The face network: overextended? (Comment on: "Let's face it: It's a cortical network" by Alumit Ishai)

We offer a critique of Ishai's [Ishai, A., 2008. Let's face it: it's a cortical network. NeuroImage. doi:10.1016/j.neuroimage.2007.10.040] comment on the value of considering the brain areas that support face perception as a network. We emphasise that this idea is not in opposition to the notion that the fusiform gyrus plays a key role in the visual analysis of faces. More important, we argue that the definition offered of the "extended" face network--areas showing a greater fMRI response to intact than scrambled face images--is too inclusive, and present data to indicate that at least two of the proposed "nodes" of this network also respond to non-face objects (compared to scrambled controls). Finally, we consider briefly how converging methodological approaches may augment the use of fMRI alone in understanding how anatomically widespread brain areas coordinate their activity in order to make sense of the human face.