The human face reveals people’s identity and a wealth of information (e.g., identity, gaze, emotion, gender, etc.) critical for social interactions. Face processing in humans is thought to rely on visual mechanisms that are specifically tailored for this category1,2. Despite decades of psychological and neuroscientific research on this topic, the implementation of face-specialized mechanisms in the human brain is still a matter of debate. Knowing what the brain processes is necessary to understand how exactly it operates (Garner, 1970). Our lab adopts this line of attack and explores the primary visual information that is used when processing faces.
Primary visual information refers to the information analyzed in the primary visual cortex (V1), namely: orientation and spatial frequency (SF). The selective encoding of orientation and SF in V1 is assumed to provide the primary building blocks of visual perception.
The primary visual bases subtending high-level face perception is still largely elusive mainly because of the implicit assumption that the primary properties encoded in V1 are fully aggregated and therefore inaccessible as the visual signal reaches higher-level stages responsible for the encoding of complex shapes, such as faces. However, our research and that of others demonstrates that it is largely driven by selective ranges of primary visual information, therefore casting a new light on the nature of the specialization of human face perception, and more generally on how complex vision builds up in the human brain.