Face recognition during naturalistic viewing

The ability to recognize familiar individuals is essential for guiding social interactions, yet the cognitive and neural mechanisms underlying familiar face processing remain incompletely understood. While prior research has emphasized the role of perceptual information in generating image-invariant visual representations for recognizing familiar faces, the influence of conceptual information—naturally associated with faces in real-world contexts—has received less attention. This thesis aims to investigate the neural correlates of familiar face perception under naturalistic conditions and to understand the role of conceptual information in the recognition process. A series of empirical studies employing natural viewing paradigms in combination with behavioural, computational and neural measures were conducted. First, fMRI was used to identify neural correlates of familiar face processing (Chapter 2). Results revealed differential activation patterns for familiar versus unfamiliar faces across a distributed brain network extending beyond the visual face regions. These findings were further examined in individuals with developmental prosopagnosia and in a case study of hyperfamiliarity for faces, highlighting the critical role of non-visual brain regions in face recognition (Chapter 3). Building on this, Chapters 4 and 5 explored the contribution of conceptual information to face familiarization during natural viewing. Findings demonstrated that faces are more accurately recognized when learned alongside contextual conceptual information, which also modulates neural responses in non-visual brain regions. In Chapter 6, individuals with developmental prosopagnosia exhibited reduced conceptual understanding during face processing, reinforcing the relationship between conceptual information and familiar face recognition. Collectively, this thesis underscores the importance of non-visual brain regions in familiar face processing and establishes conceptual information as a pivotal component of face familiarization and recognition. These findings challenge traditional models that attribute familiar face recognition solely to image-invariant visual representations within visual brain regions. By integrating perceptual and conceptual processes, this work offers novel insights into theoretical models of face recognition and advances our understanding of the interplay between visual and conceptual information in social cognition.