Successful decoding of famous faces in the fusiform face area

Abstract

What are the neural mechanisms of face recognition? It is believed that the network of face-selective areas, which spans the occipital, temporal, and frontal cortices, is important in face recognition. A number of previous studies indeed reported that face identity could be discriminated based on patterns of multivoxel activity in the fusiform face area and the anterior temporal lobe. However, given the difficulty in localizing the face-selective area in the anterior temporal lobe, its role in face recognition is still unknown. Furthermore, previous studies limited their analysis to occipito-temporal regions without testing identity decoding in more anterior face-selective regions, such as the amygdala and prefrontal cortex. In the current high-resolution functional Magnetic Resonance Imaging study, we systematically examined the decoding of the identity of famous faces in the temporo-frontal network of face-selective and adjacent non-face-selective regions. A special focus has been put on the face-area in the anterior temporal lobe, which was reliably localized using an optimized scanning protocol. We found that face-identity could be discriminated above chance level only in the fusiform face area. Our results corroborate the role of the fusiform face area in face recognition. Future studies are needed to further explore the role of the more recently discovered anterior face-selective areas in face recognition.

Fig 1. Face-selective areas of one representative participant (inflated cortex, right hemisphere).

(A) Lateral brain view: posterior STS and prefrontal face-selective areas. (B) Ventral brain view: FFA, ATL face-area and amygdala face-selective areas.

Fig 2. The face and cup stimuli used in the experiment.

(A) Eight images of Benjamin Netanyahu (the prime minister of Israel) and eight images of Shimon Peres (the president of Israel). (B) Eight images of two cup types.

Fig 3. Region of Interest-based face identity discrimination analysis (Benjamin Netanyahu and Shimon Peres identities).

(A) Average percent signal change for two face identities in the different face-selective areas and non-face selective collateral sulcus area. Error bars denote standard error of the mean. (B) Classification rates between face identities in face and non-face selective regions. The black line indicates a chance level of 50%. The error bars denote the standard error of the mean.

Fig 4. Individual classification rates of identity discrimination analysis in the right FFA.

Fig 5. Classification rates in face-selective (FFA, pSTS, ATL face-area, prefrontal face-area, amygdala) and non-face selective collateral sulcus area for different ROI sizes (10, 20, 30, 40 and 50 voxels).

The black line indicates a chance level of 50%. The error bars denote the standard error of the mean.

Fig 6. Region of Interest discrimination analysis for cups.

(A) Average percent signal change for two cup types in the face-selective areas and non-face selective collateral sulcus area. Error bars denote standard error of the mean. (B) Classification rates between two cup types in face and non-face selective regions. The black line indicates a chance level of 50%. The error bars denote the standard error of the mean.

Fig 7. Region of Interest-based face identity discrimination analysis (replication experiment; Leonardo DiCaprio and Brad Pitt identities).

(A) Average percent signal change for two face identities in the different face-selective areas and non-face selective collateral sulcus area. Error bars denote standard error of the mean. (B) Classification rates between face identities in face and non-face selective regions. The black line indicates a chance level of 50%. The error bars denote the standard error of the mean.

Fig 8. Comparison of individual classification rates of identity discrimination in the right FFA for the original (Benjamin Netanyahu and Shimon Peres) experiment and the replication (Leonardo DiCaprio and Brad Pitt).