Abstract
It is often difficult to distinguish a stranger’s permanent facial shape from their transient facial expressions; for example, whether they are scowling or rapid character judgments we make about others. Someone with narrow eyes might be judged to be untrustworthy, because of strong associations between facial anger and threat. The emotion overgeneralization hypothesis suggests that judgments of character traits are based on subtle resemblance to emotional cues in others’ neutral faces. To test this hypothesis, we investigated the trait judgments made by individuals with severe alexithymia, associated with impaired recognition of facial emotion. Participants viewed emotionally neutral faces and rated them according to how trustworthy, aggressive, attractive and intelligent they appeared. Participants then rated the same faces according to subtle expressions of six basic emotions. Consistent with the emotion overgeneralization hypothesis, alexithymic participants demonstrated reduced inter-rater consistency when judging trustworthiness, aggressiveness and intelligence of unfamiliar faces, and the presence of subtle emotions. Judgments of attractiveness were more consistent in the alexithymic than control group. Nevertheless, where alexithymics perceived, or misperceived, emotion cues, the character traits inferred thereafter were broadly typical. The finding that individuals with developmental deficits of emotion recognition exhibit atypical attribution of character traits, confirms the hypothesis that emotion recognition mechanisms play a causal role in character judgments. For attractiveness inferences, which may be more objectively made using non-emotional cues, such as symmetry and averageness, it seems that alexithymic individuals are able to use these cues more reliably, due to being influenced by emotional cues less. The fact that trait inferences were made in line with judgments of emotion in both groups suggests that all individuals base trait inferences on subtle emotional cues, regardless of their emotion recognition accuracy.
Meeting abstract presented at VSS 2015