Neural response to the visual familiarity of faces

Abstract

Recognizing personally familiar faces is the result of a spatially distributed process that involves visual perceptual areas and areas that play a role in other cognitive and social functions, such as the anterior paracingulate cortex, the precuneus and the amygdala [M.I. Gobbini, E. Leibenluft, N. Santiago, J.V. Haxby, Social and emotional attachment in the neural representation of faces, Neuroimage 22 (2004) 1628–1635; M.I. Gobbini, J.V. Haxby, Neural systems for recognition of familiar faces, Neuropsychologia, in press; E. Leibenluft, M.I. Gobbini, T. Harrison, J.V. Haxby, Mothers’ neural activation in response to pictures of their, and other, children, Biol. Psychiatry 56 (2004) 225–232]. In order to isolate the role of visual familiarity in face recognition, we used fMRI to measure the response to faces characterized by experimentally induced visual familiarity that carried no biographical information or emotional content. The fMRI results showed a stronger response in the precuneus to the visually familiar faces consistent with studies that implicate this region in the retrieval of information from long-term memory and imagery. Moreover, this finding supports the hypothesis of a key role for the precuneus in the acquisition of familiarity with faces [H. Kosaka, M. Omori, T. Iidaka, T. Murata, T. Shimoyama, T. Okada, N. Sadato, Y. Yonekura, Y. Wada, Neural substrates participating in acquisition of facial familiarity: an fMRI study, Neuroimage 20 (2003) 1734–1742]. By contrast, the visually familiar faces evoked a weaker response in the fusiform gyrus, which may reflect the development of a sparser encoding or a reduced attentional load when processing stimuli that are familiar. The visually familiar faces also evoked a weaker response in the amygdala, supporting the proposed role of this structure in mediating the guarded attitude when meeting someone new.

Introduction

In two previous fMRI studies on familiar face recognition [12], [23] we have suggested that recognizing someone we know is the result of a spatially distributed process that involves not only visual cortical areas but also areas that are involved in other social and cognitive functions such as the theory of mind areas (anterior paracingulate and posterior superior temporal sulcus) [10], [11], the precuneus, the amygdala and the insula. We have hypothesized that different areas encode different types of information that support the successful recognition of a familiar individual [13]. While the precuneus and anterior temporal cortex [14], [24], [26], [32] might be involved in retrieval of information from long-term memory, the theory of mind areas might encode information about personality traits and mental states of a familiar individual [12], [23]. The emotional response that we experience when seeing a familiar face plays an important role in successful recognition of that individual [1], [12], [23] and is reflected by changes in activity in the amygdala and the insula.

In our functional model on face perception [17], we grouped face responsive regions in two systems: the core system that includes areas involved with the visual analysis of a face and the extended system that includes areas that are involved in the extraction of other non-visual information.

We were interested in isolating the effect of simple visual familiarity in face recognition. With this purpose, neural activity was recorded with fMRI while participants viewed faces that were visually familiar, due to experimentally induced learning, with no associated semantic information.

Based on our previous findings, we hypothesized that visual familiarity would modulate activity in the “core system”, namely the fusiform gyrus. We also predicted modulation of activity in parts of the extended system, such as in the precuneus, an area involved with the retrieval of episodic memories [3] and in the amygdala, a structure involved in detecting and processing unexpected or unfamiliar events with potential biological importance [5], [7], [12], [22], [23], [31].

We did not expect any modulation of activity in areas involved with the retrieval of personal traits such as the anterior paracingulate cortex [12], [23], [25] or in areas associated with retrieval of biographical information such as the anterior temporal regions [14], [24], [26], [30].