision process. In terms of this type of model, the reduced neural response to the target on valid-cue trials indexed by the PI and N l peaks may be affecting response time by affecting the rate of evidence accumulation in a post-perceptual decision maker. Such an effect of cue validity on response selection processes would be consistent with the results of a variety of studies that have found that IOR interacts with several experimental effects that are thought to arise from conflict at the response selection stage. These effects include, the Simon effect (Ivanoff, Klein, & Lupianez, 2002), the presence of a nonresponding effector (Ivanoff & Klein, 2001), the Stroop effect (Vivas & Fuentes, 2001), and semantic priming and flanker interference (Fuentes, Vivas, & Humphreys, 1999). Again, further research is necessary to explore these possibilities. Although the current results provide new insights into the mechanisms underlying IOR, it is clear that significantly more research is needed. Some authors have expressed surprise that there has been such difficulty in explaining the results arising from such a simple experimental paradigm. However, the brain is a complex system regardless of the simplicity of the paradigms with which we choose to investigate it. The brain's complexity has evolved to allow us to function in complex environments and the results obtained in our laboratory experiments reflect this complexity. In fact, our interest in cognitive processes stems from our desire to understand real world functioning, not simply to understand the results of our experiments. Until the mechanisms of IOR are better understood, the real world significance of this laboratory phenomenon cannot be definitively established.  85  References Clarke, V . P., Fan, S., Hillyard, S. A . (1995). Identification of early visual evoked potential generators by retinotopic and topographic analyses. Human Brain Mapping, 2, 170187. D i Russo, F., Martinez, A . , Sereno, M . I., Pitzalis, S., & Hillyard, S. A . (2001). Cortical sources o f the early components o f the visual evoked potential. Human Brain Mapping, 15, 95-111. Fuentes, L . J., Vivas, A . B . , & Humphreys, G . W . (1999). Inhibitory tagging o f stimulus properties in inhibition of return: Effects on semantic priming and flanker interference. The Quarterly Journal o f Experimental Psychology, 52A, 149-164. Hunt, A . R., & Kingstone, A . (2003). Inhibition o f return: Dissociating attentional and oculomotor components. Journal o f Experimental Psychology: Human Perception and Performance, 29. 1068-1074. Ivanoff, J., & K l e i n , R. M . (2001). The presence o f a nonresponding effector increases inhibition of return. Psychonomic Bulletin & Review, 8, 307-314. Ivanoff, J., K l e i n , R. M . , & Lupianez, J. (2002). Inhibition o f return interacts with the simon effect: A n omnibus analysis and its implications. Perception & Psychophysics, 64, 318-327. Kineses, W . E . , Braun, C , Kaiser, S., Grodd, W . , Ackermann, H . , & Mathiak, K . (2003). Reconstruction o f cortical sources for E E G and M E G based on a monte-carlomarkov-chain estimator. Human Brain Mapping, 18, 100-110. K l e i n , R. (2000). Inhibition of return. Trends in Cognitive Sciences, 4(4), 138-147. K l i e n , R. M . , Schmidt, W . C , & Miiller, H . J. (1998). Disinhibition o f return: Unnecessary and unlikely. Perception & Psychophysics, 60, 862-872; Mangun, G . R. (1995). Neural mechanisms o f visual selective attention. Psychophysiology, 32, 4-18. Maylor, E . A . (1985). Facilitatory and inhibitory components o f orienting in visual space. In M . I. Posner & O. S. M . M a r i n (Eds.), Mechanisms o f Attention: Attention and Performance X I . Hillsdale, N e w Jersey: Lawrence Erlbaum Associates. Posner, M . I., Rafal, R. D . , Choate, L . S., & Vaughan, J. (1985). Inhibition o f return: Neural Basis and Function. Cognitive Neuropsychology, 2, 211-228. Ratcliff, R., & Rouder, J. N . (1998). Modeling response times for two-choice decisions. Psychological science, 9, 347-356. Ro, T., Pratt, J., & Rafal, R. D . (2000). Inhibition o f return in saccadic eye movements. Experimental Brain Research, 130, 264-268.  86 Scherg, M . S., Vajsar, J., & Picton, T. W . (1989). A source analysis o f the late human auditory evoked potentials. Journal of Cognitive Neuroscience, 1(4), 336 - 355. Schmidt, W . C . (1996). Inhibition o f return is not detected using illusionary line motion. Perception & Psychophysics, 58, 883-898. Taylor, T. L . , & K l e i n , R. M . (2000). Visual and motor effects in inhibition o f return. Journal of Experimental Psychology: Human Perception and Performance, 26, 1639-1656. V a n Rullen, R., & Thorpe, S. J. (2001). The time course o f visual processing: From early perception to decision-making. Journal o f Cognitive Neuroscience, 13, 454-461. Vivas, A . B . , & Fuentes, L . J. (2001). Stroop interference is affected i n inhibition o f return. Psychonomic Bulletin & Review, 8, 315-323.  Contributions of the Author Chapters 2, 3, and 4 of this thesis are versions of manuscripts that arouse from the collaboration of the candidate, David J. Prime, and his supervisor, Lawrence M. Ward. All of the research hypotheses were initially conceived by the candidate. The candidate designed, conducted and analyzed all the experiments presented in this thesis. In addition, the candidate was the primary author of all the chapters in this thesis. Lawrence M. Ward revised and edited Chapters 2, 3, and 4 and contributed to the Discussion section of Chapter 3. The above statement was written by the candidate, and agreed upon by the undersigned.  

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.831.1-0092368/manifest

Comment

Related Items