Categorization training changes the visual representation of face identity.

Area of Science:

• Cognitive psychology
• Neuroscience
• Visual perception

Background:

• Category learning is known to alter perceptual representations of learned dimensions.
• However, the impact of category learning on the representation of specific face identities remains under-explored.

Purpose of the Study:

• To investigate how learning to categorize faces affects the internal representations used for identifying specific face identities.
• To determine if categorization training influences the stability of face identity representations across variations in irrelevant features.

Main Methods:

• Participants were trained to categorize faces varying along two dimensions (one relevant, one irrelevant).
• Reverse correlation was employed to derive internal templates for face identification before and after training.
• Templates were assessed at different levels of the irrelevant dimension.

Main Results:

• Categorization training significantly altered the internal templates used for face identification.
• Post-training templates demonstrated increased invariance to changes in the irrelevant face dimension.
• These changes occurred despite the distinct demands of categorization versus identification tasks.

Conclusions:

• Face identity representations are malleable and can be modified by categorization experience.
• Category learning can lead to more robust and stable internal representations of face identity.
• This suggests that the brain adapts face perception mechanisms based on task demands and learned categories.