Abstract
There is renewed interest in evaluating the extent to which face processing and its underlying neural circuitry is sexually dimorphic. We recently found no sex differences in face recognition behavior (Motta-Mena et al., under review) or in patterns of neural activation during face recognition as determined using univariate voxelwise and ROI-based analyses (Scherf et al., under revision). However, it could still be the case that men and women exhibit different patterns of neural functional connectivity during face recognition. To examine this possibility, we investigated potential sex differences in the topology of directed functional neural connections within the face-processing network in typically developing young adults. Participants completed a recognition task in the scanner in which they had to identify both a male and a female target face among separate blocks of male and female distractor faces. Core and extended regions in the face-processing network were defined using a separate localizer task. Effective connectivity was modeled separately for males and females during recognition of male and female faces using unified SEM. We quantified potential differences in global network topology using graph theory and pattern recognition metrics. Both male and female participants modulated the topology of directed functional connections as they shifted between recognizing male and female faces. During female face recognition, female participants exhibited a homogenous and unique pattern of connections that were not shared by the male participants, whose networks were more heterogeneous. In contrast, there were no sex differences in the organization of network topology during recognition of male faces. In general, this pattern of results reflects similar topological organization and modulation of the face-processing network for men and women during face recognition, which is consistent with our previous findings from univariate analyses.
Meeting abstract presented at VSS 2017