Abstract
Face pareidolia is the common misperception of illusory facial features in otherwise inanimate objects. Previously we showed that rhesus monkeys (Macaca mulatta) perceive illusory faces in the same images that humans do, using a free-viewing paradigm in which pairs of visual stimuli were presented to subjects (Taubert et al., 2017, Current Biology). The aim of this study was to test the hypothesis that emotional valence will mediate the attention of monkeys to different illusory faces. To this end, we created a stimulus set whereby each example of an illusory face was matched to both a monkey face, in terms of its expression (e.g. low valence = neutral expressions and high valence = aggressive or fearful expressions), and a non-face object, in terms of its object identity (e.g. a pie, a bell pepper, or a flower). Thirty stimuli were presented to two monkeys in pairs. There were three trial types containing the following stimulus pairs: Face / Illusory Face trials, Face / Non-Face Object trials, and Illusory Face / Non-Face Object trials. These trial types were presented to the subjects in a random order. The results revealed that when the monkeys were presented Face / Non-Face Object trials, the emotional valence of the face did not systematically alter viewing behavior. In both the high and low valence conditions the subjects tended to first, more frequently, and for longer periods of time look at the conspecific faces. In contrast, the emotional valence of the illusory faces in the Illusory Face / Non-Face Object trials did alter viewing behavior. When the illusory faces depicted aggressive or fearful emotional expressions, they summoned more attention (i.e. an increased number of fixations) than when they depicted neutral facial expressions. These results indicate that the emotional cues conveyed by illusory faces modulate the free-viewing behavior of monkeys.