Target-present guessing as a function of target prevalence and accumulated information in visual search

Target prevalence influences visual search behavior. At low target prevalence, miss rates are high and false alarms are low, while the opposite is true at high prevalence. Several models of search aim to describe search behavior, one of which has been specifically intended to model search at varying prevalence levels. The multiple decision model (Wolfe & Van Wert, Current Biology, 20(2), 121-–124, 2010) posits that all searches that end before the observer detects a target result in a target-absent response. However, researchers have found very high false alarms in high-prevalence searches, suggesting that prevalence rates may be used as a source of information to make “educated guesses” after search termination. Here, we further examine the ability for prevalence level and knowledge gained during visual search to influence guessing rates. We manipulate target prevalence and the amount of information that an observer accumulates about a search display prior to making a response to test if these sources of evidence are used to inform target present guess rates. We find that observers use both information about target prevalence rates and information about the proportion of the array inspected prior to making a response allowing them to make an informed and statistically driven guess about the target’s presence.     

Low target prevalence is a stubborn source of errors in visual search tasks

In visual search tasks, observers look for targets in displays containing distractors. Likelihood that targets will be missed varies with target prevalence, the frequency with which targets are presented across trials. Miss error rates are much higher at low target prevalence (1%-2%) than at high prevalence (50%). Unfortunately, low prevalence is characteristic of important search tasks such as airport security and medical screening where miss errors are dangerous. A series of experiments show this prevalence effect is very robust. In signal detection terms, the prevalence effect can be explained as a criterion shift and not a change in sensitivity. Several efforts to induce observers to adopt a better criterion fail. However, a regime of brief retraining periods with high prevalence and full feedback allows observers to hold a good criterion during periods of low prevalence with no feedback.     

Even in correctable search, some types of rare targets are frequently missed

Socially important visual search tasks, such as airport baggage screening and tumor detection, place observers in situations where the targets are rare and the consequences of failed detection are substantial. Recent laboratory studies have demonstrated that low target prevalence yields substantially higher miss errors than do high-prevalence conditions, in which the same targets appear frequently (Wolfe, Horowitz, & Kenner, 2005; Wolfe et al., 2007). Under some circumstances, this \ldprevalence effect\rd can be eliminated simply by allowing observers to correct their last response (Fleck & Mitroff, 2007). However, in three experiments involving search of realistic X-ray luggage images, we found that the prevalence effect is eliminated neither by giving observers the choice to correct a previous response nor by requiring observers to confirm their responses. This prevalence effect, obtained when no trial-by-trial feedback was given, was smaller than the effect obtained when observers searched through the same stimuli but were given trial-by-trial feedback about accuracy. We suggest that low prevalence puts pressure on observers in any search task, and that the diverse symptoms of that pressure manifest themselves differently in different situations. In some relatively simple search tasks, misses may result from motor or response errors. In other, more complex tasks, shifts in decision criteria appear to be an important contributor.     

A bayesian optimal foraging model of human visual search

Real-world visual searches often contain a variable and unknown number of targets. Such searches present difficult metacognitive challenges, as searchers must decide when to stop looking for additional targets, which results in high miss rates in multiple-target searches. In the study reported here, we quantified human strategies in multiple-target search via an ecological optimal foraging model and investigated whether searchers adapt their strategies to complex target-distribution statistics. Separate groups of individuals searched displays with the number of targets per trial sampled from different geometric distributions but with the same overall target prevalence. As predicted by optimal foraging theory, results showed that individuals searched longer when they expected more targets to be present and adjusted their expectations on-line during each search by taking into account the higher-order, across-trial target distributions. However, compared with modeled ideal observers, participants systematically responded as if the target distribution were more uniform than it was, which suggests that training could improve multiple-target search performance.     

False feedback increases detection of low-prevalence targets in visual search

Many critical search tasks, such as airport and medical screening, involve searching for targets that are rarely present. These low-prevalence targets are associated with extremely high miss rates Wolfe, Horowitz, & Kenner (Nature, 435, 439?C440, 2005). The inflated miss rates are caused by a criterion shift, likely due to observers attempting to equate the numbers of misses and false alarms. This equalizing strategy results in a neutral criterion at 50?% target prevalence, but leads to a higher proportion of misses for low-prevalence targets. In the present study, we manipulated participants?? perceived number of misses through explicit false feedback. As predicted, the participants in the false-feedback condition committed a higher number of false alarms due to a shifted criterion. Importantly, the participants in this condition were also more successful in detecting targets. These results highlight the importance of perceived prevalence in target search tasks.     

Satisfaction at last: Evidence for the “satisfaction” hypothesis for multiple-target search errors

ABSTRACT

Multiple-target visual searches are susceptible to Subsequent Search Miss (SSM) errors—a reduced accuracy for target detection after a previous target has already been detected. SSM errors occur in critical searches (e.g., evaluations of radiographs and airport luggage x-rays), and have proven to be a stubborn problem. A few SSM theories have been offered, and here we investigate the “satisfaction” account: failing to completely finish a search after having found a first target. Accuracy on a multiple-target search task was compared to both how long participants spent searching after finding a first target and their target sensitivity in a separate vigilance task. Less time spent searching and poor vigilance predicted higher SSM error rates. These results suggest that observers who are more likely to miss a second target are less likely to thoroughly search after finding a first target, thus offering some of the first evidence for the “satisfaction” account.     

Comparing target detection errors in visual search and manually-assisted search

Subjects searched for low- or high-prevalence targets among static nonoverlapping items or items piled in heaps that could be moved using a computer mouse. We replicated the classical prevalence effect both in visual search and when unpacking items from heaps, with more target misses under low prevalence. Moreover, we replicated our previous finding that while unpacking, people often move the target item without noticing (the unpacking error) and determined that these errors also increase under low prevalence. On the basis of a comparison of item movements during the manually-assisted search and eye movements during static visual search, we suggest that low prevalence leads to broadly reduced diligence during search but that the locus of this reduced diligence depends on the nature of the task. In particular, while misses during visual search often arise from a failure to inspect all of the items, misses during manually-assisted search more often result from a failure to adequately inspect individual items. Indeed, during manually-assisted search, over 90 % of target misses occurred despite subjects having moved the target item during search.     

Colour and spatial cueing in low-prevalence visual search

In visual search, 30-40% of targets with a prevalence rate of 2% are missed, compared to 7% of targets with a prevalence rate of 50% (Wolfe, Horowitz, & Kenner, 2005). This "low-prevalence" (LP) effect is thought to occur as participants are making motor errors, changing their response criteria, and/or quitting their search too soon. We investigate whether colour and spatial cues, known to improve visual search when the target has a high prevalence (HP), benefit search when the target is rare. Experiments 1 and 2 showed that although knowledge of the target's colour reduces miss errors overall, it does not eliminate the LP effect as more targets were missed at LP than at HP. Furthermore, detection of a rare target is significantly impaired if it appears in an unexpected colour-more so than if the prevalence of the target is high (Experiment 2). Experiment 3 showed that, if a rare target is exogenously cued, target detection is improved but still impaired relative to high-prevalence conditions. Furthermore, if the cue is absent or invalid, the percentage of missed targets increases. Participants were given the option to correct motor errors in all three experiments, which reduced but did not eliminate the LP effect. The results suggest that although valid colour and spatial cues improve target detection, participants still miss more targets at LP than at HP. Furthermore, invalid cues at LP are very costly in terms of miss errors. We discuss our findings in relation to current theories and applications of LP search.     

Spatial and temporal separation fails to counteract the effects of low prevalence in visual search

Recent research has shown that, in visual search, participants can miss 30–40% of targets when they only appear rarely (i.e., on 1–2% of trials). Low target prevalence alters the behaviour of the searcher. It can lead participants to quit their search prematurely (Wolfe, Horowitz, & Kenner, 2005), to shift their decision criteria (Wolfe et al., 2007), and/or to make motor or response errors (Fleck & Mitroff, 2007). In this paper we examine whether the low prevalence (LP) effect can be ameliorated if we split the search set in two, spreading the task out over space and/or time. Observers searched for the letter “T” among “L”s. In Experiment 1, the left or right half of the display was presented to the participants before the second half. In Experiment 2, items were spatially intermixed but half of the items were presented first, followed by the second half. Experiment 3 followed the methods of Experiment 2 but allowed observers to correct perceived errors. All three experiments produced robust LP effects with higher errors at 2% prevalence than at 50% prevalence. Dividing up the display had no beneficial effect on errors. The opportunity to correct errors reduced but did not eliminate the LP effect. Low prevalence continues to elevate errors even when observers are forced to slow down and permitted to correct errors.     

Spatial and temporal separation fails to counteract the effects of low prevalence in visual search

Recent research has shown that, in visual search, participants can miss 30-40% of targets when they only appear rarely (i.e. on 1-2% of trials). Low target prevalence alters the behavior of the searcher. It can lead participants to quit their search prematurely (Wolfe et al., 2005), to shift their decision criteria (Wolfe et al., 2007) and/or to make motor or response errors (Fleck & Mitroff, 2007). In this paper we examine whether the LP Effect can be ameliorated if we split the search set in two, spreading the task out over space and/or time. Observers searched for the letter "T" among "L"s. In Experiment 1, the left or right half of the display was presented to the participants before the second half. In Experiment 2, items were spatially intermixed but half of the items were presented first, followed by the second half. Experiment 3 followed the methods of Experiment 2 but allowed observers to correct perceived errors. All three experiments produced robust low prevalence (LP) effects with higher errors at 2% prevalence than at 50% prevalence. Dividing up the display had no beneficial effect on errors. The opportunity to correct errors reduced but did not eliminate the LP effect. Low prevalence continues to elevate errors even when observers are forced to slow down and permitted to correct errors.     

Prevalence-based decisions undermine visual search

In visual search, observers make decisions about the presence or absence of a target based on their perception of a target during search. The present study investigated whether decisions can be based on observers’ expectation rather than perception of a target. In Experiment 1, participants were allowed to make target-present responses by clicking on the target or, if the target was not perceived, a target-present button. Participants used the target-present button option more frequently in difficult search trials and when target prevalence was high. Experiment 2 and 3 employed a difficult search task that encouraged the use of prevalence-based decisions. Target presence was reported faster when target prevalence was high, indicating that decisions were, in part, cognitive, and not strictly perceptual. A similar pattern of responses were made even when no targets appeared in the search (Experiment 3). The implication of these prevalence-based decisions for visual search models is discussed.     

Infrequent identity mismatches are frequently undetected

The ability to quickly and accurately match faces to photographs bears critically on many domains, from controlling purchase of age-restricted goods to law enforcement and airport security. Despite its pervasiveness and importance, research has shown that face matching is surprisingly error prone. The majority of face-matching research is conducted under idealized conditions (e.g., using photographs of individuals taken on the same day) and with equal proportions of match and mismatch trials, a rate that is likely not observed in everyday face matching. In four experiments, we presented observers with photographs of faces taken an average of 1.5 years apart and tested whether face-matching performance is affected by the prevalence of identity mismatches, comparing conditions of low (10 %) and high (50 %) mismatch prevalence. Like the low-prevalence effect in visual search, we observed inflated miss rates under low-prevalence conditions. This effect persisted when participants were allowed to correct their initial responses (Experiment 2), when they had to verify every decision with a certainty judgment (Experiment 3) and when they were permitted “second looks” at face pairs (Experiment 4). These results suggest that, under realistic viewing conditions, the low-prevalence effect in face matching is a large, persistent source of errors.     

Does within-person variability predict errors in healthy adults aged 18–90?

This study investigated within-person variability on basic psychomotor tasks in relation to errors on a higher order cognitive task. We were interested in whether more variable individuals were more prone to making errors, and whether this relationship varied with age. Variability was assessed using simple and choice reaction time, while errors of omission (misses) and commission (false alarms) were obtained from simple and complex visual search tasks. Data from 557 participants aged 18–90 years were included in the analysis. Greater variability was associated with more misses, and distribution analyses showed that slower responses were behind this effect. Variability was also associated with false alarms, but the pattern was inconsistent. Taking age into account revealed that the association between variability and misses in the simple visual search condition was stronger in older (aged 65–90 years) participants. The results suggest the relationship between greater variability and errors of omission (misses) may be related to inattention. Measures of variability may therefore provide valuable insights into individual differences in error rates and, more broadly, may also offer early warning of persons who are more prone to errors in visual search.     

Robust representations for faces: evidence from visual search.

We report evidence from visual search that people can develop robust representations for highly overlearned faces. When observers searched for their own face versus the face of an unfamiliar observer, search slopes and intercepts revealed consistently faster processing of self than stranger. These processing advantages persisted even after hundreds of presentations of the unfamiliar face and even for atypical profile and upside-down views. Observers not only showed rapid asymptotic recognition of their own face as the target, but could reject their own face more quickly as the distractor. These findings suggest that robust representations for a highly overlearned face may (a) mediate rapid asymptotic visual processing, (b) require extensive experience to develop, (c) contain abstract or view-invariant information, (d) facilitate a variety of processes such as target recognition and distractor rejection, and (e) demand less attentional resources.     

Long-term visual associations affect attentional guidance

When observers perform a visual search task, they are assumed to adopt an attentional set for what they are looking for. The present experiment investigates the influence of long-term visual memory associations on this attentional set. On each trial, observers were asked to search a display for a grayscale version of a known traffic sign. On each trial, a distractor sign was drawn in full color. This color could either be related or unrelated to the target sign. Distractors interfered more with search when their color was related (e.g. red when the target was a stop sign), implying that long-term color associations resulted in inadvertent attentional guidance, even though color was irrelevant to the task. The results add to the growing body of evidence that long-term memory representations automatically affect attentional orienting.     

Hybrid value foraging: How the value of targets shapes human foraging behavior

In hybrid foraging, observers search visual displays for multiple instances of multiple target types. In previous hybrid foraging experiments, although there were multiple types of target, all instances of all targets had the same value. Under such conditions, behavior was well described by the marginal value theorem (MVT). Foragers left the current “patch” for the next patch when the instantaneous rate of collection dropped below their average rate of collection. An observer’s specific target selections were shaped by previous target selections. Observers were biased toward picking another instance of the same target. In the present work, observers forage for instances of four target types whose value and prevalence can vary. If value is kept constant and prevalence manipulated, participants consistently show a preference for the most common targets. Patch-leaving behavior follows MVT. When value is manipulated, observers favor more valuable targets, though individual foraging strategies become more diverse, with some observers favoring the most valuable target types very strongly, sometimes moving to the next patch without collecting any of the less valuable targets.

     

Dual-target search for high and low prevalence X-ray threat targets

Recent studies have shown that the prevalence of target presentation in visual search has an impact on target detection rates, and also that searching for two targets leads to a decrement in performance (the “dual-target cost”). In the present experiments, we added to previous work that has examined the interaction between the prevalence effect and dual-target cost, and also to previous work that has examined a full range of prevalence levels. We found no interaction between the dual-target cost and prevalence effect, though the response criterion in dual-target search did interact with target prevalence in an unexpected manner. Furthermore, we detected anasymmetry in the prevalence effect: It was more marked for high than low prevalence. Implications for real-world tasks are discussed.     

The contents of perceptual hypotheses: Evidence from rapid resumption of interrupted visual search

Observers can resume a previously interrupted visual search trial significantly more quickly than they can start a new search trial (Lleras, Rensink, & Enns, 2005). This rapid resumption of search is possible because evidence accumulated during the previous exposure, a perceptual hypothesis, can carry over to a subsequent presentation. We present four interrupted visual search experiments in which the content of the perceptual hypotheses used during visual search trials was characterized. These experiments suggest that prior to explicit target identification, observers have accumulated evidence about the locations, but not the identities, of local, task-relevant distractors, as well as preliminary evidence for the identity of the target. Our results characterize the content of perceptual search hypotheses and highlight the utility of interrupted search for studying online search processing prior to target identification.     

Color singleton pop-out does not always poop out: An alternative to visual search

Folk psychology suggests that when an observer views a scene, a unique item will stand out and draw attention to itself. This belief stands in contrast to numerous studies in visual search that have found that a unique target item (e.g., a unique color) is not identified more quickly than a nonunique target. We hypothesized that this finding is the result of task demands of visual search, and that when the task does not involve visual search, uniqueness will pop out. We tested this hypothesis in a task in which observers were presented an array of letters and asked to respond aloud, as quickly as possible, with the identity of any one of the letters. The observers were significantly more likely to respond with a uniquely colored letter than would be expected by chance. In a task in which observers blurt out the first thing that they see, unique pop-out does not poop out.     

Measuring the time course of selection during visual search

In visual search tasks, observers can guide their attention towards items in the visual field that share features with the target item. In this series of studies, we examined the time course of guidance toward a subset of items that have the same color as the target item. Landolt Cs were placed on 16 colored disks. Fifteen distractor Cs had gaps facing up or down while one target C had a gap facing left or right. Observers searched for the target C and reported which side contained the gap as quickly as possible. In the absence of other information, observers must search at random through the Cs. However, during the trial, the disks changed colors. Twelve disks were now of one color and four disks were of another color. Observers knew that the target C would always be in the smaller color set. The experimental question was how quickly observers could guide their attention to the smaller color set. Results indicate that observers could not make instantaneous use of color information to guide the search, even when they knew which two colors would be appearing on every trial. In each study, it took participants 200–300 ms to fully utilize the color information once presented. Control studies replicated the finding with more saturated colors and with colored C stimuli (rather than Cs on colored disks). We conclude that segregation of a display by color for the purposes of guidance takes 200–300 ms to fully develop.