Effective connectivity of the human cortical face network through concurrent intracerebral electrical stimulation and frequency-tagged visual presentation

Abstract

The neural basis of human face recognition has been extensively studied for decades. The contribution of several face-selective brain regions in the ventral occipito-temporal cortex (VOTC; e.g., in the fusiform gyrus, FG; inferior occipital gyrus, IOG) has been established, but the functional organization of this cortical network remains largely unknown. In rare epileptic patients implanted with intracerebral electrodes in several face-selective VOTC regions, an original combination of frequency-tagging and direct electrical stimulation (DES) can provide unique information to address this issue. These depth electrodes allow us to stimulate a local node of the network while recording the functional activity of other implanted regions, with high spatial and temporal resolution. Here we report the case of a 34-year-old man, presented with 70-second sequences of natural images (objects with one face inserted every five images, at a 6Hz rate), while focal stimulation (1,0mA at 55Hz) was separately applied for 10s to four independently defined face-selective areas in the right and left VOTC (right and left lateral FG, right IOG, right anterior FG). Upon stimulation, face-selective neural activity was reduced or abolished both locally and at remote bilateral VOTC recording sites. These remote effects of DES were found in both postero-anterior (i.e., DES to lateral occipital cortex affecting fusiform face-selective activity) and antero-posterior (DES to fusiform gyrus affecting lateral occipital cortex activity) directions as well as across face-selective sites of homologous regions of the two hemispheres. Most importantly, this reduction was extremely specific to the face-selective response, as it did not extend to the general visual response (i.e., 6Hz and harmonics). Overall, these results shed original light on the functional connectivity of the human cortical face recognition network and pave the way for a more widespread and systematic development of this approach to reveal the functional and effective connectivity of human brain networks.