Ensemble Coding of Crowd with Cross-Category Facial Expressions

Area of Science:

• Cognitive Psychology
• Visual Perception
• Computational Neuroscience

Background:

• Ensemble coding enables averaging representations of element sets.
• The capacity for cross-category ensemble coding remains debated, with prior low-level stimulus studies yielding inconsistent results.

Purpose of the Study:

• To investigate if observers can extract average information from cross-category, high-level visual stimuli.
• To clarify the role of perceptual distance in cross-category ensemble coding.

Main Methods:

• Presented high-level stimuli (crowds of facial expressions) simultaneously and sequentially.
• Utilized a member judgment task to assess participants' ability to perceive average information.

Main Results:

• Participants successfully extracted average information from cross-category facial expressions.
• Effective cross-category ensemble coding was observed when perceptual distances between stimuli were short.

Conclusions:

• Demonstrates the existence of cross-category ensemble coding for high-level stimuli.
• Highlights the influence of perceptual distance on averaging diverse visual information.
• Provides foundational insights for future research on ensemble coding and visual cognition.