Activity in the right fusiform face area predicts the behavioural advantage for the perception of familiar faces

People are extremely proficient at discriminating the identity of familiar faces, but are significantly worse with unfamiliar faces. Despite this clear behavioural difference in perception, the neural correlates of the advantage for familiar faces remain unclear. Here, we use an individual differences approach to explore the link between neural responses in face-selective regions and the behavioural advantage for the perception of familiar faces. First, we compared performance on an identity matching task with either familiar or unfamiliar faces. We found that participants were significantly better at matching the identity of familiar compared to unfamiliar faces. Next, we used fMRI to measure the response to familiar and unfamiliar faces. Consistent with the behavioural data, there was a significant difference in the neural response to familiar and unfamiliar faces in face-selective regions. Finally, we asked whether interindividual variation in behavioural performance could be predicted by corresponding variation in fMRI response. We found a significant correlation in the right fusiform face area (rFFA) between the difference in response to familiar and unfamiliar faces and corresponding differences on the face-matching task. That is, participants who showed a larger response to familiar compared to unfamiliar faces in the rFFA also matched familiar faces much more accurately than unfamiliar faces. No other face-selective region showed a correlation between neural and matching accuracy. These results provide a link between activity in the rFFA and the perception of familiar faces.