Visual Cognition

ISSN: 1350-6285 (Print) 1464-0716 (Online) Journal homepage: http://www.tandfonline.com/loi/pvis20

Investigating the parameters of transsaccadic memory: inhibition of return impedes information acquisition near a saccade target

Martijn J. Schut, Jasper H. Fabius & Stefan Van der Stigchel

To cite this article: Martijn J. Schut, Jasper H. Fabius & Stefan Van der Stigchel (2016) Investigating the parameters of transsaccadic memory: inhibition of return impedes information acquisition near a saccade target, Visual Cognition, 24:2, 141-154, DOI: 10.1080/13506285.2016.1206050

To link to this article: http://dx.doi.org/10.1080/13506285.2016.1206050

Published online: 18 Jul 2016.

Submit your article to this journal

Article views: 44

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=pvis20

Download by: [University Library Utrecht] Date: 20 September 2016, At: 00:11 VISUAL COGNITION, 2016 VOL. 24, NO. 2, 141–154 http://dx.doi.org/10.1080/13506285.2016.1206050

Investigating the parameters of transsaccadic memory: inhibition of return impedes information acquisition near a saccade target Martijn J. Schut, Jasper H. Fabius and Stefan Van der Stigchel Experimental Psychology, Helmholtz Institute, Utrecht University, Utrecht, The Netherlands

ABSTRACT ARTICLE HISTORY A limited amount of visual information is retained between saccades, which is subsequently stored Received 7 March 2016 into a memory system, such as transsaccadic memory. Since the capacity of transsaccadic memory Revised 13 June 2016 is limited, selection of information is crucial. Selection of relevant information is modulated by Accepted 15 June 2016 attentional processes such as the presaccadic shift of attention. This involuntary shift of attention KEYWORDS occurs prior to execution of the saccade and leads to information acquisition at an intended fl Inhibition of return; saccade target. The aim of the present study was to investigate the in uence that another transsaccadic memory; attentional effect, inhibition of return (IOR), has on the information that gets stored into presaccadic shift of attention transsaccadic memory. IOR is the phenomenon where participants are slower to respond to a cue at a previously attended location. To this end, we used a transsaccadic memory paradigm in which stimuli, oriented on a horizontal axis relative to saccade direction, are only visible to the participant before executing a saccade. Previous research showed that items in close proximity to a saccade target are likely to be reported more accurately. In our current study, participants were cued to fixate one of the stimulus locations and subsequently refixated the centre fixation point before executing the transsaccadic memory task. Results indicate that information at a location near a saccade landing point is less likely to be acquired into transsaccadic memory when this location was previously associated with IOR. Furthermore, we found evidence which implicates a reduction of the overall amount of elements retained in transsaccadic memory when a location near a saccade target is associated with IOR. These results suggest that the presaccadic shift of attention may be modulated by IOR and thereby reduces information

acquisition by transsaccadic memory.

The way our visual system deals with the challenge of The crucial factor to select relevant information for processing visual information across eye movements storage and subsequently store this information is (known as saccades) is both complex and fascinating. visual attention (Irwin & Gordon, 1998; Prime et al., While viewing a scene, the eyes saccade about three 2007). Johnson et al. (2008) concluded that visual times per second to foveate specific parts of the attention plays a general role in both maintaining scene for high acuity processing (Hollingworth & Hen- feature memory and binding features into represen- derson, 1998). One of the challenges of integrating tations of objects. After being selected, visual infor- visual information is the inability of the visual system mation is acquired in a memory storage, such as to process all information simultaneously (Wolfe, transsaccadic memory. Transsaccadic memory can 1994). Selection of visual information is therefore be loaded as a result from selection by endogenously crucial as we are only able to attend to a small orientated attention and by automatic, exogenous amount of information at any given time and unat- attention. For instance, before executing an eye move- tended information is lost (Wolfe, Reinecke, & Brawn, ment attention is shifted to the intended location of 2006). Loss of information can occur during different the eye movement (Deubel & Schneider, 1996; stages of visual processing and is especially true Hoffman & Subramaniam, 1995). This presaccadic whilst executing a saccade. Between saccades a shift of attention is thought to be involuntary and limited amount of objects are retained in a memory mandatory. Information at the intended target trace, known as transsaccadic memory (Bays & location of an eye movement is attended before the Husain, 2008; Irwin, 1991; Irwin & Gordon, 1998; movement is executed, due to the presaccadic shift Prime, Tsotsos, Keith, & Crawford, 2007). of attention (Deubel & Schneider, 1996; Irwin &

CONTACT Martijn J. Schut [email protected] Experimental Psychology, Helmholtz Institute, Heidelberglaan 1, Utrecht 3584 CS, The Netherlands © 2016 Informa UK Limited, trading as Taylor & Francis Group 142 M. J. SCHUT ET AL.

Gordon, 1998). How the presaccadic shift of attention by IOR. Under normal circumstances, information in affects transsaccadic memory has been explored in the near proximity of a saccade target is typically the paradigm used by Irwin and Andrews (1996). In acquired into transsaccadic memory. We hypothesize this paradigm a stimulus array is presented before that when information is located near a saccade executing an eye movement. As the eye movement target at a location that is associated with IOR, that is cued, and subsequently executed, the stimulus is this information would be acquired into transsaccadic removed. Results showed that the information in memory to a lesser extent. As a consequence, infor- near proximity to the intended saccade target is mation at that particular location should be reported better acquired than information away from a less accurately. To test this hypothesis we used an saccade target, due to the presaccadic shift of adapted version of the original transsaccadic attention. memory paradigm by Irwin and Gordon (1998) Besides the presaccadic shift of attention, other where one of the stimulus locations was previously attentional processes may modulate information fixated by the participants. This paradigm allowed us acquisition by transsaccadic memory, yet these to investigate how IOR influences the overall amount factors still remain unknown. One particular candidate of items acquired in transsaccadic memory. If only is inhibition of return (IOR; see Posner & Cohen, 1984). the information that appears at a saccade goal is IOR has been described as an attentional effect where acquired due to presaccadic shift of attention and a slowed reaction time is observed to information at a IOR would reduce information acquisition we should previously attended location (Pratt, Hillis, & Gold, observe a reduction in overall number of items 2001). In the IOR paradigm by Posner and Cohen acquired by transsaccadic memory. However, if obser- (1984) the participants are presented with a cue, fol- vers have the ability to compensate for this proposed lowed by a target to which the participant responds. loss of information, by more accurately acquiring The presented cue is uninformative and draws atten- stimuli further away from the saccade target, the tion by its onset (exogenous cue). When the interval hypothesized effect of IOR would not decrease

between cue presentation and target is short, a facili- overall transsaccadic memory capacity. tation of reaction time is observed. However, when this interval is relatively long a slowed reaction time is observed, resulting from attentional disengagement Experiment 1 (Castel, Pratt, & Craik, 2003; Posner & Cohen, 1984). Method Other than reaction time, IOR may also affect Participants sensory-perceptual processes and thereby influence performance on certain discrimination tasks (Reuter- Fourteen participants, 10 female, aged 19 to 35 (M = Lorenz, Jha, & Rosenquist, 1996). Research has indi- 23.3) from the Utrecht University community partici- cated a reduced ability for human observers to dis- pated for monetary compensation of 6 Euros per criminate contrast differences at a location hour. Participants completed 500 trials in a two and associated with IOR (Sapir, Jackson, Butler, Paul, & a half hour session. All participants reported normal Abrams, 2013). Yet, it is currently unclear how IOR or corrected-to-normal vision and were naïve to the modulates information acquisition across saccades purpose of the study. Written informed consent was by transsaccadic memory. Converging evidence obtained from all participants. The study was reviewed suggests that IOR is closely related to the oculomotor and approved by the Faculty Research Ethics Commit- system and that IOR and oculomotor activation are tee (FETC) of the University of Utrecht. integrally linked together (Hilchey, Klein, & Satel, 2014). As acquisition of information into transsaccadic Stimuli and apparatus memory is dependent on attentional processes such Stimuli were white letters (font height 1° visual angle) as the presaccadic shift of attention, we inferred that on a black background. The letters used were drawn transsaccadic memory and IOR may also interact at a from a subset of 12 letters (A, S, D, F, G, H, J, K, L, V, target discrimination level. B and N) with no repeating letters present in one The aim of the current study was to investigate how trial. Fixation dots and probes were presented as information acquisition across saccades is modulated white circles of 0.6° with a centred black circle of VISUAL COGNITION 143

0.1°. The fixation dot was placed in the centre of the Python 2.7 and R 3.1.3 was used for statistical pro- screen. Stimuli and probes could appear at 10 cedures (Ihaka & Gentleman, 1996). The lme4 R locations, placed on two imaginary horizontal axes package was used for statistical analyses (Kuznetsova, (3° upwards and downwards from fixation; Figure 1). Brockhoff, & Christensen, 2013). The stimuli and probes were evenly spaced along these horizontal axes subtending 4° left and right from fixation. During the experiment two saccade Procedure targets appeared along the horizontal axis on the The experiment was divided into 10 blocks of 50 trials. same height as the fixation target, also subtending Prior to starting the experiment the eyetracker was cali- 4° left and right. brated using a standard 9-point calibration procedure. The experiment was performed in a darkened room. After that a practice block of 15 trials was initiated. If the Stimuli were presented on a LG 24MB65PM LCD participant did not perform correct saccades (within 2° monitor with a spatial resolution of 1280 by 800 of the saccade targets) on at least five trials the practice pixels and a refresh rate of 60 Hz. The screen size was block was repeated (one participant). Practice trials 50.8 cm wide and 33.9 cm high. The participant was were identical to regular trials. Before every trial a seated 70 cm from the monitor. Participants placed drift check was performed. If drift was greater than their heads on a desk-mounted chin rest to reduce 1.5° visual angle, the eye tracker was recalibrated. head movement. Eye movements were tracked with The procedure of the experiment is depicted in an Eyelink 1000 (SR Research Ltd., Canada) sampling Figure 2. Participants fixated on a central stimulus at 1000 Hz. All participants were calibrated using a 9- for 750, 1000 or 1250 ms. The variable interval was point calibration procedure. For offline saccade detec- used to prevent pre-emptive saccades. A probe tion the SR research algorithm was used, where an eye would then appear at one of 10 locations. This location movement was classified as a saccade when the move- will be referred to as the onset location. Participants ment velocity exceeded 35°/s or when the movement were instructed to make a saccade to this stimulus 2 acceleration was greater than 9500°/s . The left eye as quickly as they could in response to its onset. was tracked for all participants. When the stimulus disappeared, participants had The paradigm was programmed in Python 2.7. The 500 ms to make a saccade back to the fixation point. Pygaze library was used to connect to the eye tracker After this, the letter stimulus array and the saccade using the native Pygaze event detection to detect sac- targets appeared. After a variable interval of 750 to cades online (Dalmaijer, Mathôt, & Van der Stigchel, 1250 ms a tone was played for 400 ms, alerting partici- 2013). Eyetracker data files were analyzed with pants to initiate a saccade. In half of the trials this was a low tone of 200 Hz, instructing a leftward saccade. In the other half the tone was a high tone of 1000 Hz, instructing a rightward saccade. When the saccade was completed, a period of 500 ms passed, after which a target probe briefly appeared at one of the 10 locations for 100 ms. The location of this probe will be referred to as the answer location. Participants then typed in their answer by keyboard. Responses were shown on screen and could be changed until participants confirmed their answer by pressing the Enter key or the Space bar. The subset of letters was chosen due to the location of the letters on the key- board, which eased the difficulty of typing in the dark. To further facilitate this, participants received Figure 1. The possible locations of the stimuli used in Exper- feedback on their performance at the end of every iment 1 and Experiment 2. The dashed circles are placeholders block of trials. To reduce fatigue in participants, at and were not presented during the experiment. The figure pre- sented here is not drawn to scale and is shown in inverted the end of every block of trials the participants had contrast. the opportunity to take a short break. We also gave 144 M. J. SCHUT ET AL.

Figure 2. Sequence of events as they appeared in Experiment 1. Dashed lines with arrows indicate eye movement and the eye graphic indicates eye position. The tone played is either a high tone, indicating a saccade to the right, or a low tone, indicating a saccade to the left. The panels in the figure are presented in inverted contrast and are not drawn to scale. the participants a 10 minute break outside of the lab any saccades missing their target by more than 2° or when they were halfway through the experiment. by participants not returning to the central fixation Combinations of onset location and answer dot after fixating the onset position. Trials were exe- location were counterbalanced per participant and cuted on the basis of saccade latencies using a recur- appeared equally often, but in quasi-random order. sive trimming procedure (Van Selst & Jolicoeur, 1994). Thus, there was a 10% chance that the answer location Saccade latencies deviating more than 2.5 SD within and the onset location were the same. Participants the participant’s median latency were excluded. Two were informed that probes were equally likely to participants were excluded from further analysis due appear at any location. The location of the onset was to not meeting these criteria on more than 50% of therefore completely uninformative of the location their trials. Of the remaining participants, 14.2% of of the answer. Right and leftward saccades were also the trials were excluded due to saccades not landing counterbalanced with the combination of onset and within 2° of a saccade target, either the cued location, fi answer locations. Lastly, the stimuli could not appear saccade back to xation or to saccade target. Another more than twice in succession at the same location. 7.5% of the trials were excluded on the basis of Locations of onset and answers were re-randomized saccade latencies. if an experiment was generated in which stimuli did Answer locations were collapsed vertically to appear more than twice at the same location in analyze the effect of the location of the onset relative succession. to the saccade direction. This collapsing was per- To determine whether IOR would be induced at the formed after tagging the trials as either onset and onset location, as expressed in prolonged reaction answer congruent or incongruent trials, to ensure times, we performed a pilot study with a different that only the trials in which the participants fixated set of participants. The same locations, timing and the exact same location as where the answer stimuli were used as in Experiment 1. One saccade appeared were tagged as target congruent trials. By was made to an onset location and subsequently collapsing the positions of the answer in relation to back to fixation. After this, the stimulus array appeared the saccade target, two stimuli appearing on the with one target letter highlighted. The letter was high- same vertical axis were encoded as one answer lighted using a grey circle (2° diameter). Participants location to a total of five answer locations. The were encouraged to type a letter as quickly as poss- locations were encoded relative to the saccade ible. A slowed response of 30.6 ms (SD = 125.6 ms) target: Location 1 being the two answer stimuli was observed for stimuli that were previously cued, appearing above and below the saccade target, in F(1, 11) = 5.09, p < .05. We therefore concluded that equal steps up to Location 5 (the two stimuli furthest IOR, in terms of prolonged reaction times, could be away from the saccade target). The performance is induced with the onset cues used in this paradigm. expressed as a proportion of correct answers. Statistical analyses included a repeated measures Data analysis analysis of variance and a generalized linear mixed Analyses and associated data are available online (GLM) model. The repeated measures analysis of var- (Schut, 2016). Trials were excluded on the basis of iance used the mean proportion of answers correct VISUAL COGNITION 145 as dependent variable. Independent variables were was cued by tone to a saccade target) as fixed effects location of the answer (five locations, from closest to to the GLM model to test their predictive qualities. saccade target to furthest from saccade target) and Saccade latencies were similarly controlled for and whether the location of the answer probe had been investigated using a repeated measures analysis of previously fixated (either true or false). A significance variance. criterion of p < .05 was used for these analyses, additionally effect sizes are reported as eta squared Results (η2). Post-hoc tests consisted of Holm-Bonferroni cor- rected paired t-tests, Cohen’s d is reported as a The main results found are shown in the Data points of measure of effect size for post-hoc analyses. Figure 3. A main effect was found for answer location GLM analyses allowed us to examine the fit of the in relation to saccade target F(4, 44) = 11.84, p < .01, participant data to the model, as analyzing pro- η2 = 0.29. Post-hoc tests exploring the main effect for portions per participant may be less informative for location show that participants were more accurate the small amount (10) of critical trials per participant at the location closest to the saccade target (Location in this experiment (Bell et al., 2010; Kenward, Roger, 1: M = 0.49, SD = 0.50) as compared to locations in the Process, Lodge, & Ox, 2013; Lindstrom & Bates, middle of the array: Location 2 (M = 0.19, SD = 0.39), 1990). GLM models may be a stronger approach in Location 3 (M = 0.17, SD = 0.38) and Location 4 (M = examining participant variance than using pro- 0.21, SD = 0.41), p < .01, 0.67 < d ≤ 0.72. Participants portions, as trial by trial data is not collapsed in a were more accurate at the location furthest from the GLM (like in traditional statistics), but added as a saccade target as compared to the locations in the random effect to the model. As a result, GLM is less middle of the stimulus array, p < .05, 0.39 < d ≤ 0.44. sensitive to unbalanced data than a type 3 sum of Importantly, participants were more accurate at squares analysis of variance (Cnaan, Laird, & Slasor, Location 1 (the location closest to the saccade 2014; Lindstrom & Bates, 1990). The GLM model target) than at Location 5 (furthest from the saccade

included correct answers per trial as dependent vari- target, M = 0.36, SD = 0.48), p < .01, d = 0.27. able, with location of answer probe as a second No main effect was found for onset location F(1, 11) degree polynomial and answer probe appearing at = 1.93, p = .19, η2 = 0.02. An interaction effect was previously fixated location as fixed effects. The present for answer location and onset location F(4, second degree polynomial was added as the response 44) = 6.14, p < .01, η2 = 0.09. Post-hoc tests compared pattern in the task used typically shows a U shape, the onset location with each answer location in where end-point locations are reported better than relation to the saccade target. Visually examining the central locations (Irwin & Gordon, 1998). A random data (as shown in Figure 3) suggests that there is an intercept per participant was added to the model, to effect present for the location closest to the saccade account for variance per observer. Significance of target. The interaction effect comparing performance fixed effects were tested by a z test of which the p for answer at Location 1 with onset at the same values are reported. Lastly, the model was compared location and answer at Location 1 with onset at a to a null GLM model, which excluded any interaction different location was significant (difference in mean between fixed effects. A chi square test was used to proportion = −0.21, p = .04, d = 0.40). Comparisons test the two models against each other and Bayesian for the other four locations (e.g., answer at location Information Criteria (BIC) were reported for both x, onset at exact same location or at a different models. location) did not show significance (ranging all ps As saccades are somewhat inaccurate (Coëffé & > .8, .02 < d ≤ 0.17). These results suggest that only O’Regan, 1987; Cohen & Ross, 1978) and foveated the information closest to the saccade target is signifi- points thus differ between trials, we controlled for cantly less acquired by prior orienting to the onset. An saccade landing and starting position. It may be poss- additional post-hoc analysis shows that performance ible that these properties that differ between trials at Location 1 is significantly better than at Location would bias which information is acquired. To this 5 in trials where the onset and probe were presented end, we added the deviation from the perfect starting at different locations, p < .01, Cohen’s d = 0.27. In con- and landing position (with regards to the saccade that trast, in trials in which the onset and the probe 146 M. J. SCHUT ET AL.

Figure 3. Generalized linear mixed model and mean data for mean proportion of answers correct in relation to saccade target between onset appearing at same location as the answer and onset not appearing at the same location as the answer. The shaded region rep- resents the 95% confidence interval for the model, error bars around the data points show standard error. The x axis shows an illus- tration of where the answer probe appeared. The dashed black line shows performance at chance level (0.08). In the figure the two conditions are horizontally offset as to not overlap visually. appeared at the same location there was no significant saccade target. A logistic regression on the participant performance difference between Location 1 and data suggests that half of the participants show signifi- Location 5, p = .18, d = 0.12. Therefore, participants cant reduction in their performance when onset and only scored significantly higher at the location probe were presented at the same location, p < .05, closest to the saccade target in trials in which the as shown in Figure 4. To further quantify the contri- answer probe was not presented at a previously bution of individual participants a GLM model was fixated location. fit, the results of which are shown Model points of Data for individual participants is shown in Figure 4. Figure 3. This model shows significance for the fixed This figure shows individual scores for the interaction effects: answer probe at same location as the onset, effect we found, which is the location closest to the z = 3.73, p < .01 and probe location in relation to

Figure 4. Mean proportion of answers correct per participant at the location closest to the saccade target. Each orange and blue bar, separated by ticks at the bottom of the graph, represents one participant. Significant differences between the onset appearing at the location at the onset or a new location are indicated by asterisk for each participant. For visualization purposes, error bars are standard error calculated from a binomial probability function. The dashed black line is performance at chance level (0.08). VISUAL COGNITION 147 saccade target, z = −7.89, p < .01. The interaction loss of information was always present when the effect between the two fixed factors were similarly sig- onset appeared near the saccade target. This assump- nificant, z = −2.51, p < .01. We found that this model tion had to be made due to restrictions in our data set, (BIC = 5435) significantly outperformed a null model as only one answer/onset location combination was (BIC = 5910) in which no interaction effects were reported per trial. The onset position was recoded to present, χ2 = 474.79, p < .01. Thus, the results from be either near the saccade target (onset appeared at this analysis converge with our previous results. Visu- Location 1) or away from it (onset appeared at ally inspecting the mean data suggests that the Locations 2 to 5). This allowed us to examine model fits the data well. Analyzing the fit of this whether Locations 2 through 5 received an increase model per participant indicated that six out of 12 par- in accuracy from the onset appearing near the ticipants significantly fitted the model (odds ratio > saccade target. To analyze the effect of the recoded 1.2). Furthermore, investigating the fit of the partici- onset position on accuracy we used a 2 × 5 repeated pants for the location closest to the saccade target measures analysis of variance as before. We entered reveals that seven out of 12 participants fit the two factors into the analysis: the recoded onset model for this location (odds ratio > 1.2). location (2 levels: onset at Location 1 or onset at To examine whether saccade starting and landing Location 2 through 5) and location of answer probe positions biased the information that was reported (Locations 1 through 5). As before, the analysis in any way we tested our previously constructed showed a main effect for location F(4,44) = 12.41, p GLM model to a model that also included saccade < .01, η2 = 0.38. No main effect was observed for the starting and landing position as fixed effects. We recoded onset probe locations, F(1,11) = 1.03, p = .33, found that the GLM model without saccade landing η2 < 0.01. Lastly, no interaction effect was identified position or starting position (BIC = 5435), as used in between the recoded onset probe locations and the the previous analysis, outperformed the models that answer locations, F(4,44) = 0.47, p = .90, η2 = 0.01 included saccade landing (BIC = 5890) or saccade suggesting that no difference was observed when

starting position (BIC = 5880). Thus, we conclude that onset appeared at the location closest to the the actual starting and landing position of the saccade target or further away from it in terms of saccade to a cued saccade target did not affect the overall accuracy for information at Locations 2 to ability to report information near that saccade target. 5. Therefore, locations further from the saccade To control for saccade latency we analyzed the target received no compensation in information effect of onset location on saccade onset. As IOR acquisition from the reduction of information acqui- may affect saccade latency (Briand, Larrison, & sition after the onset appeared near the saccade Sereno, 2000; Klein & MacInnes, 1999) it was crucial target. to exclude a possible interaction between delayed sac- cades towards a location associated with IOR. No indi- Discussion cations of an effect was found for onset location on saccade latency, F(4, 44) = 0.54, p = .48, η2 < 0.01. In this experiment we studied the association between Median latency for Location 1 was 314 ms with a stan- IOR and transsaccadic memory using an adapted dard deviation of 328 ms, average latency for the transsaccadic memory paradigm previously used by other locations was 325 ms with a standard deviation Irwin and Andrews (1996). We hypothesized that IOR of 333 ms. would interact with transsaccadic memory by redu- To follow-up on our findings, we investigated cing the information acquisition at a previously cued whether the reduction of accuracy, when answers location. Our results indicate an interaction between and onset appeared in close proximity to the IOR and transsaccadic memory: we found a reduction saccade target, would also mean that less elements of correct answers for elements associated with IOR were retained overall. Our results thus far imply that compared to elements that were not associated with when the onset appeared near the saccade target a IOR. This effect was only present for the elements at reduction of accuracy is shown for elements that the closest proximity to the saccade target and was appear at the same location as the onset. Thus, we not present for elements further from the saccade recoded the onset position, and assumed that this target. This indicates that processes prior to the 148 M. J. SCHUT ET AL. execution of the saccade were affected by IOR which as early research has shown that task demands may resulted in a reduction of accuracy for elements near dictate the size of the attentional focus (Castiello & a saccade target. Umiltà, 1990). Thus, we have shown that the presacca- In the current experiment IOR only affected trans- dic shift of attention can at least have an asymmetric saccadic memory for elements in near proximity to a spread of 3.0° with the task described in our current saccade target. To explore how this interaction study. affects the overall memory capacity of transsaccadic In this paradigm, overall target discrimination was memory we recoded our data relative to the location best at the location closest to the saccade target. where the onset appeared. In this manner, we could This effect has been previously observed and attribu- investigate whether the less accurately reported infor- ted to the presaccadic shift of attention in the study mation, that was previously fixated, was compensated by Irwin and Andrews (1996). However, the proportion for by higher accuracy for items that were not pre- of correct answers on the location furthest away from viously fixated. Our results show that no benefitwas the saccade target was also better than for stimuli that observed for the other locations when the information appeared in the middle of the stimulus array. We near the saccade target was previously cued and thus suggest that these results might have been due to less correctly reported. These results suggest that the the design of the paradigm. To reiterate, in the reduction of accuracy, when the onset and stimuli current study the stimulus array was shown after sac- both appeared at the same location close to the cading to a cued location. The stimulus array and saccade target, did not relocate attentional resources saccade targets were then visible until a saccade was to other locations. executed to a saccade target. These saccades were IOR has been associated with increasing in saccade cued with a tone that had a mean onset of 1000 ms latency when saccading to a previously cued location after stimulus presentation. The participants may (Briand et al., 2000). For this reason we wanted to have been pre-emptively attending both saccade examine the possibility that our results could be targets before the onset of the tone, as the saccade

explained through differences in saccade latencies. targets were relevant to perform the task correctly. IOR being associated with a location within the stimu- This form of attending in preparation of the tone lus array did not cause delays in saccade execution to would explain the increase in performance at locations the saccade targets and does therefore not add any furthest from saccade target as compared to stimuli explanatory value to our results. Similarly, we investi- that appeared in the middle of the array. Our results gated whether information acquisition in the para- show that overall performance was significantly digm used influenced by saccade landing and higher at the location closest to the saccade target starting position. We found no such effect, which is when compared to the location furthest from the in line with the studies indicating that the presaccadic saccade target. This indicates a possible added shift of attention moves to the intended location of an benefit in memory acquisition of executing the eye movement, rather than the actual landing position saccade. Interestingly, no effect was present for (Deubel & Schneider, 1996; Van der Stigchel & De target discrimination when onset and answer Vries, 2015). appeared furthest from the saccade target and at We note that in the previous implementations of the same location. Therefore, we postulate that IOR the paradigm (Irwin, 1991; Irwin & Andrews, 1996; may affect attentional processes linked to the Irwin & Gordon, 1998) the stimuli to be attended to saccade, such as the presaccadic shift of attention, via the presaccadic shift of attention were located in by disrupting the benefit in information acquisition closer proximity than in our current study. Although of executing a saccade to the saccade target. the distance between the saccade target and the However, our model of the effect may be supported stimuli were increased (from a maximum of 2.2° in by the similar properties between the presaccadic shift the original paradigm to 3.0° in the current paradigm) of attention and exogenously captured attention. IOR we found a pattern of results that matched the results has been associated with exogenous cueing (Posner & found previously (i.e., information near saccade end- Cohen, 1984), which directs attention automatically. points is better reported when blanked during sacca- Similarly, the presaccadic shift of attention is drawn dic suppression). This effect is not entirely unexpected, involuntarily to a saccade location. The area that the VISUAL COGNITION 149 presaccadic shift of attention is directed to has a range array are less accurately reported and may therefore that is bigger than the target itself (Deubel & Schnei- be less attended to, due to not receiving a presaccadic der, 1996; Irwin & Andrews, 1996; Irwin & Gordon, attentional benefit. We propose that flanker stimuli 1998). Through this, the participants were better will not receive this presaccadic benefit in a similar able to report information not only at the saccade manner and therefore not interfere with performance target, but also near a saccade target. The increased in this paradigm. However, visual crowding has been performance for stimuli near the saccade target linked to reducing the ability to identify stimuli suggests that the location that was cued and sub- (Whitney & Levi, 2011) and thus adding more stimuli sequently associated with IOR also falls within this to the array would reduce the accuracy for the range of attention. Research has shown that the size elements at the end positions. To test for this possi- and shape of the attentional window can be modu- bility we added four flanker stimuli next to the end lated and can contain fields in which information is positions in the stimulus array. unattended (Castiello & Umiltà, 1990; Müller & Hübner, 2002). This leads us to postulate that infor- mation acquisition in transsaccadic memory may be Experiment 2 affected by IOR through inhibition of the presaccadic Method shift of attention. We do note that IOR is an effect Participants and procedure which may be disentangled in a motor and attentional component (Kingstone & Pratt, 1999; Martín-Arévalo, The procedure for the second experiment was identi- Kingstone, & Lupiáñez, 2012). In the current study cal to Experiment 1 except for the following details. IOR was induced by the participants executing a The study consisted of 12 participants from the saccade to a probe, rather than through attentional Utrecht University community (11 female) aged 18 orienting alone. Although our current design would to 25, M = 21.8, that completed 120 trials in one 30 most likely elicit both attentional and oculomotor minute session. We chose this amount of trials for fi Experiment 2 to closely match the amount of critical IOR, it is possible that the effect found may be speci c to IOR induced by executing a saccade to a location. trials in Experiment 1. All participants reported It should be noted that we base our conclusions on normal or corrected-to-normal vision. a relative effect that is the combination of two effects. The procedure used for the experiment is shown in First, we observed an increase in performance near the Figure 5. The main difference is that we only included saccade target relative to the performance furthest a crowding condition, where in half of the trials four from the saccade target, which may be attributed to additional stimuli appeared next to the end point presaccadic attentional processes. Second, we stimuli. In the remaining half of the trials the four observed an absence of this increase when IOR was additional stimuli did not appear. The crowding induced near the saccade target that we suggest stimuli were interchanging between the letter T and affects the initial presaccadic shift of attention. To the letter Q. We opted for these crowding stimuli as verify the validity of our conclusions, we wanted to they are highly similar to the target stimuli (a necessity ensure that the observed increase in performance for crowding to occur; Kooi, Toet, Tripathy, & Levi, near the saccade targets was not due to confounding 1994; Levi, 2008; Whitney & Levi, 2011). With regards factors. As noted by Irwin and Gordon (1998), perform- to Experiment 1, the onset stimulus has been ance is generally increased for targets at the end pos- removed and thereby the IOR condition. Timing for itions in this task and in linear arrays in general. Irwin the experiment remained unchanged as compared and Gordon argue that this advantage is due to less to Experiment 1. visual crowding, since the stimuli at the endpoints have only one stimulus besides them. As we wanted Data analysis to control for this possibility, a follow-up experiment Trials were excluded on basis of the saccade missing was performed to examine the effects of flanking its target by more than 2° or saccade latencies stimuli on the observed increase in performance greater than 2.5 standard deviations from the near the saccade target. Our results thus far indicate median saccade latency. By these criteria 10.8% of that stimuli that are in the middle of the stimulus the trials were excluded. Similar to Experiment 1, 150 M. J. SCHUT ET AL.

Figure 5. Sequence of events as they appeared in Experiment 2. Dashed lines with arrows indicate eye movement and the eye graphic indicates eye position. Top panel second from the left shows an example trial with crowding stimuli. The figure is presented in inverted contrast and is not drawn to scale. statistical analyses included a GLM model and a for location of answer probe, z = 1.99, p = .04, the repeated measures analysis of variance. The fixed repeated measures analysis confirmed this, F(1,11) = factors in the GLM model were location (five levels, 11.69, p < .01, η2 = 0.08. Post-hoc paired t-tests show from closest to saccade target to furthest) and that the proportion of letters correct is higher at the whether the trial contained crowding stimuli or not location closest to the saccade target (Location 1, M (two levels). A random effect was added for partici- = 0.51, SD = 0.50) as compared to locations further pant. For analyses between locations we used Holm- from the saccade target (Locations 2 through 4, M = Bonferroni corrected paired t-tests. 0.24, SD = 0.43) with p < .01, d = 0.58. Moreover, per- formance at the location furthest from the saccade fi Results target was signi cantly higher (M = 0.42, SD = 0.50) than at the centre locations, p = .01, d = 0.39. Perform- We investigated the effects that crowding stimuli have ance at the location closest to the saccade target was on the linear array used in Experiment 1. To this end, significantly better than at the location furthest from we added stimuli next to the end point positions of the saccade target (Location 5), p = .03, d = 0.18. No the linear array in half of the trials and compared main effect was present for presence of the crowding these with performance on trials without the extra stimuli, z = −0.12, p = .89 and no interaction effect was stimuli. The results are shown in Figure 6. First present for location of answer probe and presence of results for the GLM show a main effect was found crowding stimuli, z = −0.20, p = .84. A repeated

Figure 6. Mean proportion of correct answers for locations in relation to saccade target between crowding stimuli and no crowding stimuli. Error bars show standard error. The dashed black line shows performance at chance level (0.08). In the figure the two conditions are horizontally offset as to not overlap visually. VISUAL COGNITION 151 measures analysis of variance confirmed these results: further away from the saccade target are not reported no significant effect of crowding stimuli, F(1,11) = 0.02, as accurately. The flanker stimuli appeared at a similar p = .89, η2 < 0.01 and no evidence for an interaction distance from the saccade target as the stimuli that effect between presence of crowding stimuli and were further from the saccade target. We suggest location of answer, F(1,11) = 0.03, p = .87, η2 < 0.01. that crowding stimuli were not selected by the presac- To provide further evidence that the added crowd- cadic shift of attention in a similar manner as the less ing stimuli did not interfere with performance on the accurately reported stimuli in the middle of the stimu- task we performed a Bayesian analysis. We choose a lus array. Thus, the crowding stimuli would not Bayesian approach since it allowed us to further compete with the stimuli closest to the saccade examine whether the null hypothesis (i.e., crowding target for storage in transsaccadic memory and does not affect accuracy on the transsaccadic thereby not elucidate effects of performance on the memory paradigm) was rightly rejected or that our transsaccadic memory task. test was inconclusive (Dienes, 2011). This distinction cannot be made by using p values. To quantify this General discussion concept: if the Bayes factor is less than 0.33 it indicates evidence for the null hypothesis and there is evidence In this study we investigated how IOR affects infor- to support the alternative hypothesis when the Bayes mation acquisition into transsaccadic memory. To factor is greater than 3. Conversely, if the Bayes factor this end we used a transsaccadic memory paradigm is between 0.33 and 3 it indicates a lack of sensitivity previously used by Irwin and Gordon (1998) with an and neither the null hypothesis nor the alternative added exogenous cue to which an additional hypothesis is supported. We conservatively modelled saccade had to be made. In both our experiment the prior distribution as a uniform distribution with a and the original paradigm it was found that infor- lower bound of 0 (no correct answers) and an upper mation that is presented near a saccade target is bound of 1 (all answers correct). This analysis indicated reported more accurately, which has been attributed fi signi cant support for the null hypothesis, BF = 0.08. to the presaccadic shift of attention. The presaccadic Therefore, the null hypothesis was rightly accepted, shift of attention has been described as an effect as our data was 12.5 times as likely to be explained that facilitates information acquisition, whereas IOR by the null hypothesis over the alternative hypothesis. is seen as a negative attentional effect, increasing We conclude that the crowding stimuli did not affect reaction times and reducing target discrimination. the performance on the transsaccadic memory task. Our main finding was that IOR near a saccade target reduces accuracy for stimuli at that same location. Discussion This implies that these elements were less accurately acquired by transsaccadic memory and thus affected We added crowding stimuli to the paradigm described the participants’ ability to report this information. We in Experiment 1 to assess the influence that endpoint did not observe a reduced performance for stimuli advantages have for correctly reporting the stimuli that were positioned further from the saccade near the saccade targets. Our conclusions in Exper- target, or when IOR was induced further away from iment 1 may be confounded by the finding that the the saccade target. These results may implicate the end point stimuli have an advantage over the centre attentional benefit provided by the presaccadic shift stimuli by lack of visual crowding. Our results for of attention being reduced by the negative attentional Experiment 2 indicate that endpoint advantages, effect of IOR. due to a lack of crowding, did not influence perform- As we can only observe a relative effect in the para- ance. To elucidate these results: in both Experiment 1 digm used, we examined the validity of our con- and Experiment 2 a main effect for location was found, clusions by investigating possible confounding where elements in nearest proximity to the saccade factors. Linear stimulus arrays, as the one used here, targets were reported more accurately than elements have been associated with endpoint advantages further from the saccade target. These results are in (Irwin & Gordon, 1998). Thus, we needed to exclude line with the findings in the original experiment by the possibility that the increased performance for Irwin and Andrews (1996). The stimuli that are stimuli at the endpoint positions in Experiment 1 152 M. J. SCHUT ET AL. was not due to a lack of visual crowding. In Experiment exclusively store this visual information between sac- 2 we investigated the effects of crowding stimuli on cades (Deubel, Schneider, & Bridgeman, 1996; Irwin the linear stimulus array used in Experiment 1, by & Andrews, 1996). Research has shown that transsac- using crowding stimuli which were similar to the cadic memory carries many similarities to VSTM, experimental stimuli. We found that the linear array such as the limited amount of objects retained and used was not affected by added visual crowding. information being selected by similar attentional pro- This indicates that the increased performance at the cesses (for a review, see Luck & Hollingworth, 2008). As endpoint positions in Experiment 1 were not due end- such, transsaccadic memory may be a subsystem of point advantages. Thus, even though the effect we VSTM (Hollingworth, Richard, & Luck, 2008). Impor- observed in Experiment 1 is relative, it seems unaf- tantly, previous studies have linked VSTM and IOR fected by the lack of crowding for the elements near closely together. Both VSTM and IOR are thought to the saccade targets. However, the effect we have be used in visual search tasks, as to respectively observed in Experiment 1 was found by inducing remember the locations of previous fixations and to IOR through instructing the participants to execute a inhibit revisiting previously attended locations saccade to a probe and refixating the centre stimulus. (Castel et al., 2003; Klein & Ivanoff, 2000; McCarley, Therefore, we cannot disentangle the oculomotor and Wang, Kramer, Irwin, & Peterson, 2003). Although our attentional component of IOR in our current series of study does not necessarily support the notion that experiments. We hypothesize this effect to be most VSTM and transsaccadic memory may consist of the strongly present in the design as described in Exper- same system, our study adds that previously fixated iment 1, as the effects of IOR may be reduced in the locations within the field of the presaccadic shift of absence of saccadic eye movements (Kingstone & attention are affected in terms of information acqui- Pratt, 1999). sition, which may implicate both VSTM and transsac- Previous research suggests that a limited amount of cadic memory. objects are stored in transsaccadic memory (Irwin & To summarize, this study provides evidence that

Gordon, 1998; Luck & Vogel, 1997; Ma, Husain, & IOR and transsaccadic memory are functionally Bays, 2014; Woodman, Vogel, & Luck, 2012). Thus, linked. Our results are in line with the notion that we analyzed whether the reduction of accuracy IOR prioritizes information presented at uninspected when a location near a saccade target is associated locations (Pratt et al., 2001), and adds that there is with IOR has an effect on the total amount of elements less benefit from the attentional processes that are stored in transsaccadic memory. To this end we closely linked to execution of saccades at locations recoded the data relative to the onset position. Our associated with IOR. This loss of benefit is not compen- results show that when the elements near a saccade sated for by increased performance at other locations target are less accurately reported, other elements in the visual field and thus results in lessened acqui- remain unaffected and are reported as accurately as sition of information. otherwise. This implicates that lessened information acquisition, in close proximity to a saccade target, is not compensated for by increased performance at Disclosure statement other locations. This suggests that IOR not only modu- No potential conflict of interest was reported by the authors. lates the accuracy of information acquisition, but also may reduce the amount of elements that are acquired between saccades. Funding The nature of transsaccadic memory has been a This research was funded by a Netherlands Organization for point of discussion since it was named. Similarities Scientific Research [VIDI grant number 452-13-008] to SVdS. between visual short-term memory (VSTM) and trans- saccadic memory have been noted since the term has arisen (Irwin, 1991). VSTM is the memory storage References where a small amount of visual information is held Alvarez, G. A., & Cavanagh, P. (2004). The capacity of visual for a limited time (Alvarez & Cavanagh, 2004; Cowan, short-term memory is set both by visual information load 2001). Transsaccadic memory is described to and by number of objects. Psychological Science: A Journal VISUAL COGNITION 153

of the American Psychological Society, 15(2), 106–111. doi:10. Hoffman, J. E., & Subramaniam, B. (1995). The role of visual 1167/2.7.273 attention in saccadic eye movements. Perception & Bays, P. M., & Husain, M. (2008). Dynamic shifts of limited Psychophysics, 57(6), 787–795. doi:10.3758/BF03206794 working memory resources in human vision. Science, 321 Hollingworth, A., & Henderson, H. J. (1998). Does consistent (5890), 851–854. doi:10.1126/science.1158023 scene context facilitate object perception? Journal of Bell, B. A., Morgan, G. B., Schoeneberger, J. A., Loudermilk, B. L., Experimental Psychology: General, 127(4), 398–415. Kromrey, J. D., & Ferron, J. M. (2010). Dancing the sample-size Hollingworth, A., Richard, A. M., & Luck, S. J. (2008). limbo with mixed models: How low can you go? SAS Global Understanding the function of visual short-term memory: Forum, 4,11–14. Transsaccadic memory, object correspondence, and gaze Briand, K. A., Larrison, A. L., & Sereno, A. B. (2000). Inhibition of correction. Journal of Experimental Psychology, 137(1), 163– return in manual and saccadic response systems. Perception 181. & Psychophysics, 62(8), 1512–1524. doi:10.3758/BF03212152 Ihaka, R., & Gentleman, R. (1996). R: A language for data analysis Castel, A. D., Pratt, J., & Craik, F. I. M. (2003). The role of spatial and graphics. Journal of Computational and Graphical working memory in inhibition of return: Evidence from Statistics, 5(3), 299–314. doi:10.2307/1390807 divided attention tasks. Perception & Psychophysics, 65(6), Irwin, D. E. (1991). Information Integration across saccadic eye 970–981. doi:10.3758/BF03194827 movements. Cognitive Psychology, 23, 420–456. Castiello, U., & Umiltà, C. (1990). Size of the attentional focus Irwin, D. E., & Andrews, R. V. (1996). Integration and accumu- and efficiency of processing. Acta Psychologica, 73(3), 195– lation of information across saccadic eye movements. 209. doi:10.1016/0001-6918(90)90022-8 Attention and performance: Information integration in percep- Cnaan, A., Laird, N. M., & Slasor, P. (2014). Using the general tion and communication, 16, 125–155. linear mixed model to analyse unbalanced repeated Irwin, D. E., & Gordon, R. D. (1998). Eye Movements, attention measures and longitudinal data. Journal of Neuroscience and trans-saccadic memory. Visual Cognition, 5, 127–155. Methods, 509(1), 805–820. doi:10.1002/(SICI)1097-0258 doi:10.1080/713756783 (19971030) Johnson, J. S., Hollingworth, A., & Luck, S. J. (2008). The role of Coëffé, C., & O’Regan, J. K. (1987). Reducing the influence of attention in the maintenance of feature bindings in visual non-target stimuli on saccade accuracy: Predictability and short-term memory. Journal of Experimental Psychology, 34, latency effects. Vision Research, 27(2), 227–240. doi:10.1016/ 41–55. 0042-6989(87)90185-4 Kenward, M. G., Roger, J. H., Process, L. D., Lodge, C. H., & Ox, O. fi

Cohen, M. E., & Ross, L. E. (1978). Latency and accuracy charac- (2013). Small sample inference for xed effects from teristics of saccades and corrective saccades in children and restricted maximum likelihood. Biometrics, 53(3), 983–997. adults. Journal of Experimental Child Psychology, 26(3), 517– Kingstone, A., & Pratt, J. (1999). Inhibition of return is composed 27. doi:10.1016/0022-0965(78)90130-3 of attentional and oculomotor processes. Perception & Cowan, N. (2001). The magical number 4 in short-term memory: Psychophysics, 61(6), 1046–54. doi:10.3758/BF03207612 A reconsideration of mental storage capacity. Behavioral and Klein, R. M., & Ivanoff, J. (2000). Inhibition of return. Trends in Brain Sciences, 24,87–114. Cognitive Sciences, 4(4), 138–147. Dalmaijer, E. S., Mathôt, S., & Van der Stigchel, S. (2013). PyGaze: Klein, R. M., & MacInnes, W. J. (1999). Inhibition of return is a An open-source, cross-platform toolbox for minimal-effort foraging facilitator in visual search. Psychological Science, 10 programming of eyetracking experiments. Behavior (4), 346–352. doi:10.1111/1467-9280.00166 Research Methods, 46(4), 1–16. doi:10.3758/s13428-013- Kooi, F. L., Toet, A., Tripathy, S. P., & Levi, D. M. (1994). The effect 0422-2 of similarity and duration on spatial interaction in peripheral Deubel, H., & Schneider, W. X. (1996). Saccade target selection vision. Spatial Vision, 8(2), 255–279. and object recognition: Evidence for a common attentional Kuznetsova, A., Brockhoff, P. B., & Christensen, R. H. B. (2013). mechanism. Vision Research, 36(12), 1827–1837. doi:10. lmerTest: Tests for random and fixed effects for linear mixed 1016/0042-6989(95)00294-4 effect models (lmer objects of lme4 package). Retrieved from Deubel, H., Schneider, W. X., & Bridgeman, B. (1996). https://cran.r-project.org/web/packages/lmerTest/lmerTest.pdf Postsaccadic target blanking prevents saccadic suppression Levi, D. M. (2008). Crowding—An essential bottleneck for object of image displacement. Vision Research, 36(7), 985–996. recognition: A mini-review. Vision Research, 48(5), 635–654. doi:10.1016/0042-6989(95)00203-0 doi:10.1016/j.visres.2007.12.009 Dienes, Z. (2011). Bayesian versus orthodox statistics: Which Lindstrom, M. J., & Bates, D. M. (1990). Nonlinear mixed effects side are you on? Perspectives on Psychological Science, 6(3), models for repeated measures data. Biometrics, 46(3), 673– 274–290. doi:10.1177/1745691611406920 687. Hilchey, M. D., Klein, R. M., & Satel, J. (2014). Returning to “inhi- Luck, S. J., & Hollingworth, A. (2008). Visual memory (S. J. Luck & bition of return” by dissociating long-term oculomotor IOR A. Hollingworth, Eds.). New York, NY: Oxford University Press. from short-term sensory adaptation and other nonoculomo- doi:10.1177/105345126600100312 tor “inhibitory” cueing effects. Journal of Experimental Luck, S. J., & Vogel, E. K. (1997). The capacity of visual working Psychology. Human Perception and Performance, 40(4), memory for features and conjunctions. Nature, 390(1996), 1603–16. doi:10.1037/a0036859 279–281. doi:10.1038/36846 154 M. J. SCHUT ET AL.

Ma, W. J., Husain, M., & Bays, P. M. (2014). Changing concepts of Sapir, A., Jackson, K., Butler, J., Paul, M. A., & Abrams, R. A. (2013). working memory. Nature Neuroscience, 17(3), 347–56. doi:10. Inhibition of return affects contrast sensitivity. The Quarterly 1038/nn.3655 Journal of Experimental Psychology, 67(7), 1305–1316. doi:10. Martín-Arévalo, E., Kingstone, A., & Lupiáñez, J. (2012). Is 1080/17470218.2013.859282 “Inhibition of Return” due to the inhibition of the return Schut, M. J. (2016). P01: Inhibition of return in transsaccadic of attention? The Quarterly Journal of Experimental memory. doi:10.17605/OSF.IO/56BJE Psychology, 66(2), 1–13. doi:10.1080/17470218.2012.711844 Van der Stigchel, S., & De Vries, J. P. (2015). There is no atten- McCarley, J. S., Wang, R. F., Kramer, A. F., Irwin, D. E., & Peterson, tional global effect: Attentional shifts are independent of M. S. (2003). How much memory does oculomotor search the saccade endpoint. Journal of Vision, 15(15), 1–12. have? Psychological Science, 14(5), 422–426. doi:10.1111/ doi:10.1167/15.15.17 1467-9280.01457 Van Selst, M., & Jolicoeur, P. (1994). A solution to the effect of Müller, M. M., & Hübner, R. (2002). Can the spotlight of attention sample size on outlier elimination. The Quarterly Journal of be shaped like a doughnut? Evidence from steady-state Experimental Psychology, 47A(3), 631–650. doi:10.1080/ visual evoked potentials. Psychological Science: A Journal of 14640749408401131 the American Psychological Society, 13(2), 119–124. doi:10. Whitney, D., & Levi, D. M. (2011). Visual crowding: A fundamen- 1111/j.0956-7976.2002.t01-1-.x tal limit on conscious perception and object recognition. Posner, M. I., & Cohen, Y. (1984). Components of visual orient- Trends in Cognitive Sciences, 15(4), 160–168. doi:10.1016/j. ing. Attention and Performance: Control of Language tics.2011.02.005.Visual Processes, 32, 531–556. doi:10.1162/jocn.1991.3.4.335 Wolfe, J. M. (1994). Guided search 2.0 a revised model of visual Pratt, J., Hillis, J., & Gold, J. M. (2001). The effect of the physical search. Psychonomic Bulletin & Review, 1(2), 202–238. characteristics of cues and targets on facilitation and inhi- Wolfe, J. M., Reinecke, A., & Brawn, P. (2006). Why don’t we see bition. Psychonomic Bulletin & Review, 8(3), 489–495. doi:10. changes? The role of attentional bottlenecks and limited 3758/BF03196183 visual memory. Visual Cognition, 14(4–8), 749–780. doi:10. Prime,S.L.,Tsotsos,L.,Keith,G.P.,&Crawford,J.D.(2007). Visual 1080/13506280500195292 memory capacity in transsaccadic integration. Experimental Woodman,G.F.,Vogel,E.K.,&Luck,S.J.(2012). Flexibility Brain Research, 180(4), 609–628. doi:10.1007/s00221-007-0885-4 in visual working memory: Accurate change detection Reuter-Lorenz, P. A., Jha, A., & Rosenquist, J. N. (1996). What is inhib- in the face of irrelevant variations in position. ited in inhibition of return? JournalofExperimentalPsychology: Visual Cognition, 20,1–28. doi:10.1080/13506285.2011. –

HumanPerceptionandPerformance, 22(2), 367 378. 630694