Ensemble representation of animacy could be based on mid-level visual features.

Area of Science:

• Cognitive Science
• Visual Perception
• Psychology

Background:

• Object animacy perception is crucial for understanding the environment.
• Previous research suggests mid-level features contribute to object animacy.
• The role of feature level in ensemble animacy perception remains unclear.

Purpose of the Study:

• To investigate whether ensemble animacy perception is based on high-level or mid-level visual features.
• To determine if mid-level features alone are sufficient for ensemble animacy judgment.

Main Methods:

• Utilized five image types: color, grayscale, silhouettes, texforms (preserving mid-level texture/shape), and scrambled images.
• Conducted Experiments 1 (10-point animacy rating for single/set images) and 2 (2-alternative forced-choice task for set comparison).
• Employed behavioral tasks to assess participants' perception of animacy from different image representations.

Main Results:

• Participants could accurately report the mean animacy of texform image sets.
• This perception occurred without explicit access to high-level object features.
• Results were consistent across both rating and forced-choice paradigms.

Conclusions:

• Ensemble animacy perception can be effectively driven by mid-level visual features, specifically shape and texture.
• High-level object information is not necessary for judging the average animacy of a set of images.
• The findings support the hypothesis that mid-level features are fundamental to ensemble animacy processing.