Early temporal negativity is sensitive to perceived (rather than physi- cal) facial identity Stella J. Faerber a,n , Jürgen M. Kaufmann a,b , Stefan R. Schweinberger a,b a DFG Research Unit Person Perception, Friedrich Schiller University of Jena, Germany b Department of Psychology, Friedrich Schiller University of Jena, Germany article info Article history: Received 15 October 2014 Received in revised form 29 April 2015 Accepted 21 May 2015 Available online 23 May 2015 Keywords: Adaptation ERPs Face recognition Mental representation Priming abstract Increased early negativity to repeated faces over inferior temporal regions around 200–300 ms has been related to the reactivation of mental representations of individual identities of familiar faces. Since this modulation is larger for same-image (compared to different-image) repetitions of a familiar face, it is debated whether it reflects physical stimulus similarity between prime and target, or reactivation of perceived representations of identity. In Experiment 1 participants performed a four-choice identification task on famous target faces, which were always preceded by the same average face. This average face served as prime stimulus. Crucially, by adapting participants to specific anti-faces, we induced different illusory facial identities (cf. Leopold, D.A., O’Toole, A.J., Vetter, T., Blanz, V., 2001. Prototype-referenced shape encoding revealed by high-level after effects. Nature Neuroscience 4 (1), 89–94) in the same physical prime. Importantly, temporal ERPs ( 155–235 ms) were significantly more negative for “Primed” than for “Unprimed” trials. In Experiment 2, we determined whether this effect was due to the encoding of shape information, by using anti-shape adaptors with constant average reflectance in- formation. Priming by these anti-shape adapted average primes did not elicit a similar temporal ERP modulation. We conclude that these early temporal ERP modulations evoked by a combination of anti- face adaptation and immediate-repetition priming represent a neural correlate of the activation of mental representations of individual familiar faces. These identity specific representations can be trig- gered even by physically identical prime stimuli, when preceded by corresponding anti-face adaptation. & 2015 Elsevier Ltd. All rights reserved. 1. Introduction The cognitive neuroscience of face perception has been the subject of considerable research (e.g., Calder et al., 2011). Although progress has been made in understanding the mechanisms med- iating the recognition of familiar faces (Bruce and Young, 1986; Young and Bruce, 2011), the neuronal implementation of familiar face recognition – as contrasted with unfamiliar face processing – is subject to debate. For instance, Haxby et al.'s (2000) model, based on fMRI data, has become widely accepted. It suggests three “core” areas for face perception in the inferior occipital gyrus (IOG), the lateral fusiform gyrus (FG), and the superior temporal sulcus (STS). Nevertheless, each of these regions is similarly acti- vated by familiar and unfamiliar faces, such that neither region has been unequivocally linked to familiar face recognition (Natu and O’Toole, 2011). Rather, familiar compared to unfamiliar faces were reported to activate a widespread network of various brain areas in an “extended” face processing system (e.g., Leveroni et al., 2000; Nasr and Tootell, 2012), and the responses of those areas largely depend on personal (and not just visual) familiarity with a face. Neurophysiological studies using EEG have also revealed a number of face-sensitive responses. Not unlike the situation in fMRI research, the most prominent and easily identified of those responses – the occipito-temporal N170 – is typically found to be equivalent in amplitude and topography for familiar and un- familiar faces (Eimer, 2000; Schweinberger, 2011). By contrast, a slightly later ventral temporal negativity to repeated compared to unrepeated faces – the N250r (r for repeated) – has been con- sistently found to be larger for familiar than for unfamiliar faces (Doerr et al., 2011; Herzmann et al., 2004; Schweinberger et al., 1995). Source modelling of EEG and MEG data suggested that the N250r (and M250r, respectively) is likely generated in fusiform cortex (Schweinberger et al., 2004, 2007). Moreover, immediate repetitions of previously learned (but not novel) faces elicited a relatively view-independent N250r response (Zimmermann and Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/neuropsychologia Neuropsychologia http://dx.doi.org/10.1016/j.neuropsychologia.2015.05.023 0028-3932/& 2015 Elsevier Ltd. All rights reserved. n Correspondence to: Friedrich Schiller University of Jena, DFG Research Unit Person Perception (PPRU), General Psychology and Cognitive Neuroscience, Am Steiger 3/Haus 1, D-07743 Jena, Germany. Fax: þ49 3641 9 45182. E-mail address: stella.faerber@uni-jena.de (S.J. Faerber). Neuropsychologia 75 (2015) 132–142