Elsevier

NeuroImage

Volume 23, Issue 3, November 2004, Pages 905-913
NeuroImage

Distinct representations for facial identity and changeable aspects of faces in the human temporal lobe

https://doi.org/10.1016/j.neuroimage.2004.07.060Get rights and content

The neural system underlying face perception must represent the unchanging features of a face that specify identity, as well as the changeable aspects of a face that facilitate social communication. However, the way information about faces is represented in the brain remains controversial. In this study, we used fMR adaptation (the reduction in fMRI activity that follows the repeated presentation of identical images) to ask how different face- and object-selective regions of visual cortex contribute to specific aspects of face perception. We report that activity in the face-selective region of the fusiform gyrus (FG) was reduced following repeated presentations of the same face. Adaptation in this area was not sensitive to changes in image size, but was sensitive to changes in viewpoint. In contrast, face-selective regions in the superior temporal lobe failed to adapt to identical presentations of the same face, but showed an increased response when the same face was shown from different viewpoints and with different expressions. These results reveal a largely size-invariant neural representation in the inferior temporal lobe that could be involved in the recognition of facial identity, and a separate face-selective region in the superior temporal lobe that could be used to detect changeable aspects of faces. The absence of fMR-adaptation in object-selective regions of visual cortex challenges the idea that a more distributed network of areas is used to represent information about faces.

Introduction

Recognising complex objects, such as faces, is a simple and effortless process for most human observers. However, the apparent ease with which recognition takes place belies its inherent complexities and ambiguities. For example, as we move about or as gaze or expression change, the size and shape of a face image on the retina also changes. To be useful, the visual system must take into account these sources of variation to facilitate recognition, but at the same time be able to detect changeable aspects of faces that are important in social communication. Although models of face processing have proposed ways to deal with these different tasks, it remains unclear how these mechanisms might be implemented in visual cortex.

One model of human face processing proposes that information is processed in specialised modules (Breen et al., 2002, Bruce and Young, 1986, Haxby et al., 2000). This conception is supported by several physiological studies that show specific regions of the temporal lobe are more responsive to faces than to other complex objects (Allison et al., 1994, Kanwisher et al., 1997, Kreiman et al., 2000). These findings are consistent with brain lesion studies that report specific deficits in recognising, identifying and naming faces following damage to the inferior temporal lobe (Damasio et al., 1982, McNeil and Warrington, 1993). Despite this deficit in face perception, such individuals have a largely preserved ability to recognise other objects (McNeil and Warrington, 1993). In contrast, lesions to other areas of the temporal lobe can leave face recognition intact, but impair an individual's ability to identify other objects (Moscovitch et al., 1997).

An alternative model of face processing appeals to a more distributed representation across a large network of visual cortex. In this theory, the representation of a face is not restricted to those areas that respond maximally to this object category. This is because non-face, object-selective regions such as the lateral occipital complex (LOC) and the parahippocampal place area (PPA) also respond to the presentation of a face, albeit less than to non-face objects (Andrews and Schluppeck, 2004, Ishai et al., 1999). Indeed, a recent study has shown that our perception of faces could be based on a distributed pattern of response across the whole temporal lobe rather than on the activity of a few specialised modules (Haxby et al., 2001). Moreover, because of the spatial limitations of fMRI, it is possible that a weak response to a face in a particular brain region does not reflect a sub-optimal activation, but reveals the activation of a small proportion of face-selective neurons (Avidan et al., 2002).

In the present study, we have used the technique of fMR-adaptation (the decreased activity that occurs following repeated presentation of the same image) to determine how different aspects of face processing are represented in visual cortex (Grill-Spector and Malach, 2001). In two previous studies, adaptation to faces was reported in the lateral occipital complex (LOC) (Avidan et al., 2002, Grill-Spector et al., 1999). Because this region of the brain has been characterised as an object-selective area (Malach et al., 1995), these findings could be taken as support for a distributed representation underlying face perception. However, in these studies, the analysis was restricted to face-selective areas of the LOC. Indeed, it is possible that the regions studied may have included the face-selective region in the fusiform gyrus (Kanwisher et al., 1997). In the present study, we have defined face- and object-selective areas in the occipital and temporal lobe in terms of their anatomical location and functional responses and have asked how they are involved in specific aspects of face perception. Our hypothesis was that those regions of the brain that are involved in the recognition of identity would show a reduction in response to repeated presentations of the same face, and that this reduction in response would be invariant to changes in image size or viewpoint. In contrast, if an area was involved in representing changeable aspects of faces, we would not expect to find adaptation to repeated images of the same face identity, but rather we would expect responses to be sensitive to changes in viewpoint that are important in social communication.

Section snippets

Subjects

All eight observers had normal or corrected to normal visual acuity. Informed consent was obtained from all subjects and the study was approved by the Central Oxford Research Ethics Committee (COREC 98.161). Stimuli (approximately 9° × 9°) were back-projected (Focus LP1000, Unicol Engineering, Oxford UK) on to a screen placed at a distance of 280 cm from the subject's eyes. Subjects lay supine in the magnet bore and viewed the back-projection screen outside the bore through prism glasses.

Imaging parameters

All

Localiser scan

Spatially discrete face- and object-selective areas were localised using a blocked design (Fig. 1A and Table 1). In each subject, a region of the fusiform gyrus showed significant activation for faces versus non-face objects. This activation was predominantly in the right hemisphere. The Talairach coordinates of this area suggest that it is analogous to the FFA (Kanwisher et al., 1997) and area LO-a/pFs (Avidan et al., 2002, Grill-Spector et al., 1999). In addition, a more posterior region on

Discussion

The aim of this study was to determine how information about faces is represented in visual cortex. Specifically, we were interested in asking which regions of visual cortex are involved in forming an invariant representation of a face that could be used for recognition, and which areas process changeable aspects of faces that are important in social communication.

Consistent with previous studies, regions in the inferior and superior regions of the temporal lobe responded more to photographs of

Acknowledgments

We grateful to Caroline Johnson for her involvement in the early stages of this project. We thank Peter Hobden, Dave Flitney and Paul Matthews for their help during the study, and we are grateful to Tony Atkinson for providing helpful criticism of the manuscript. Functional imaging was carried out at the Oxford Centre for Functional Magnetic Resonance Imaging of the Brain (FMRIB). This work was supported by a grant from the Royal Society to TJA. MPE is supported by studentship from the

References (44)

  • M.J. Tarr et al.

    Mental rotation and orientation-dependence in shape-recognition

    Cogn. Psychol.

    (1989)
  • T. Allison et al.

    Human extrastriate visual cortex and the perception of faces, words, numbers and colors

    Cereb. Cortex

    (1994)
  • T.J. Andrews et al.

    Correlated size variations in human visual cortex, lateral geniculate nucleus, and optic tract

    J. Neurosci.

    (1997)
  • T.J. Andrews et al.

    Activity in the fusiform gyrus predicts perception when viewing Rubin's vase-face stimulus

    NeuroImage

    (2000)
  • I. Biederman

    Recognition-by-components—A theory of human image understanding

    Psychol. Rev.

    (1987)
  • N. Breen et al.

    Models of face recognition and delusional misidentification: a critical review

    Cogn. Neuropsychiatry

    (2002)
  • V. Bruce et al.

    Understanding face recognition

    Br. J. Psychol.

    (1986)
  • H.H. Bulthoff et al.

    Psychophysical support for a 2-dimensional view interpolation theory of object recognition

    Proc. Natl. Acad. Sci. U.S.A.

    (1992)
  • G.A. Calvert et al.

    Silent lipreading activates the auditory cortex

    Science

    (1997)
  • J. Capgras et al.

    L'illusion des sosies dan un delire systematise chronique

    Bull. Soc. Clin. Med. Ment.

    (1923)
  • A.R. Damasio et al.

    Prosopagnosia-anatomic basis and behavioral mechanisms

    Neurology

    (1982)
  • R. Epstein et al.

    A cortical representation of the local visual environment

    Nature

    (1998)
  • Cited by (300)

    • The cultural learning account of first impressions

      2022, Trends in Cognitive Sciences
    View all citing articles on Scopus
    View full text