Purchase this article with an account.
Vadim Axelrod, Galit Yovel; Invariant representation of face identity in the fusiform face area (FFA): The effect of external facial information. Journal of Vision 2008;8(6):404. doi: https://doi.org/10.1167/8.6.404.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Several studies have shown that the fusiform face area (FFA) codes the identity of individual faces. The extent to which this representation is invariant to various face changes is still unclear. Here we examined the effect of external facial information such as hair or a cap on the representation of internal facial features. External information is usually excluded in most face studies, but it may have a significant influence on the representation of the internal face features. In our study we presented pairs of faces of same or different identity in three different variations: in a cap condition faces wear a cap that covers their hair, in the hair (no-cap) condition the hair is shown and in a cap-hair condition one face had a cap on and the other did not. We used fMR-adaptation to examine whether identity representation is influenced by external information. Consistent with previous studies we found a robust release from adaptation for the cap and hair conditions in the FFA. However, there was no adaptation for faces of same identity that differ in external information (cap-hair condition). Interestingly, our behavioral data show that although identity discrimination for faces that differ in external information (cap-hair condition) is somewhat lower than when external information is similar (cap & hair conditions), discrimination level is still very high. We suggest two possible interpretations to the lack of invariance to face identity in the FFA: either the FFA is sensitive to any facial information rather than just the identity of internal facial features or that inconsistent external information (hair-cap) may modify the perceived identity of the internal features (e.g. a composite effect). To decide between these two possibilities we currently conduct the same experiment with faces in which the top hair/cap part is misaligned with respect to the internal features.
This PDF is available to Subscribers Only