Cortical areas involved in facial physical features and face identity processing: preliminary data by rTMS

Face recognition is enabled through a distributed neural network but the relative contribution of the single components of this network has not been clarified. Functional neuroimaging data suggest a role of faceselective occipito-temporal region in the perception of facial physical features (occipital face area, OFA) and invariant aspects of faces such as face identity (posterior superior temporal sulcus, pSTS). We used an on-line interference appoach by focal rTMS to invesigate the role of the OFA and pSTS in face processing. Methods. Fifteen healthy volunteers participated in a match-to-sample discrimination task. Stimuli were manip- ulated by morphing technique that changes (morphs) one image into another at different levels. Two faces were successively presented for 250 ms: a match-face and a target face. The target face could be the same (no morphing), different in terms of physical features (low morphing), or different both for physical features and identity (high morphing) compared to the match face. Focal 20-Hz rTMS at an intensity of 90% resting motor threshold was delivered simultaneously to the target face presentation (250 ms). Four blocks of 45 pairs of stimuli (15 for each level of morphing) were randomly presented in four different experimental conditions: 1) baseline (without rTMS); 2) rTMS of the right OFA; 3) rTMS of the right pSTS; and 4) rTMS over Cz (cortical control area). Dependent variables were accuracy (percentage of correct response), reaction times (RT) and the response confidence on a scale from 1 to 3. Results. rTMS of the right OFA did not produce signifi- cant effect on accuracy ad RT. The interference with the right pSTS function produced a slight worsening of the accuracy in the ‘high morphing’ condition (75 ± 12%) compared to baseline (82 ± 16%), rTMS of the right OFA (85 ± 15%) and rTMS over Cz (83 ± 12%) conditions. However, this effect did not reach statistical significance. Conclusion. These preliminary data suggest that OFA could not play a crucial role in the holistic face process- ing. The present task implied a global stimulus analysis that likely involves neural regions at higher level of the face processing stream. In contrast, OFA is considered to be involved in the early processing of face parts. The evaluation of a larger number of subjects is necessary to clarify the role of the pSTS in this task.