Adaptation reveals that facial expression averaging occurs during rapid serial presentation

Area of Science:

• Cognitive Psychology
• Neuroscience
• Visual Perception

Background:

• The visual system processes complex stimuli, but how it interprets rapidly changing faces remains unclear.
• Visual adaptation is a key method for studying short-term plasticity in visual perception.
• Ensemble coding and integration are proposed mechanisms for interpreting multiple faces presented temporally.

Purpose of the Study:

• To investigate if ensemble coding or averaging occurs when viewing emotional faces in rapid serial visual presentation (RSVP).
• To determine if adaptation aftereffects from RSVP faces match those from statistically averaged faces.
• To explore how variations in emotional faces within a stream affect perception.

Main Methods:

• Utilized a visual adaptation paradigm with rapid serial visual presentation (RSVP) of distinct emotional faces.
• Measured facial expression aftereffects (FEAs) induced by RSVP face streams and their averaged counterparts.
• Conducted four experiments manipulating emotion, temporal frequency, and face identity.

Main Results:

• RSVP of emotional faces induced significant facial expression aftereffects (FEAs) for happy and sad emotions.
• The magnitude of FEAs from RSVP faces was comparable to and correlated with those from averaged faces.
• FEAs were influenced by the mean emotion of the face stream, not individual face variations or temporal frequency.
• The average emotion of the stream, not individual identities, determined the FEAs.

Conclusions:

• The visual system employs ensemble coding to interpret rapidly presented faces, suggesting an averaging mechanism.
• This process implies the formation of a 'facial expression norm' within the visual system's face space.
• Findings support the role of averaging in perceiving emotional expressions from dynamic visual input.