Scalp topography and analysis of intracranial sources of face-evoked potentials.

Several reports have described that a positive vertex peak of an evoked potential varied in amplitude and latency specifically when images of faces were the eliciting stimulus. The scalp topography and the possible underlying dipole sources of this peak are the subject of this report. We presented black-and-white photographs of human faces, flowers and leaves to 16 healthy subjects and recorded the evoked brain potentials from 31 scalp electrodes. We found the previously described higher amplitude of the positive vertex peak when faces were the crucial stimulus, but the latency of this peak was the same (180 ms) for all three categories of stimulus. At the posterior temporal electrodes, the face waveforms showed a negative peak at 175 ms, which was only rudimentary in the waveforms elicited by the other stimuli. Since in most previous reports a mastoid reference was used, it is most likely that the previously described latency shift of the positive vertex peak associated with face stimuli was due to the interaction with this posterior temporal peak. The dipole analysis of the possible generators of the recorded potentials suggested the sequential activation of occipital, lateral temporal and mesio-temporal brain structures during the perception of a human face.