UBC Undergraduate Research

Where is the Inhibition in Inhibition of Return? Jarman, Jennifer 2006

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
[if-you-see-this-DO-NOT-CLICK]
Jarman Thesis.pdf [ 92.86kB ]
[if-you-see-this-DO-NOT-CLICK]
Metadata
JSON: 1.0086050.json
JSON-LD: 1.0086050+ld.json
RDF/XML (Pretty): 1.0086050.xml
RDF/JSON: 1.0086050+rdf.json
Turtle: 1.0086050+rdf-turtle.txt
N-Triples: 1.0086050+rdf-ntriples.txt
Original Record: 1.0086050 +original-record.json
Full Text
1.0086050.txt
Citation
1.0086050.ris

Full Text

Where is the Inhibition in Inhibition of Return?  By: Jennifer Jarman  A project report submitted in partial fulfillment of the requirements for the degree of Bachelor of Arts (Honours), in the department of Psychology University of British Columbia Okanagan  April 2006  2 Abstract Inhibition of return (IOR) refers to a slowing in response time for targets that appear at previously inspected locations. Using a two-location paradigm in which participants were asked to respond to one target (the “go” target) and to withhold their response to another target (the “no-go” target), Ivanoff and Klein (2001) demonstrated support for a criterion shift account of IOR. This account posits that responses are slower at previously inspected locations because participants are more reluctant to respond at these locations as compared to uninspected locations. In their study, Ivanoff and Klein found more of false alarms (responses to no-go targets) at an uncued cued vs. a cued location, supporting a criterion shift account of IOR. In the present study, Ivanoff and Klein’s paradigm was extended by increasing the number of locations from two to six and the number of cues from one to three. If a criterion shift underlies the multiple-location IOR effect, then the greatest number of false alarm responses will occur at uncued than cued locations, and successively fewer will occur from the least recently to the most recently cued location. Results failed to support a criterion shift view of multiple-location IOR.  3 Table of Contents Chapter 1: Introduction..…………………………………………………………………………..6 Chapter 2: Experiment 1…………………………………………………………………………14 Chapter 3: Experiment 2…………………………………………………………………………20 Chapter 3: Experiment 3…………………………………………………………………………24 Chapter 4: Conclusions…………………………………………………………………………..29 References………………………………………………………………………………………..31  4 List of Tables Table 1: Percentages of errors by trial type in Experiments 1-3…………………………………19 ……………………………….  5 List of Figures Figure 1: Schematic representation of Posner and Cohen’s (1984) inhibition of return study……6 Figure 2: Depiction of perceptual information based on the attentional account of IOR………..11 Figure 3: Depiction of perceptual information based on the criterion shift account of IOR….…12 Figure 4: Schematic representation of the display sequences in Experiment 1…….……………16 Figure 5: Results of Experiment 1…………………………………………………….…………17 Figure 6: Results of Experiment 2…………………………………………………….…………21 Figure 7: Results of Experiment 3…………………………………………………….…………27  6 Introduction Inhibition of return (IOR) refers to a slowing of response times for items appearing at previously inspected locations. Posner and Cohen (1984) were the first to identify IOR and suggested that it evolved to maximize the efficiency of visual search by encouraging the inspection of novel locations. In a paradigm consisting of three outlined boxes arranged horizontally (see Figure 1) one of the peripheral boxes was brightened briefly (i.e., cued) to draw attention from fixation to this location and, following a variable delay, a target appeared in one of the three boxes. To ensure that attention was removed from the peripheral location prior to target onset there was a high probability that a target would appear in the center box and a lesser, but equal, probability that it would occur in one of the two peripheral boxes (i.e., at the cued or the uncued location). When the cue-target latency (or stimulus onset asynchrony - SOA) was shorter than 150 ms, response time (RT) was faster for a target appearing in a cued than an uncued location. This “facilitation effect” has been demonstrated in numerous studies and is thought to reflect speeded perceptual processing of the target (Jonides, 1981; Müller & Rabbitt, 1989). More interestingly, Posner and Cohen (1984) reported that when the SOA was longer than 200 ms, RT was now slower for a target appearing in a cued than an uncued location. This slowing of RT was later termed “inhibition of return” to reflect the idea that attention is inhibited from returning to a previously attended location (Posner, Rafal, Choate, & Vaughn, 1985). Importantly, it has been shown that IOR is observed when target probability is not manipulated. For instance, Posner and Cohen (1984) found IOR when the target appeared equiprobably in one of the two peripheral locations if the peripheral cue was followed by a central cue to reorient attention away from the cue. Furthermore, IOR has been readily observed even if a central cue does not follow the noninformative peripheral cue (Maylor, 1985; Rafal, Calabresi, Brennan, &  7 Sciolto, 1989). This latter finding suggests that attention is voluntarily withdrawn from the cued location and is supported by the finding that children who have an incomplete understanding of probability require a reorienting event to remove attention from the peripheral location prior to target onset (MacPherson, Klein, & Moore, 2003). In addition, patients with damage to the interior frontal gyrus fail to show IOR in their contralesional visual field (Snyder & Chatterjee, in press).  Time  Figure 1. Paradigm used by Posner and Cohen (1984, Experiment 1): the center box is fixated; one of the peripheral boxes is brightened (i.e., cued); following a variable delay the target (black dot) appears. In this example, the target appears at the cued location.  Why is IOR important in visual search? Many authors agree with Posner and Cohen (1984) that IOR improves search efficiency (Klein, 1988; Posner & Cohen, 1984; Tipper, Weaver, & Watson, 1996). Importantly, Klein (1988) provided empirical support that IOR serves to maximize search efficiency by demonstrating IOR in a visual search paradigm. In a standard visual search paradigm (Treisman & Gelade, 1980), participants search for a target item that differs from distractors on the basis of one feature (i.e., feature search), such as shape, or on the basis of a conjunction of features (i.e., conjunction search), such as shape and colour. In feature search, targets are detected preattentively while in serial search detecting targets requires attention to be allocated to multiple display items to conjoin features. In his experiment, Klein (1988) presented a probe on half the trials at either a previously inspected location or a previously uninspected location. He found slowed responses for detecting the probe at previously  8 inspected locations in the conjunction task, which required attentional allocation, but not the feature task, which did not require attention, suggesting that IOR serves to prevent attention from returning to previously inspected locations (see also Müller & von Mühlenen, 2000; Takeda & Yagi, 2000). Although simple detection tasks, such as that used by Posner and Cohen (1984), provide evidence that IOR promotes search efficiency; the task lacks the complexity of real world searches. For example, searching for one’s car in a parking lot requires head and eye movements, discrimination between different cars, and searching multiple locations to find your car. In contrast to simple detection tasks that require searching for a target in one of two locations, real world searches involve movements of the eyes, head, or body. Researchers have readily observed IOR when participants respond using eye movements rather than a key press (Maylor, 1985; Posner & Cohen, 1984; Rafal, Calabresi, Brennan, & Sciolto, 1989; Reuter-Lorenz, Jha, & Rosenquist, 1996). Also, Posner et al. (1985) found that observers were biased to make saccades away from locations that had previously been fixated. When searching a complex scene, Klein and MacInnes (1999) found that saccadic RT was slowest to a probe presented at a previously searched location and that there was a decrease in saccadic RT as the distance from this location increased, which is what would be expected if IOR helps search. In addition to eye movements, real world searches often require discrimination between similar objects, rather than detection of one. In early studies, many researchers failed to find IOR in discrimination tasks, disputing IOR’s role as a facilitator of efficient search (Pontefract & Klein, 1988 as cited in Klein & Taylor, 1994; Tanaka & Shimojo, 1996; Terry, Valdes, & Neill, 1994). This issue was put to rest with the observation that IOR can be observed in discrimination tasks (Danziger, Kingstone, & Snyder, 1998; Pratt, 1995; Pratt, Kingstone, & Khoe, 1997; Pratt,  9 & Abrams, 1999). Importantly, Lupianez, Milan, Tornay, Madrid, and Tudela (1997) explained previous failures to observe IOR in discrimination tasks by demonstrating that IOR occurs at longer SOAs in discrimination than in detection tasks. Although IOR can be found in discrimination tasks, the simple IOR paradigm does not approximate the complexity of real world searches. Specifically, search often involves more than one location and, therefore, IOR should be found when more than one location is cued. Accordingly, several studies have reported robust IOR effects at up to three sequentially cued locations and that the magnitude of this IOR effect is greatest at the most recently cued location and smaller at less recently cued locations (Danziger et al., 1998; Snyder & Kingstone, 2001; Tipper, Weaver, & Watson, 1996). In addition, Snyder and Kingstone (2000) demonstrated that IOR is observed at five and possibly six cued locations. Thus, the observation of IOR in detection and discrimination tasks and at multiple locations suggests that IOR serves to increase search efficiency. However, the question remains: where is the inhibition in inhibition of return? As previously stated, the first explanation of IOR was that it is the inhibition of attention from returning to a previously searched location (Posner et al., 1985). Support for this account first came from Klein (1988) who demonstrated that IOR is present only in search tasks that require attention. Similarly, Snyder and Kingstone (2001) demonstrated that multiple-location IOR is only observed if participants attend all peripheral events, regardless of why attention is committed. When a manipulation was introduced that eliminated the need for participants to attend peripheral cues, IOR was eliminated at all but the most recently cued location, indicating that attention accounts for at least some of the multiplelocation IOR effect. Moreover, Reuter-Lorenz, Jha, and Rosenquist (1996) suggested that if IOR results from an inhibition of attention then it should share similar characteristics with attention.  10 They hypothesized that, if IOR is an attentional effect then factors known to influence the magnitude of attention, such as target modality and target intensity, should similarly influence the magnitude of IOR. As hypothesized, they found that the magnitude of IOR, like the magnitude of attention, is greater for visual than for auditory stimuli and is greater for low intensity targets than for high intensity targets. Further support for the attentional account comes from Handy, Jha, and Mangun (1999) who suggested that IOR characterized as an inhibition in processing at that location. Handy et al. (1999) suggested two results of this inhibition: that IOR slows the rate at which perceptual information accumulates or that IOR decreases the total amount or quality of the perceptual information that accumulates. They hypothesized that if IOR decreases perceptual processing, sensitivity for detecting targets should be lower at an uncued location. To test this hypothesis, Handy et al. used a go/no-go task in which participants respond to one target (the “go” target) and withhold their response to a second target (the “no-go” target) to assess differences in accuracy at cued and uncued locations on trials in which false alarms (i.e., responses to no-go targets) occurred. They found that a slower RT at a cued location was associated with lower sensitivity, supporting the view that IOR results from an inhibition of attention at a cued location. Handy et al. (1999) offered two explanations of this inhibition (see Figure 2). One explanation was that IOR slows the rate at which perceptual information accumulates. The second explanation, which was the one they favoured, was that IOR decreases the total amount or quality of the perceptual information that accumulates.  11  uncued  Time (ms)  cued  uncued  cued  Time (ms)  Figure 2. Depiction of Handy et al.’s (1999) possible explanations of the source of IOR: (left panel) that perceptual information accrues more slowly at a cued than at an uncued location or (right panel) that less perceptual information accrues or the information is of poorer quality.  Although there is considerable evidence to suggest that IOR reflects the inhibition of attention, this is not the only possible explanation. Evidence has also been found to support a response bias account of IOR. That is, IOR may reflect an inhibition of responding, rather than an inhibition of attention. One such account comes from Klein and Taylor (1994) and was based on initial failures to observe IOR in discrimination tasks. They suggested that, because IOR was present for localization tasks but not for discrimination tasks, that IOR resulted from a motor bias. Specifically, they suggested that when an abrupt onset occurs, an oculomotor response is generated for that location. If the onset is a cue, then a response must be inhibited and when a target occurs in that same location, this inhibition must be overcome, resulting in longer RT at this location. In support of this idea, Rafal et al. (1989) found that simply preparing, but not executing, an oculomotor response was as efficient at producing IOR as actually making an eye movement to the cued location. Presently, most authors agree that both attention and a response bias underlie the IOR effect (Hunt & Kingstone, 2003; Kingstone & Pratt, 1999; Taylor & Klein, 2000). A more recent version of the response bias account comes from Ivanoff and Klein (2001) who proposed that a shift in response criterion accounts for at least some, if not all, of the IOR effect. Although they were primarily interested in the effect of a nonresponding effector on IOR,  12 their go/no-go paradigm allowed them to assess differences in false alarms for a cued and an uncued location. They found that the slower RT associated with IOR was accompanied by a lower rate of false alarms. They suggested that participants are more reluctant to respond to a target appearing at a cued than at an uncued location. According to Ivanoff and Klein’s results, perceptual information accrues at the same rate but the slower RT at a cued location reflects the need for more information to establish target identity at a cued than at an uncued location (see Figure 3). This finding is in contrast to that of Handy et al. (1999) who did not find such differences in false alarms at a cued vs. an uncued location. Recall that Handy et al. (1999) attributed the IOR effect to lowered sensitivity to detect a target at a cued location. Ivanoff and Klein (2001) suggested that task differences could account for the equal rate of false alarms found by Handy et al. (1999). Specifically, Handy et al. presented a mask after target onset so that any information that had been gathered regarding the target would have begun to decay so faster responses would have been more accurate than slower ones. Therefore, a difference in sensitivity may have been found because responses are slower at the cued location so more target information would have decayed, not because perceptual processing for targets appearing at the cued location was degraded.  uncued  cued  Time (ms)  Figure 3. Depiction of Ivanoff and Klein’s (2001) explanation that longer RT at cued locations reflects a shift in response criterion, rather than a difference in the accrual of perceptual information.  13 In sum, IOR is found in simple detection and discrimination tasks using eye movements and manual responses, suggesting that IOR facilitates search efficiency. The finding of IOR at multiple locations also supports this contention. However, there is still debate concerning the source of this inhibition. Considerable evidence suggests that IOR results from an inhibition of attention at the cued location (Handy et al., 1999; Klein, 1988; Posner et al., 1985, ReuterLorenz et al., 1996; Snyder & Kingstone, 2001). However, there is also ample evidence to suggest that IOR is due to a response bias, either a motor bias (Klein & Taylor, 1994) or a criterion shift (Ivanoff & Klein, 2001) With regard to the criterion shift account, if Ivanoff and Klein’s contention that a criterion shift underlies the IOR effect, then such a shift should occur in more complex search situations. Thus, the goal of the present study was to determine whether the criterion shift account of IOR would occur in multiple-location IOR. Using a sequential cueing paradigm that was paired with a go/no-go task, it was expected that IOR would be observed at all cued locations with slower RT at a cued location than at an uncued location and slower RT would be accompanied by fewer false alarms.  14 Experiment 1 The go/no-go procedure employed by Ivanoff and Klein (2001) was modified to test whether a criterion shift underlies multiple-location IOR. This modification included increasing the number of locations from two to six and the number of cues from one to three. If a criterion shift is responsible for any of the multiple-location IOR effects, then a faster RT at an uncued location should be accompanied by more false alarms than a slower RT at a cued location. A strict interpretation of this account predicts that a greater number of false alarms will occur at the least recently cued location than at the most recently cued location, given the typical decrease in RT, and hence IOR, across less recently cued locations. Method Participants. Twenty-nine1 undergraduate students from the University of British Columbia Okanagan participated in this experiment for course credit or a payment of $10/hour. All participants were naïve to the purpose of the experiment and had normal or corrected-tonormal vision. Apparatus and Stimuli. Stimuli were displayed on a 44 cm computer monitor situated 50 cm from participants. The stimulus display consisted of a black background with six light grey outline boxes measuring 1.1° of visual angle (va) and placed equidistant around an imaginary circle, with the first box located at 0° va (see Figure 4). The circle’s radius, from its center to the center of each box, measured 6.3° va. A light grey fixation dot, measuring 0.3° va, marked the center of the imaginary circle. Cueing was accomplished by changing a light grey box to white. The two target types (i.e., go and no-go) were a white “x” and a white “+,” each measuring 0.6° va x 0.6° va. RT was recorded by the computer and measured in ms. 1  Forty-one subjects were tested, 12 were excluded due to go errors > 10% or no-go errors > 25%.  15 Design. At the start of the session, participants received 24 practice trials with equal numbers of go targets that required a response and no-go targets that required withholding a response. Following the practice trials, participants received 486 experimental trials in nine blocks of 54 trials. Following the procedure of Ivanoff and Klein (2001), a ratio of 2:1 go to nogo trials was used, with each block containing 36 go (6 cued one-back, 6 cued two-back, 6 cued three-back, 18 uncued) and 18 no-go (3 cued one-back, 3 cued two-back, 3 cued three-back, 9 uncued) trials. For half the participants, the go target was an “x” and for the other half it was a “+.” The cues did not predict target location, and the target could appear equiprobably at any of the six possible locations. Procedure. Participants were seated in a dimly lit room, in front of the computer monitor. They were instructed: (1) that three sequential cues would precede target onset; (2) that the cues did not predict where the target would occur; (3) to maintain fixation on the central dot; (4) to respond quickly and accurately to the go target and not to respond to the no-go target; and (5) that the ratio of go to no-go trials was 2:1. Each trial began with a 100 ms warning tone. Following a 400 ms delay, three sequential cues preceded target onset, with each event separated by an SOA of 1,000 ms. Cue duration was 350 ms and target duration was 1,000 ms. Importantly, the location of each cue was randomly selected with the constraint that the same location could never be cued more than once on a trial. Trials were terminated when a key press response was executed or 1,000 ms after the target appeared, whichever came first. An error tone that was easily distinguishable from the warning tone provided response feedback on incorrect responses. Errors on go trials could be as follows: (1) anticipations (i.e., responses to a target in < 150 ms); (2) false alarms (i.e., responses prior to target onset); (3) misses (i.e., failure to respond  16 to a target in < 1000 ms). Errors on no-go trials could only be false alarms- either to the cue or to the target. Each participant completed one 60 min session.  x  Figure 4. An example of a trial in Experiment 1, with three peripheral cues preceding target onset (an “x”). Cue durations were 350 ms, all stimulus onset asynchronies (SOAs) were 1000 ms. In this example, the target appears at the location cued second to last.  Results Before presenting the results, it is important to clarify terminology. When a target occurs in the same location as the last cue, it is referred to as being at a cued one-back location as it is one-back from target onset. When a target occurs in the location of the second to last cue, it is at a cued two-back location. And when a target occurs at the location of the third to last cue, it is at a cued three-back location. Mean correct RT for target detection as a function of target location is presented in Figure 5. Response time. A repeated measures analysis of variance (ANOVA) was conducted on mean correct RT with target location (cued one-back, cued two-back, cued threeback, uncued locations) as a factor. The results demonstrated a significant main effect of target location, F(3, 28) = 2.87, MSe = 156.32, p < .05. A contrast of cued vs. uncued locations revealed the presence of IOR, F(1, 28) = 7.972, MSe = 1246.14, p < .01, with RT longer at a cued than an uncued location (565 ms vs. 557 ms). Additional contrasts revealed no RT differences between the cued-back locations (all Fs < 1), suggesting that IOR did not vary as a function of cued-back location.  17  600  *  *  *  cued one-back  cued two-back  cued three-back  575  550  525  500 0 uncued  Target Location  Figure 5. Mean correct RT as a function of target location. The asterisk indicates a significant difference between a cued-back location and an uncued location.  Response accuracy. Error data for all three experiments in this study are presented in Table 1. A repeated measures ANOVA was conducted on mean accuracy with trial type (go, nogo) and target location as factors. The results demonstrated a significant main effect for trial type, F(1, 28) = 41.72, MSe = .06, p < .001, with higher accuracy on go trials than no-go trials (96% vs. 75%), but no other significant main effects or interactions (both Fs <1). A second repeated measures ANOVA was conducted on mean accuracy for the no-go trials with target location as a factor to assess possible differences in false alarms. This analysis failed to reach significance, F(3, 28) = 1.15, MSe = 32.44, p > .05, suggesting that there were no differences in false alarms at a cued vs. an uncued location.  18 Table 1 Percentages of Errors by Trial Type in Experiments 1-3 Trial type (% Errors) Experiment  Error Type  Go  No-go  1  False Alarm  0.3  10.0  Anticipatory Response  0.0  Miss  1.0  False Alarm  0.2  3.9  Anticipatory Response  0.0  0.0  Miss  1.1  0.0  False Alarms  0.4  11.2  Anticipatory Response  0.0  0.0  Miss  1.6  0.0  2  3  Discussion This experiment tested the criterion shift account of IOR. It was hypothesized that if a criterion shift underlies multiple-IOR effects, then more false alarms should occur at an uncued vs. a cued location. In addition, false alarms should decrease from the cued three-back to the cued one-back location, as RT increased. The results demonstrated that a small (8 ms) IOR effect was present with no differences in the magnitude of the effect at the three cued locations. This finding is in contrast to previous observations (Danziger et al, 1998; Snyder & Kingstone, 2000; Snyder & Kingstone, 2001; Tipper et al., 1996) that the magnitude of IOR is greater for the more recently cued locations. In contrast to the findings of Ivanoff and Klein (2001), false alarms did not differ for cued and uncued locations. This finding does not appear to support a criterion shift account of IOR. However, there are major differences between this experiment and the study conducted by Ivanoff and Klein. As noted, the paradigm used in the present study presented  19 three sequential cues at three of six possible locations prior to target onset, whereas their paradigm consisted of a single cue at one of two locations. It is possible that a paradigm with increased number of locations and multiple cues increases task demands. This contention is supported in that nearly twice as many false alarms occurred in the present experiment compared with Ivanoff and Klein, suggesting that a more liberal criterion for responding was adopted by participants. This liberal criterion may have obscured differences in false alarms between the cued and uncued locations. It is conceivable that reducing the overall false alarms may allow such differences to emerge.  20 Experiment 2 As in Experiment 1, the goal of Experiment 2 was to test the criterion shift account for multiple-location IOR. To decrease the overall false alarm rate, the ratio of go to no-go trials was reduced to 1:1. If a criterion shift underlies multiple-IOR effects, then more false alarms should occur at an uncued location, with the number of false alarms decreasing from a cued three-back to a cued one-back location. Method Participants. Seventeen undergraduate students from the University of British Columbia Okanagan participated in this experiment for course credit or a payment of $10/hour. All participants were naïve to the purpose of the experiment and had normal or corrected-to-normal vision. Apparatus and stimuli. As in Experiment 1. Design. As in Experiment 1, with the following exceptions. Following the practice trials, participants received 648 experimental trials in eight blocks of 36 trials. Each block of trials contained 36 go (6 cued one-back, 6 cued two-back, 6 cued three-back, 18 uncued) and 36 no-go (6 cued one-back, 6 cued two-back, 6 cued three-back, 18 uncued) trials. Procedure. As in Experiment 1, with the exception that participants were instructed that the ratio of go to no-go trials was 1:1. Results Data analyses were conducted as in Experiment 1. Mean correct RT for target detection as a function of target location is presented in Figure 4. Response time. A repeated measures ANOVA was conducted on mean correct RT with target location as a factor. As in Experiment 1, the results demonstrated a significant main effect  21 of target location, F(3, 16) = 3.67, MSe = 147.91, p < .05. A contrast of cued vs. uncued locations revealed the presence of IOR, F(1, 16) = 10.23, MSe = 1513.79, p < .01, with RT longer at a cued than an uncued location (581 ms vs. 569 ms). Again, additional contrasts did not reveal RT differences between the cued-back locations (all Fs < 1). 600  *  *  *  cued one-back  cued two-back  cued three-back  575  550  525  500 0 uncued  Target Location  Figure 6. Mean correct RT as a function of target location. The asterisk indicates a significant difference between a cued-back location and an uncued location.  Response accuracy. A repeated measures ANOVA was conducted on mean accuracy with trial type and target location as factors. As in Experiment 1, results demonstrated a significant main effect for trial type, F(1, 16) = 15.05, MSe = .003, p < .01, with higher accuracy on go trials than no-go trials (98% vs. 94%). No other main effect or interaction was significant (both Fs < 1.60, both ps > .05). A second repeated measures ANOVA conducted on accuracy for the no-go trials with target location as a factor failed to reach significance, F(3, 16) = 1.43, MSe = 3.98, p > .05, indicating that there were no differences in false alarms at a cued vs. an uncued location. Discussion This experiment represents a second attempt at replicating Ivanoff and Klein’s (2001) finding of increased false alarms accompanying faster RT at an uncued location in support of a criterion shift account of multiple-location IOR. The rationale was to reduce the overall number of false alarms found in Experiment 1 by inducing a more conservative criterion for responding.  22 It was possible that this manipulation might have allowed a difference in false alarms between a cued and an uncued location to emerge. As in Experiment 1, IOR was present with no observable differences in the overall magnitude of IOR, F(2, 112) = .009, MSe = 161.56, p > .05, across the experiments (8 ms vs. 11 ms, for Experiments 1 and 2, respectively). As was the case in Experiment 1, the magnitude of IOR was consistent across cued-back locations. And, importantly, no differences in false alarms were found between cued and the uncued locations despite a significant improvement in accuracy (90% vs. 94%, for Experiments 1 and 2, respectively) on no-go trials, F(1, 56) = 10.63, MSe = .102, p < .01. Taken together, the results of Experiments 1 and 2 fail to support the criterion shift account of multiple-location IOR. However, before it can be concluded that a criterion shift does not influence multiple-location IOR, two points must be considered. First, it must be noted that multiple-location IOR paradigms requiring detection or discrimination rather than go/no-go responses yield a greater magnitude IOR effect than was observed in the present study. Second, multiple-location IOR paradigms requiring detection or discrimination rather than a go/no-go response yield greater IOR at more recently than less recently cued locations. Thus, it is possible that differences in false alarms might be observed with more robust IOR effects. Snyder and Kingstone (2001) provided evidence that multiple-location IOR is contingent on attention being committed to each cued location, regardless of why attention is committed there. In their experiment, participants were presented with a multiple-location IOR paradigm in which one, two, or three cues could precede target onset with an incrementing numeral at fixation coinciding with each cue and the target onset. In one condition, both groups did not know how many cues would precede target onset and would have to attend each onset event to determine whether it was a cue or a target. In a second condition, participants were informed regarding the number of  23 cues that would precede target onset and could use the numeral to prepare for target onset. Results revealed that, in the first condition when the numeral was uninformative, attention was committed to each onset event and a robust IOR effect was found. However, in the second condition when the numeral was informative, attention was not committed to each onset event and the IOR effect was abolished at all but the most recently cued location. In the present experiments, three cues always preceded target onset and it is possible that the weak IOR effects reflect limited attention being committed to the cued locations. By creating uncertainty as to whether an onset event is a cue or a target, it is hoped that a more robust IOR effect will emerge, and, with it, the characteristic decline in RT from the most to the least recently cued locations, allowing possible differences in false alarms to emerge.  24 Experiment 3 Experiments 1 and 2 failed to replicate Ivanoff and Klein’s (2001) finding of increased false alarms at uncued locations, supporting a criterion shift account of multiple-location IOR. The goal of the final experiment was to increase attentional allocation at cued locations by creating uncertainty (Danziger, Kingstone, & Snyder, 1998; Snyder & Kingstone, 2001), yielding a more typical pattern of multiple-location IOR. It was hypothesized that the increase in attention would produce greater IOR effects that declined linearly from the most recently cued location to the less recently cued locations. As in the previous experiments, it was expected that the greater the IOR effect (i.e., the slower the RT), the fewer the false alarms. Given that neither the magnitude of IOR nor a differential rate of false alarms were observed with a 1:1 ratio of go to no-go trials, a 2:1 ratio of go to no-go trials was used. Method Participants. Sixteen2 undergraduate students from the University of British Columbia Okanagan participated in this experiment for course credit or a payment of $10/hour. All participants were naïve to the purpose of the experiment and had normal or corrected-to-normal vision. Apparatus and stimuli. As in Experiment 1. Design. As in Experiment 1, with the following exceptions. At the start of the session, participants received 40 practice trials, with equal number of go and no-go trials. Following the practice trials, participants received 504 experimental trials in eight blocks of 63 trials each. Each block contained 42 go trials and 21 no-go trials. Each block of go trials consisted of six zero-cue, 12 one-cue (2 cued one-back, 10 uncued), 12 two-cue (2 cued one-back, 2 cued two-back, 8 2  Thirty-three subjects were tested, 13 were excluded due to an error in programming, 3 were excluded due to go errors > 10% or no-go errors > 25%.  25 uncued) and 12 three-cue (2 cued one-back, 2 cued two-back, 2 cued three-back, 6 uncued) trials. Each block of no-go trials consisted of three zero-cue, six one-cue (1 cued one-back, 5 uncued), six two-cue (1 cued one-back, 1 cued two-back, 4 uncued), and six three-cue (1 cued one-back, 1 cued two-back, 1 cued three-back, 1 uncued) trials. Procedure. As in Experiment 1, with the exception that participants were instructed that zero, one, two, or three cues could precede target onset. Results Separate repeated measures ANOVAs were conducted for each cued location (cued oneback, cued two-back, cued three-back). Mean correct RT for target detection as a function of the number of cues and target location is presented in Figure 5. Target-only trials. The mean correct RT for a zero-cue trial was 614.35 ms. Cued one-back trials. A repeated measures ANOVA was conducted on mean correct RT with number of cues (1, 2, 3) and target location (cued one-back, uncued) as factors. Results revealed a significant main effect of number of cues, F(2, 15) = 29.07, MSe = 685.88, p < .001 and contrasts revealed a decrease in RT as the number of cues increased (all Fs > 8.56, all ps < .01). This result is consistent with previous research suggesting a preparation effect that decreases as number of cues increases (Danziger et al., 1998; Snyder and Kingstone, 2000; Snyder and Kingstone, 2001). A significant main effect of target location, F(1, 15) = .00, MSe = 367.53, p > .05, was not observed. However, a significant number of cues x target location interaction, F(6, 15) = 4.13, MSe = 273.60, p < .05, was found. Contrasts revealed the presence of IOR for a one-cue trial (578 ms vs. 564 ms), F(1, 15) = 5.47, MSe = 1497.77, p < .05, but not for a two-cue or three-cue trial (both Fs < 4.13, both ps > .05).  26 A repeated measures ANOVA was conducted on mean accuracy with trial type, number of cues, and target location as factors. As in previous experiments, a significant main effect of trial type, F(1, 15) = 32.54, MSe = .02, p < .001,was observed, with higher accuracy on go trials than no-go trials (98% vs. 88%). Importantly, no other main effects or interactions were significant (all Fs < 1). A second repeated measures ANOVA conducted on mean accuracy for no-go trials with number of cues and target location as factors revealed no significant main effects or interactions (all Fs < 1), indicating that there were no differences in false alarms as a function of target location. Cued two-back trials. A repeated measures ANOVA was conducted on mean correct RT with number of cues (2, 3) and target location (cued two-back, uncued) as factors. Results demonstrated a significant main effect for number of cues, reflecting a preparation effect similar to that observed at the cued one-back location, F(1, 15) = 8.90, MSe = 519.22, p < .01. However, neither a significant main effect of target location nor a significant number of cues x target location interaction were observed (both Fs < 2.80, both ps > .05), indicating that IOR was not present at the cued two-back location. A repeated measures ANOVA was conducted on mean accuracy with trial type, number of cues, and target location as factors. Again, results revealed a significant main effect of trial type, F(1, 15) = 30.94, MSe = .01, p < .001, with higher accuracy on go trials than no-go trials (98% vs. 87%). And, once again, no other main effects or interactions were significant (all Fs < 4.87, all ps > .05). The second repeated measures ANOVA conducted on mean accuracy for nogo trials with number of cues and target location as factors failed to reveal any significant main effects or interactions (all Fs > 1), again indicating that there were no differences in false alarms as a function of target location.  27 Cued three-back trials. A repeated measures ANOVA was conducted on mean correct RT with target location (cued three-back, uncued) as a factor. Results revealed a significant main effect of target location, F(1, 15) = 4.98, MSe = 192.23, p < .05, indicating the presence of IOR, with longer RT at a cued three-back than at an uncued location (535 ms vs. 524 ms). An ANOVA was conducted on mean accuracy with trial type and target location as factors. The results revealed a significant main effect of trial type, F(1, 15) = 21.74, MSe = .01, p < .01, with higher accuracy on go than no-go trials (98% vs. 89%), but again failed to reveal a significant main effect of target location or an interaction involving trial type and target location (all Fs < 1). The ANOVA conducted on accuracy for no-go trials with target location as a factor failed to reveal a significant main effect (F < 1) again indicating that there were no differences in false alarms as a function of target location. no cue  625  cued  600  uncued  * 575  Cued One-back  Cued Two-back  Cued Three-back  550  * 525  500 0 0 cue  1 cue  2 cues  3 cues  2 cues  3 cues  3 cues  Number of Cues  Figure 7. Mean correct RT as a function of number of cues and target location. The asterisk indicates a significant difference in RT between that cued location and uncued location.  Discussion This experiment was aimed at providing a final test of the criterion shift account of multiple-location IOR by attempting to generate the magnitude of IOR effect typically observed in a multiple-location paradigm with the accompanying decline in IOR across cued-back locations. However, results revealed that the attentional manipulation intended to produce more robust IOR effects failed. Specifically, IOR was present only in the initially cued location for  28 both one- and three-cue trials and absent in the two-cue trials. In addition, when IOR was observed, the magnitude remained small (14 and 9 ms for the cued one- and cued three-back locations, respectively). As in Experiments 1 and 2, no differences in false alarms were observed between the cued and uncued locations. Rather than finding support for the criterion shift account of multiple-location IOR, Experiment 3 provides some support for an attentional account of multiple-location IOR. As indicated, the manipulation used in this experiment has reliably produced robust IOR effects in the past. Thus, the finding that this manipulation failed to produce such effects suggests that the high attentional demands of the go/no-go task left little attention to be committed to cued locations. This finding supports the conclusion of Snyder and Kingstone (2001) that multiple-location IOR is, at least in part, attention-based.  29 Conclusions Two views have been proposed to explain where the inhibition comes from in the IOR phenomenon. The attentional account proposes that IOR is the result of an inhibition of attention, whereas the response bias account proposes that IOR is the result of an inhibition of a response. Specifically, one response bias account proposes that IOR is the result of inhibiting the execution of a motor response to a cued location (Klein & Taylor, 1994). A more recent account posits that IOR is the result of a criterion shift, such that participants require more information to respond to a target at a cued than at an uncued location (Ivanoff & Klein, 2001). The present study was aimed at extending Ivanoff and Klein’s supporting evidence of increased false alarms at an uncued location to multiple-location IOR. Thus, the number of cues was increased from one to three and the number of locations was increased from two to six. It was hypothesized that if a criterion shift underlies at least part of the inhibition in multiple-location IOR, then a larger number of false alarm errors would occur at an uncued than at a cued location, with the number of false alarms decreasing from a cued three-back to a cued one-back location, reflecting a more conservative response criterion. The results of Experiments 1 did not support this hypothesis. Although IOR was observed, no differences in false alarms between cued and uncued locations were observed. It was suggested that the liberal response criterion adopted may have obscured any differences in false alarms. Therefore, Experiment 2 was aimed at producing a more conservative response criterion by equalizing the ratio of go to no-go trials. Although overall accuracy on no-go trials increased from Experiment 1 to 2, and IOR was observed, again no differences in false alarms were revealed. Taken together, the results of Experiments 1 and 2 suggest that the criterion shift account does not generalize to multiple-location IOR. However, there were two problems with Experiments 1 and 2. First, only a small IOR effect was observed  30 and, second, this effect was not larger for more recently cued locations. To address this concern, a manipulation shown by Snyder and Kingstone (2001) to produce robust multiple-location IOR effects was introduced. Snyder and Kingstone (2001) demonstrated that attention must be committed to each location in order to observe robust IOR effects, regardless of why it is committed. The aim of Experiment 3 was to generate robust IOR effects by increasing attentional allocation at cued locations to reveal any potential differences in false alarm errors at cued and uncued locations. However, this manipulation, which has not failed in the past (Danziger et al., 1998; Snyder and Kingstone, 2000; Snyder and Kingstone, 2001; Tipper et al., 1996), did not produce robust IOR effects in the present study. It appears that the attentional demands of the go/no-go task left no attention to be committed to cue onsets and generate robust IOR. Rather than supporting a criterion shift account of IOR, this finding supported the conclusion of Snyder and Kingstone’s (2001) conclusion that multiple-location IOR is attentionbased. However, the debate regarding the source of the inhibition in IOR will continue.  31 References Danziger, S., Kingstone, A., & Snyder, J.J. (1998). Inhibition of return to successively stimulated locations in a sequential visual search paradigm. Journal of Experimental Psychology: Human Perception and Performance, 24, 1467-1475. 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. 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. Long & A. Baddeley (Eds.), Attention and performance IX (pp. 187-203). Hillsdale, NJ: LEA. Kingstone, A, & Pratt, J. (1999). Inhibition of return is composed of attentional and oculomotor processes. Perception & Psychophysics, 61, 1046-1061. Klein, R.M., Kingstone, A., & Pontefract, A. (1992). Orienting of visual attention. In K. Rayner (Ed.), Eye movements and visual cognition: Scene perception and reading (pp. 46-63). North-Holland: Elsevier Science Publishers. Klein, R.M., & Taylor, T.L. (1994). Categories of cognitive inhibition with reference to attention. In D. Dagenbach & Carr, T.H. (Eds.), Inhibitory processes in attention, memory, and language (pp. 113-150). San Diego, CA: Academic Press. Klein, R. (1988). Inhibitory tagging system facilitates visual search. Nature, 334, 430-431.  32 Klein, R. M., & MacInnes, W.J. (1999). Inhibition of return is a foraging facilitator in visual search. Psychological Science, 10, 346-352. Lupianez, J., Milan, E.G., Tornay, F.J., Madrid, E., & Tudela, P. (1997). Does IOR occur in discrimination tasks? Yes, it does, but later. Perception & Psychophysics, 59, (12411254). MacPherson, A. C., Klein, R. M., & Moore, C. (2003). Inhibition of return in children and adolescents. Journal of Experimental Child Psychology, 28, 337-351. Maylor, E. A. (1985). Facilitatory and inhibitory components of orienting in visual space. In M. I. Posner & O. S. M. Marin (Eds.), Attention and performance XI (pp. 189-207). Hillsdale, NJ: LEA. Muller, 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, 315-330. Muller, H.J., von Muhlenen, A. (2000). Probing distractor inhibition in visual search: Inhibition of return. Journal of 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: Control of language processes (pp. 531556). 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 & Review, 2, 117-120.  33 Pratt, J., Kingstone, A., & Khoe, W. (1997). Inhibition of return in location- and identity-based choice decision tasks. Perception & Psychophysics, 59, 964-971. Pratt, J., & Abrams, R. A. (1999). Inhibition of return in discrimination tasks. Journal of Experimental Psychology: Human Perception and Performance, 25(1), 229-242. Rafal, R.D., Calabresi, P.A., Brennan, 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., 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. Snyder, J.J., & Kingstone, A. (2000). Inhibition of return and visual search: How many separate loci are inhibited? Perception & Pyschophysics, 62, 452-458. Snyder, J.J., & Kingstone, A. (2001). Inhibition of return at multiple locations in visual search: When you see it and when you don’t. Quarterly Journal of Experimental Psychology, 54A, 1221-1237. Snyder, J.J., & Chatterjee, A. (in press). The frontal cortex and exogenous attentional orienting. Journal of Cognitive Neuroscience. Takeda, Y., & Yagi, A. (2000). Inhibitory tagging in visual search can be found if search stimuli remain visible. Perception & Psychophysics, 62, 927-934. Tanaka, Y., & Shimojo, S. (1996). Location vs. feature: Reaction time reveals dissociation between two visual functions. Vision Research, 36, 2125-2140. Terry, K.M., Valdes, L.A., & Neill, W.T. (1994). Does “inhibition of return” occur in discrimination tasks? Perception & Psychophysics, 55, 279-286.  34 Tipper, S.P., Driver, J., & Weaver, B. (1991). Object-centered inhibition of return of visual attention. Quarterly Journal of Experimental Psychology, 43A, 289-298. Tipper, S.P., Weaver, B., & Watson, F.L. (1996). Inhibition of return to successively cued spatial locations: Commentary on Pratt and Abrams (1995). Journal of Experimental Psychology: Human Perception and Performance, 22, 12998-1293. Treisman, G.A., & Gelade, G. (1980). A feature-integration theory of attention. Cognitive Psychology, 12, 97-136.  

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
China 23 0
United States 16 1
Canada 7 0
Japan 4 0
France 3 0
Germany 2 0
Russia 1 0
Malaysia 1 0
City Views Downloads
Beijing 19 0
Ashburn 12 0
Vancouver 7 0
Unknown 5 0
Shenzhen 4 0
Tokyo 4 0
Wilmington 1 0
Mountain View 1 1
Sunnyvale 1 0
Saint Petersburg 1 0
University Park 1 0
Petaling Jaya 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats

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.42496.1-0086050/manifest

Comment

Related Items