The effect of perceptual adaptation to dynamic facial expressions

We present three experiments investigating the perceptual adaptation to dynamic facial emotional expressions. Dynamic expressions of six basic emotions were obtained by video recording of a poser’s face. In Experiment 1 participants (n=20) evaluated the intensity of 6 emotions, neutral state, genuineness and naturalness of dynamic expressions. The validated stimuli were further used as adaptors in Experiments 2 and 3 aimed at exploring the structure of facial expressions perceptual space by adaptation effects. In Experiment 2 participants (n=16) categorized neutral/emotion morphs after adaptation to dynamic expressions. In Experiment 3 (n=26) the task of the first stage was to categorize static frames derived from video records of the poser. Next individual psychometric functions were fitted for each participant and each emotion, to find the frame with emotion recognized correctly in 50% trials. These latter images were presented on the second stage in adaptation experiment, with dynamic video records as adaptors. Based on the three experiments, we found that facial expressions of happiness and sadness are perceived as opponent emotions and mutually facilitate the recognition of each other, whereas disgust and anger, and fear and surprise are perceptually similar and reduce the recognition accuracy of each other. We describe the categorical fields of dynamic facial expressions and of static images of initial phases of expression development. The obtained results suggest that dimensional and categorical approaches to perception of emotions are not mutually exclusive and probably describe different stages of face information processing. The study was supported by the Russian Foundation for Basic Research, project № 15-36-01281 “Structure of dynamic facial expressions perception”.

Abstract licence: CC BY-NC

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.