Ambiguous at the second sight: Mixed facial expressions trigger late electrophysiological responses linked to lower social impressions

Abstract

Social interactions require quick perception, interpretation, and categorization of faces, with facial features offering cues to emotions, intentions, and traits. Importantly, reactions to faces depend not only on their features but also on their processing fluency, with disfluent faces suffering social devaluation. The current research used electrophysiological (EEG) and behavioral measures to explore at what processing stage and under what conditions emotional ambiguity is detected in the brain and how it influences trustworthiness judgments. Participants viewed male and female faces ranging from pure anger, through mixed expressions, to pure happiness. They categorized each face along the experimental dimension (happy vs. angry) or a control dimension (gender). In the emotion-categorization condition, mixed (ambiguous) expressions were classified relatively slower, and their trustworthiness was rated relatively lower. EEG analyses revealed that early brain responses are independent of the categorization condition, with pure faces evoking larger P1/N1 responses than mixed expressions. Some late (728-880 ms) brain responses from central-parietal sites also were independent of the categorization condition and presumably reflect familiarity of the emotion categories, with pure expressions evoking larger central-parietal LPP amplitude than mixed expressions. Interestingly, other late responses were sensitive to both expressive features and categorization task, with ambiguous faces evoking a larger LPP amplitude in frontal-medial sites around 560-660 ms but only in the emotion categorization task. Critically, these late responses from the frontal-medial cluster correlated with the reduction in trustworthiness judgments. Overall, the results suggest that ambiguity detection involves late, top-down processes and that it influences important social impressions.

Effects of emotion and emotional ambiguity on early visual event-related potentials.

“…However, the emotional facial expressions used in the latter study were presented at higher intensities (up to 100%) than in the present study (maximum of 74%), so it is possible that our stimuli were not sufficiently intense or unambiguous to modulate the early sensory ERPs such as the P1 and N1. Similarly to our study, the N170 in the study of Kaminska et al (2020) was not influenced by emotional ambiguity. It is thus possible that this component reflects the perception of faces, regardless of the ambiguity of their emotional expressions.…”

“…For the N170, we hypothesized that it would also be affected by emotion but had no assumptions as to which specific emotions would influence this component considering the conflicting results of previous studies (Durston & Itier, 2021;Müller-Bardorff et al, 2016;Recio et al, 2014;Rellecke et al, 2011Rellecke et al, , 2012Schindler et al, 2019;Turano et al, 2017). Finally, we had no a priori hypotheses as to whether or not the P1, N1, and/or N170 would be modulated by emotional ambiguity given that only one study has examined the effect of this factor on early sensory ERPs (Kaminska et al, 2020).…”

What’s in a face? The role of facial features in ratings of dominance, threat, and stereotypicality

“…These judgments may play a role in how people are perceived and may relate to important applied decisions, such as political elections (Todorov et al, 2005), military rank (Mazur et al, 1984;Mueller & Mazur, 1998), and court system outcomes relating to sentence severity and guilty verdicts (Blair et al, 2004a(Blair et al, , 2004bKleider-Offutt et al, 2017a, 2017bPorter et al, 2010). These face trait judgments occur for race-and gender-ambiguous faces, suggesting that susceptibility to biased assessment may be ubiquitous (Ito et al, 2011;Kaminska et al, 2020). However, in scientific research and the news media, Black faces specifically garner biased judgment (Dixon, 2017;Dixon & Azocar, 2007;Kleider-Offutt, 2019;Kleider-Offutt et al, 2017a, 2017b.…”

From likely to likable: The role of statistical typicality in human social assessment of faces

“…Interestingly, reducing the average negative LL of observed data under the assumed model, known as cross-entropy minimization, is also a popular and effective tool for representation learning in modern machine learning and artificial intelligence (46). Separately, empirical evidence indicates that face stimuli whose representation in the brain requires greater neural activity are indeed aversive (47)(48)(49).…”