Abstract
Previous studies relying on small samples and using resting state functional MRI have shown that the fusiform face area (FFA) is connected to other face responsive regions (Turk-Browne et al., 2010; Zhang et al., 2009). Here, we used a sample of 296 participants to estimate connectivity maps from each region of the "core system" of face processing (Haxby et al., 2000) and quantify the coupling of these regions to make inferences about their organization. For each participant, we computed the correlation between the time series of seed regions corresponding to the face core system (FFA, occipital face area (OFA), and superior temporal sulcus (STS)) with that of each other voxel in the brain. Seed regions were defined based on peak MNI coordinates from previous studies. We averaged connectivity maps for each seed region across sub-samples of about 50 participants each. For each region of the face core system, we observed consistent connectivity maps across each independent sub-sample. We further estimated the magnitude of correlations between the core regions and found preferential coupling of the STS with the FFA, relative to the OFA, thus suggesting a revision of models (Haxby et al., 2000) of the organization of these regions. Finally, we tested seven participants using both resting-state fMRI and a functional localizer. Average connectivity maps from individually defined face-responsive regions in these participants were consistent to the maps estimated in the larger samples, and there was also preferential coupling of FFA and STS in comparison to that between OFA and STS. These analyses reveal the intrinsic connectivity among the FFA, OFA and STS, as well as the connections between each of these regions and the rest of the brain, providing novel insights into the organization of the face recognition system.
Meeting abstract presented at VSS 2013