Fusiform Gyrus Face Selectivity Relates to Individual Differences in Facial Recognition Ability

Abstract

Regions of the occipital and temporal lobes, including a region in the fusiform gyrus (FG), have been proposed to comprise a “core” visual representation system for faces, in part because they show face selectivity and face repetition suppression. But recent fMRI studies of developmental prosopagnosics (DPs) raise questions about whether these measures relate to face processing skills. Although DPs manifest deficient face processing, most studies to date have not shown unequivocal reductions of functional responses in the proposed core regions. We scanned 15 DPs and 15 non-DP control participants with fMRI while employing factor analysis to derive behavioral components related to face identification or other processes. Repetition suppression specific to facial identities in FG or to expression in FG and STS did not show compelling relationships with face identification ability. However, we identified robust relationships between face selectivity and face identification ability in FG across our sample for several convergent measures, including voxel-wise statistical parametric mapping, peak face selectivity in individually defined “fusiform face areas” (FFAs), and anatomical extents (cluster sizes) of those FFAs. None of these measures showed associations with behavioral expression or object recognition ability. As a group, DPs had reduced face-selective responses in bilateral FFA when compared with non-DPs. Individual DPs were also more likely than non-DPs to lack expected face-selective activity in core regions. These findings associate individual differences in face processing ability with selectivity in core face processing regions. This confirms that face selectivity can provide a valid marker for neural mechanisms that contribute to face identification ability.

Quantifying interindividual variability and asymmetry of face-selective regions: A probabilistic functional atlas

“…As for the functional features, the FSRs showed significant variability in their selectivity for faces, consistent with previous studies (Furl et al, 2011;Gauthier et al, 2005;Yovel and Kanwisher, 2005). For the spatial features, to our knowledge, only a few studies have provided quantitative estimates of intersubject variability of the FSRs, mostly focusing on the variability of the location of peak activation.…”

“…Such interindividual functional variability likely provides a neural basis for interindividual differences in face recognition. For example, individual differences in face selectivity of the OFA and FFA predict differences in behavioral performance in face recognition (Furl et al, 2011;Huang et al, 2014). In addition, differences in the volume of the FFA were found to account for the developmental differences in face recognition memory (Golarai et al, 2007).…”

Reduced sensitivity to contrast signals from the eye region in developmental prosopagnosia

“…This is consistent with fMRI findings which show that DPs have relatively normal levels of activation within the face-selective regions of the core face processing network that are assumed to generate the N170 (Avidan & Behrmann, 2009;Avidan, Hasson, Malach & Behrmann, 2005;Furl, Garrido, Dolan, Driver, & Duchaine, 2011), and also with the fact that DPs do not have problems distinguishing faces from non-face objects. However, the presence of face-sensitive N170 components in DP does not necessarily imply that perceptual face processing mechanisms operate in exactly the same way in DPs and in control participants with unimpaired face recognition abilities.…”

Face identity matching is selectively impaired in developmental prosopagnosia

“…This has been suggested by studies showing that DPs are relatively normal in their ability to recognise categorically distinct basic emotions (Duchaine, Parker, & Nakayama, 2003;Humphreys, Avidan, & Behrmann, 2007;Palermo et al, 2011), more subtle and complex expressions (Duchaine et al, 2003;Duchaine, et al, 2007;Palermo et al, 2011) and are also able to successfully complete expression matching tasks (Bentin, DeGutis, D'Esposito, & Robertson, 2007;Garrido et al, 2009;Lee et al, 2010). DPs also show typical neural responses to emotional versus neutral faces (Avidan et al, 2014;Dinkelacker et al, 2011, Furl et al, 2011Van den Stock et al, 2008;. However, some DPs do report having difficulty reading expression in their daily lives (e.g.…”

“…Neuroimaging studies of DP have shown that in contrast to face recognition disorders caused by brain injury (acquired prosopagnosia; Bodamer, 1947), the occipito-temporal "core" face processing network (e.g., Haxby, Hoffman, & Gobbini, 2000;Kanwisher, McDermott, & Chun, 1997) appears to be largely intact in DP (Avidan & Behrmann, 2009;Avidan, Hasson, Malach, & Behrmann, 2005;Avidan et al, 2014;Furl, Garrido, Dolan, Driver, & Duchaine, 2011;Hasson, Avidan, Deouell, Bentin, & Malach, 2003; but see also Berhmann, Avidan, Gao, & Block, 2007;Garrido et al, 2009). However, investigations of face-specific event-related potential (ERPs) in DP are now beginning to reveal systematic differences between DPs and control participants, both at early visual-perceptual stages of face processing, and at later memory-related stages associated with the recognition of facial identity (see Towler, Fisher, & Eimer, in press, for review).…”