Presentation Schedule


Presenter Registration Banner 5

Ensemble Bias of Facial Expression Shows Asymmetry for Angry and Happy Faces (105362)

Session Information:

Tuesday, 24 March 2026 16:00
Session: Poster Session 3
Room: Orion Hall (5F)
Presentation Type: Poster Presentation

All presentation times are UTC + 9 (Asia/Tokyo)

Ensemble perception enables observers to extract summary statistics from groups of objects, but it remains unclear whether the perceived expression of a single face is biased toward the average emotion of a heterogeneous facial ensemble. Prior evidence for such biases has largely come from ensembles containing only one emotional category. Here, we tested ensemble bias in mixed-emotion contexts and examined whether attentional allocation contributes to potential asymmetries between positive and negative expressions.

In an online experiment with 300 participants, each display contained one angry face, one happy face, and two neutral faces. Participants first performed a visual search task to locate an angry or happy target, then completed a same/different memory judgment for one of the previously shown faces. Across conditions, participants were more likely to judge a probe face as “same” when it was shifted toward the ensemble average, indicating reliable ensemble bias even in mixed-emotion arrays. Notably, this bias was larger for angry than for happy faces, whereas memory accuracy was higher for happy than for angry faces, revealing a negative–positive asymmetry. However, attentional measures did not correlate with bias magnitude, suggesting that attentional priority alone cannot account for the effect.

We propose that this asymmetry may reflect categorical perception of facial emotion: the anger–neutral boundary may be more perceptually diffuse than the happier and sharper happy–neutral boundary, making anger more susceptible to assimilation toward the ensemble mean.

Authors:
Jun Saiki, Kyoto University, Japan
Hiroki Yamada, Miidas Co., Ltd, Japan
Nobuyuki Jincho, Miidas Co., Ltd, Japan
Xiyan Ji, Kyoto University, Japan


About the Presenter(s)
Ms.Ji is curretly a doctoral student of cognitive psychology at Kyoto University.

See this presentation on the full scheduleTuesday Schedule



Conference Comments & Feedback

Place a comment using your LinkedIn profile

Comments

Share on activity feed

Powered by WP LinkPress

Share this Presentation

Posted by James Alexander Gordon

Last updated: 2023-02-23 23:45:00