Abstract
Background: The “anger-superiority hypothesis” states an angry facial expression is detected faster than a non-threatening facial expression when either is embedded in a crowd of neutral faces. Presumably, this occurs because an angry face signals impending direct-threat to the observer. However, it is not known whether angry facial expressions are detected more efficiently than other negative expressions that signal danger. Objective: In this study we compared search efficiency for angry and fearful expressions. Methods: Subjects completed a visual search task in which the goal was to detect either an angry or fearful facial expression amid a crowd of neutral expression distractor faces. Crowd size (small vs. large) and view (frontal vs. side) were manipulated. As an angry face signals direct-threat, but a fearful face only signals indirect threat, we expected better search efficiency for angry expressions. Furthermore, we hypothesized angry faces would be more efficiently detected in frontal-view, given that this signals a direct threat from the image to the observer, but that there would be no effect of view on fearful faces. Results: Subjects were faster and more efficient in detecting fearful than angry faces. Target interacted with crowd size: performance declined with increasing crowd size for angry targets, crowd size did not affect performance for fearful targets. We confirmed that angry targets were detected more efficiently in frontal-view, but view had no effect for fearful targets. Conclusions: Contrary to the anger-superiority hypothesis, we found even better performance for fearful faces. This may indicate paradoxically that an indirect threat, possibly shared by the subject and observer, is more salient than a direct-threat. Detection of direct threat is greatest in a view where the threat can be perceived as directed at the observer, but a similar view effect is not seen for the indirect-threat represented by fearful faces.
Meeting abstract presented at VSS 2015