What's behind a face: person context coding in fusiform face area as revealed by multivoxel pattern analysis

Area of Science:

• Cognitive Neuroscience
• Neuroimaging
• Human Perception

Background:

• Face identification involves visual features and conceptual knowledge.
• Existing research often overlooks the dissociation between these processing types.
• The fusiform face area (FFA) is typically studied for its role in face identity recognition.

Purpose of the Study:

• To investigate conceptual face processing independently of visual facial input.
• To determine if the fusiform face area (FFA) processes semantic contexts.
• To explore the role of the FFA beyond mere facial feature detection.

Main Methods:

• Compared brain activity using word-based semantic categories versus person-specific conceptual contexts.
• Utilized an 8x8 word matrix to ensure visual input was consistent across conditions.
• Employed univariate statistical contrasts and a machine learning classification algorithm.

Main Results:

• Univariate analyses showed no significant differences in FFA activation between the two conditions.
• Machine learning successfully classified distinct semantic contexts based on FFA response patterns.
• FFA activation patterns encode different semantic contexts, indicating a role beyond visual processing.

Conclusions:

• The fusiform face area (FFA) demonstrates the capacity for processing abstract semantic information.
• FFA's functional role extends beyond visual face recognition to include conceptual understanding.
• Findings challenge the view of FFA as solely dedicated to facial feature processing.