Abstract
Introduction. The ability to recognize familiar faces across different viewing conditions contrasts with the inherent difficulty found in the perception of unfamiliar faces across similar image manipulations. We used fMR-adaptation to ask whether this difference in recognition might be reflected by an image-invariant representation for familiar faces and an image-dependent representation for unfamiliar faces in face-selective regions. Methods. In the first experiment, 20 participants were scanned while viewing images of faces in a blocked design. Each block contained 8 face images with the same identity. There were 4 conditions: (1) 8 repetitions of the same face image, (2) 4 repetitions of 2 face images, (3) 2 repetitions of 4 face images, (4) 1 repetition of 8 face images. In the second experiment, we used the same design but the images were taken from different identities. fMRI responses were evaluated statistically in regions of interest that were identified independently by a localizer scan for each participant. Results. In the first experiment, we found a gradual release from adaptation in the FFA with increases in the number of face images within a block. For example, there was no significant difference between the response of the FFA to the repetition of the same face image and the response to 2 face images with the same identity. In contrast, in the second experiment, we found there was a complete release from adaptation in the FFA when 2, 4 or 8 face images with different identities were presented within a block. Conclusion. Although these results are consistent with an image-invariant representation of faces in the FFA, we found no difference in the pattern of response to familiar and unfamiliar faces. This suggests that differences in the perception of familiar and unfamiliar faces may not depend on differences in image invariance within regions such as the FFA.