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

Abstract

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, 439-440, 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.Keywords Signal detection theory . Visual search . Low prevalence . FeedbackVisual search tasks are something we partake in every day. While some searches are rather trivial in nature (e.g., looking for the shirt that we want to wear, finding food in the refrigerator, or locating our car keys), other search tasks play a vital role in our wellbeing. Airport security, radiologists, and military personnel all perform critical visual search tasks that can have serious repercussions if the targets that they are searching for are not detected. Just recently, security officials at LAX failed to detect a loaded gun in a handbag (Blankstein & Sewell, 2011) and caused a scare when they mistook an insulin pump for a gun (Blankstein, 2012). These critical search tasks are more difficult when the targets are rare (i.e., have a low prevalence rate), as is often the case. The likelihood of missing a target is substantially higher for low-prevalence targets, a finding termed the low-prevalence (LP) effect (Wolfe, Horowitz, & Kenner, 2005). Wolfe et al. (2005) found that target miss rates were only 7 % when a target appeared in 50 % of the trials, but miss rates increased to 30 % when a target appeared in only 1 % of the trials. This effect has serious implications in a critical search task such as medical screening, where the prevalence of a target can be less than 1 % (Fenton et al., 2007).Analyses using signal detection theory (SDT; Green & Swets, 1966) have revealed that the LP effect is the result of a criterion shift rather than of a loss in sensitivity (Wolfe et al., 2007;Wolfe & Van Wert, 2010). As the prevalence of a target decreases, observers become biased against the "target detected" response. Some evidence has also suggested that a speed-accuracy trade-off could contribute to the LP effect (Fleck & Mitroff, 2007), but further research revealed that the speed-accuracy trade-off was primarily responsible for misses due to motor-response errors, not misses resulting from a criterion shift (Rich et al., 2008;Van Wert, Horowitz, & Wolfe, 2009). Wolfe et...

Prevalence-based decisions undermine visual search

“…When observers were falsely informed that their miss rates were higher than PREVALENCE-BASED DECISIONS 543 was actually the case, they were able to shift their criteria to more optimal levels and find more targets. Both of these results indicate that implicit feedback (identification of the target; Wolfe et al, 2007) and explicit feedback (provided after each trial; Schwark et al, 2012) are used by observers to estimate target prevalence.…”

“…This criterion shift impacts search times and appears to be due to the observer's attempt to equate the number of false alarms and misses that they commit (Schwark et al, 2012;Wolfe et al, 2007;Wolfe & Van Wert, 2010). Low target prevalence results in a more conservative decision criterion and shorter search times since there is little need to search extensively for a target that is unlikely to be present.…”

“…A scoring system identical to those used in previous studies (Schwark et al, 2012;Wolfe et al, 2007) was used to discourage misses and imitate Figure 1. Sample target-present stimulus (easy difficulty) used in Experiment 1.…”

“…This feedback presumably led observers to (falsely) believe that the overall prevalence of the target was higher. False feedback can also mediate the LP effect (Schwark et al, 2012). When observers were falsely informed that their miss rates were higher than PREVALENCE-BASED DECISIONS 543 was actually the case, they were able to shift their criteria to more optimal levels and find more targets.…”

“…According to SDT, estimates of target prevalence are used when observers are setting their decision criteria, but do not influence an observer's sensitivity to the target (Schwark, Sandry, MacDonald, & Dolgov, 2012;Wolfe et al, 2007). This criterion shift impacts search times and appears to be due to the observer's attempt to equate the number of false alarms and misses that they commit (Schwark et al, 2012;Wolfe et al, 2007;Wolfe & Van Wert, 2010).…”