A growing body of literature suggests that human individuals differ in their ability to process face identity. These findings mainly stem from explicit behavioral tasks, such as the Cambridge Face Memory Test (CFMT). However, it remains an open question whether such individual differences can be found in the absence of an explicit face identity task and when faces have to be individualized at a single glance. In the current study, we tested 49 participants with a recently developed fast periodic visual stimulation (FPVS) paradigm [Liu-Shuang, J., Norcia, A. M., & Rossion, B. An objective index of individual face discrimination in the right occipitotemporal cortex by means of fast periodic oddball stimulation. Neuropsychologia, 52, 57–72, 2014] in EEG to rapidly, objectively, and implicitly quantify face identity processing. In the FPVS paradigm, one face identity (A) was presented at the frequency of 6 Hz, allowing only one gaze fixation, with different face identities (B, C, D) presented every fifth face (1.2 Hz; i.e., AAAABAAAACAAAAD…). Results showed a face individuation response at 1.2 Hz and its harmonics, peaking over occipitotemporal locations. The magnitude of this response showed high reliability across different recording sequences and was significant in all but two participants, with the magnitude and lateralization differing widely across participants. There was a modest but significant correlation between the individuation response amplitude and the performance of the behavioral CFMT task, despite the fact that CFMT and FPVS measured different aspects of face identity processing. Taken together, the current study highlights the FPVS approach as a promising means for studying individual differences in face identity processing.

Concepts are composed of features related to different sensory and motor modalities such as vision, sound, and action. It is a matter of controversy whether conceptual features are represented in sensory-motor areas reflecting the specific learning experience during acquisition. In order to address this issue, we assessed the plasticity of conceptual representations by training human participants with novel objects under different training conditions. These objects were assigned to categories such that for one class of categories, the overall shape was diagnostic for category membership, whereas for the other class, a detail feature affording a particular action was diagnostic. During training, participants were asked to either make an action pantomime toward the detail feature of the novel object or point to it. In a categorization task at test, we assessed the neural correlates of the acquired conceptual representations by measuring electrical brain activity. Here, we show that the same object is differentially processed depending on the sensory-motor interactions during knowledge acquisition. Only in the pantomime group did we find early activation in frontal motor regions and later activation in occipito-parietal visual-motor regions. In the pointing training group, these effects were absent. These results show that action information contributes to conceptual processing depending on the specific learning experience. In line with modality-specific theories of conceptual memory, our study suggests that conceptual representations are established by the learning-based formation of cell assemblies in sensory-motor areas.