Daniel Fiset¹ (), Kim Calvé¹, Jérémy Lamontagne¹, Marie-Claude Desjardins¹, Mélodie Potvin-Poirier¹, Caroline Blais^1,2, Sara Tremblay; ¹Université du Québec en Outaouais, ²The Royal’s Institute of Mental Health Research

Previous attempts to modulate face-recognition performance with non-invasive brain stimulation over the right FFA have yielded mixed results (e.g., Brunyé et al., 2017; Wu et al., 2021; Kho et al., 2023). We asked whether high-definition tDCS (HD-tDCS) over right FFA changes face-recognition performance and the use of diagnostic information as a function of individual ability. Thirty adults were randomly assigned to anodal, cathodal, or sham HD-tDCS (N = 10 per group) targeting right FFA with a 5-electrode montage. In Session 1, participants completed the Cambridge Face Memory Test – Australian version (CFMT-Aus). In Session 2, they received offline HD-tDCS or sham and then performed the CFMT+ and 700 identity-recognition trials using the Bubbles technique (Gosselin & Schyns, 2001), in which faces are revealed through randomly positioned Gaussian apertures, allowing us to derive classification images indexing the use of visual information (in terms of facial features). Group-level ANOVAs on CFMT change scores (post–pre) showed no reliable difference between stimulation and sham conditions [F(2,27) = 0.62, p = 0.55]. However, regression analyses revealed a robust interaction between baseline CFMT and stimulation: in both anodal and cathodal groups, poorer recognizers improved after HD-tDCS, whereas better recognizers tended to worsen; no such pattern emerged under sham [F(1,18) = 13.00, p = 0.002, β = -0.65, r² = 0.42]. We then compared classification images across groups by matching participants on ability. Among low-ability observers, HD-tDCS over right FFA (cathodal or anodal) increased left-eye use relative to low-ability observers in the sham group, whereas among high-ability observers, HD-tDCS decreased left-eye use relative to high-ability sham participants, with no systematic change for the right eye. These findings provide evidence that FFA modulation reconfigures face-recognition performance and visual information use in a state-dependent manner, consistent with models where tDCS perturbs excitation–inhibition balance rather than exerting polarity-specific effects.

Acknowledgements: Natural Sciences and Engineering Research Council of Canada