Brain Advance Access originally published online on July 22, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Brain, Vol. 126, No. 11, 2381-2395,
November 2003
© 2003 Guarantors of Brain
doi: 10.1093/brain/awg241
A network of occipito-temporal face-sensitive areas besides the right middle fusiform gyrus is necessary for normal face processing
1 Faculté de Psychologie et des Sciences de l‘Education, 2 Département de Radiologie, Hopital Universitaire de Genève and 3 Unité de Neuropsychologie, Hopital Universitaire de Genève, Geneva, Switzerland, and 4 Unité de Neurosciences Cognitive et Laboratoire de Neurophysiologie, Université Catholique de Louvain, Louvain-la-Neuve, Belgium
Correspondence to: Bruno Rossion, Unité de Neurosciences Cognitives (NESC), Université Catholique de Louvain, 10 Place du Cardinal Mercier, 1348 Louvain-la-Neuve, Belgium E-mail: bruno.rossion{at}psp.ucl.ac.be
Received November 11, 2002. Revised February 26, 2003. Second revision May 1, 2003. Accepted May 19, 2003.
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Neuroimaging studies have identified at least two bilateral areas of the visual extrastriate cortex that respond more to pictures of faces than objects in normal human subjects in the middle fusiform gyrus [the ‘fusiform face area’ (FFA)] and, more posteriorly, in the inferior occipital cortex [‘occipital face area’ (OFA)], with a right hemisphere dominance. However, it is not yet clear how these regions interact which each other and whether they are all necessary for normal face perception. It has been proposed that the right hemisphere FFA acts as an isolated (‘modular’) processing system for faces or that this region receives its face-sensitive inputs from the OFA in a feedforward hierarchical model of face processing. To test these proposals, we report a detailed neuropsychological investigation combined with a neuroimaging study of a patient presenting a deficit restricted to face perception, consecutive to bilateral occipito-temporal lesions. Due to the asymmetry of the lesions, the left middle fusiform gyrus and the right inferior occipital cortex were damaged but the right middle fusiform gyrus was structurally intact. Using functional MRI, we disclosed a normal activation of the right FFA in response to faces in the patient despite the absence of any feedforward inputs from the right OFA, located in a damaged area of cortex. Together, these findings show that the integrity of the right OFA is necessary for normal face perception and suggest that the face-sensitive responses observed at this level in normal subjects may arise from feedback connections from the right FFA. In agreement with the current literature on the anatomical basis of prosopagnosia, it is suggested that the FFA and OFA in the right hemisphere and their re-entrant integration are necessary for normal face processing.
Keywords: prosopagnosia; face recognition; visual cortex; fusiform gyrus; FFA; feedback
Abbreviations: BORB= Birmingham Object Recognition Battery; FFA = fusiform face area; fMRI = functional MRI; OFA = occipital face area; rFFA = right fusiform face area; rOFA = right occipital face area
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Understanding how the human brain perceives and recognizes faces is one of the most exciting and debated areas of research in cognitive neuroscience (e.g. Haxby et al., 2000
; Tarr and Gauthier, 2000). Pursuing this general goal, neuroimaging studies (mostly PET and functional MRI) have identified a number of relatively small functional areas lying outside the retinotopic cortex that respond more to pictures of faces than other objects in the normal human brain. The most robust difference in activity between faces and objects has been described in the lateral middle fusiform gyrus, bilaterally, but often stronger in the right hemisphere (e.g. Halgren et al., 1999; Haxby et al., 1999; Kanwisher et al., 1997; McCarthy et al., 1997; Rossion et al., 2000). This is the region that has been named the ‘fusiform face area’ (FFA) and defined as a module for face perception (Kanwisher et al., 1997). Posterior to the FFA, a region of the inferior occipital gyrus (termed OFA for occipital face area, Gauthier et al., 2000) also responds more to faces than objects, with a larger differential response in the right hemisphere as well (Sergent et al., 1992; Halgren et al., 1999; Haxby et al., 1999; Gauthier et al., 2000; Rossion et al., 2000). A smaller number of studies have also disclosed larger activation for faces than objects in the posterior part of the superior temporal sulcus, but these observations are much less consistent (e.g. Puce et al., 1998; Haxby et al., 1999). Although their relative functions in face processing remain unclear, these regions are considered to form a common neural network for face perception, providing outputs to further areas and processes related for instance to facial expression processing, or retrieval of semantic knowledge about persons (see Haxby et al., 2000).
Besides the precise function of these regions, there are several important questions regarding the neuro-functional basis of face processing which cannot be addressed by neuroimaging alone. First, neuroimaging cannot indicate whether these functional brain regions are necessary for the successful perception and recognition of faces. It may well be that some of these regions are activated in response to the presentation of faces in normal humans and yet do not play a crucial role in any important aspects of face recognition. As a matter of fact, selective lesions of the regions where the largest proportion of face-selective cells (
20%, see Gross, 1992) are usually found in the monkey brain (namely in the superior temporal sulcus) do not appear to prevent monkeys from recognizing faces (Heywood and Cowey, 1992). A second, related, question that is difficult to resolve by neuroimaging experiments alone is the nature of the interactions between the different functional brain areas involved in face processing. For instance, in a recent neuro-functional model of face processing, Haxby and colleagues (see also Halgren et al., 1999) proposed that the most anterior face-sensitive regions of the visual cortex, such as the FFA, receive their input from the OFA, which would be involved in a basic analysis of facial features (Haxby et al., 2000). However, this hypothesis has not yet been verified by neuroimaging studies of normal subjects.
Neuroimaging studies of brain-damaged patients may help resolve these questions (Price and Friston, 2001). One of the most spectacular deficits observed following brain damage is prosopagnosia, a deficit classically defined as the inability to recognize faces of conspecifics, despite normal intellectual abilities and apparently normal recognition of other object categories. Although a few cases had been reported at the end of the 19th century, the term ‘prosopagnosia’ was coined by Bodamer in 1947 (Bodamer, 1947). Since then, despite its rarity, a certain number of cases with a major deficit of face recognition have been described (see Farah, 1990; for more recent cases, e.g. Sergent and Signoret, 1992; Clarke et al., 1997; Gauthier et al., 1999; Laeng and Caviness, 2001). Prosopagnosic patients are usually able to distinguish between a face and another object category (‘face detection’), but are unable to identify familiar faces—including famous faces, friends and relatives or even their own face—and are also unable to learn new faces (Damasio, 1985). Different types of prosopagnosia have been described in the literature (e.g. Damasio et al., 1990), but it is generally acknowledged that every patient is unique and most studies focus on the detailed analysis of single cases in order to clarify current theoretical debates regarding normal face processing (see Farah, 1990).
The lesions that cause prosopagnosia are usually found in ventral occipito-temporal cortex, involving the lingual and fusiform gyri, and are bilateral in most of the cases (Damasio et al., 1982; Sergent and Signoret, 1992) although right unilateral lesions can be the cause of the syndrome (e.g. Landis et al., 1988; Wada and Yamamoto, 2001; Uttner et al., 2002). However, establishing relationships between structures and function based on clinical studies alone has always been problematic; the lesions of prosopagnosic patients are not always clearly identifiable, can be highly variable from case to case, and are usually very large and not limited to the structures subtending face processing. Relative to the study of patients with naturally occurring brain lesions, neuroimaging thus affords far greater anatomical precision and avoids the confounding factors associated with patient studies, such as any compensatory functional reorganization of the brain (Sergent et al., 1992).
In the present paper, we combine a detailed neuropsychological testing with a neuroimaging investigation of a prosopagnosic patient (P.S.) in order to: (i) test whether a structurally and functionally normal FFA is sufficient for normal face processing or if other high-level visual areas besides the right FFA are necessary for normal face perception; and (ii) test the hypothesis that the FFA activation is dependent on face-sensitive feedforward inputs from the OFA.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Subjects
Patient P.S.
P.S. is a 52-year-old woman (born in 1950; right-handed) who sustained a closed head injury (hit at the back of the head by a bus) in 1992. CT scans first indicated contusions in the occipital and parieto-occipital regions, and the left cerebellum. A recent MRI scan revealed lesions to the lateral part of the occipital and temporal lobes, bilaterally, as well as in the anterior part of the left cerebellum. The right hemisphere lesion extends from the posterior part of the inferior occipital gyrus to the posterior fusiform gyrus. The left hemisphere region is more anterior and covers a large part of the fusiform gyrus, extending into the lower part of the temporal lobe (Fig. 1).
|
Following months of neuropsychological rehabilitation and spontaneous recovery, P.S. was normal at all the non-visual functions that were slightly impaired following the injury (motor aspects of language, calculation, imagery and memory), yet she was slow at all visual tasks requiring a fine-grained discrimination (September 1993: overlapping figures, dots counting tasks), presented a deficit in contrast sensitivity and a massive prosopagnosia. Helped by a neuropsychological rehabilitation programme, P.S. returned to her work as a kindergarten teacher, and she has to be trained regularly at a neuropsychological rehabilitation centre to use external cues to distinguish among children of her new class (see Mayer et al., 1999
). P.S. complains of a profound difficulty in recognizing faces, including those of her own family, and her own face. To determine a person’s identity, P.S. is reliant on cues such as haircut, hat, moustache or earrings, but also on the person’s voice, his/her posture and clothes. We presented P.S. with 60 faces of famous people, whose faces were well-known by her before her accident (selected on the basis of a questionnaire that she had to fill up months before testing), and 60 unfamiliar faces. Each face was presented for 3000 ms and P.S. had 5000 ms to press one of two keys according to the familiarity/unfamiliarity of the face. Whereas she correctly classified all the unfamiliar faces, P.S. pressed the key ‘familiar’ for 14 out of the 60 famous faces only. After the experiment, she was shown the famous faces she classified correctly (14) and asked to name them or give some information about them; she was correct only for four faces. P.S. does not complain of any difficulty at object recognition, but she reads more slowly than she used to.
The performance of P.S. on standard clinical tests of visual perception and recognition is given in Table 1. P.S.’s visual acuity is good, but not perfect and she is slightly slower than normal controls at detecting letters and numbers in her right visual field. She is also slower than normal controls at a simple reaction time task, detecting a cross in the centre of the screen, as typically observed in brain damaged patients (e.g. Benton, 1986). Her contrast sensitivity is in the normal range except at higher frequencies (>22 c/degree). She is perfect at all tests of low level visual processing as tested by the Birmingham Object Recognition Battery (BORB) (Riddoch and Humphreys, 1993; tests 1 to 6), including matching objects under different viewpoints (tests 7 and 8), and she has no difficulty in perceiving and identifying objects, including semantic associations and naming (Table 1 Object recognition). She was also tested in an object decision task on a computer (line drawing objects taken from Snodgrass and Vanderwart, 1980; non-objects taken from Kroll and Potter, 1984), which she performed perfectly. Her short-term and long-term visual memory are in the normal range (Table 1), as well as reading (slower than normal controls but perfectly accurate) and visual imagery (BORB, test 9). She was presented with the full set of colorized Snodgrass and Vanderwart’s objects (Rossion and Pourtois, 2001) to name, each object being presented for 3000 ms or until response on a computer screen. She was perfect at this task, always giving the most common object label or a close label also given by normal young subjects (see Rossion and Pourtois, 2001) without any hesitation.
|
These observations contrast with her face perception and recognition performance. She achieved a score of 27 out of 54 in the Benton face perception test, which rank her as highly impaired (<39). Her score at the Warrington face recognition test was 18 out of 25, being also significantly less accurate than controls.
Normal controls
In addition to P.S., six age- and sex-matched normal subjects performed all the behavioural tasks described in the sections below. Eleven normal controls (five females, mean age 29.8 years, all right-handed) were used in the neuroimaging study. Young normal controls of both genders were used in the neuroimaging study mainly for practical reasons, but there is no evidence of gender differences in face-sensitive responses, which also appear to remain stable until the ninth decade of age (Brodtmann et al., 2002). Both P.S. and the normal controls in the neuroimaging study gave informed consent to the experiments, which were approved by the Ethics Committee of the University of Geneva (Hopital Universitaire).
Materials and procedure
Behavioural tests
Computer tests of face processing. P.S.’s face processing abilities were tested extensively using a variety of computer tasks with well-face stimuli (see Fig. 2). Sex decision on faces was tested with 50 face stimuli (25 males) presented on the centre of the screen for 10 s maximum (next trial as soon as key press). P.S. and control subjects had to indicate the sex of the person by pressing one of two keys as accurately and as quickly as possible. Facial expression analysis was tested in two blocks. In the first block, three kinds of facial expression were presented: joy; fear; and anger (each face for 10 s maximum; next trial on key press), and subjects had to press one of three keys according to the kind of expression presented (second block: choice among disgust; sadness; and surprise). Age judgement was also a three-forced choice task: subjects were presented with 40 successive faces (10 s maximum) of different age and had to categorize them in three age classes: child (10 trials); adult (17 trials); and old person (13). Face matching, as indicated in Table 2, was tested using a same/different task with two faces presented simultaneously (a full front face and a three-quarters face) for 10 s maximum (two blocks of 35 trials).
|
|
Discrimination of non-face objects from visually homogenous categories. P.S. was presented with a matching task with a set of homogeneous objects: cars (Fig. 2). Seventy-two pairs of three-quarter profile cars were presented successively in the centre of the screen for 7000 ms each maximum or until the subject’s response (a blank screen of 1000 ms separated two trials). Subjects simply had to press a key if the two cars were the same (half of the trials) as accurately and quickly as possible. P.S. was also tested with a matching task using different sets of multi-parts novel objects (Fig. 2). In the first twoblocks, objects were presented simultaneously in the same viewpoint. For each block, there were 60 trials (7000 ms maximum, 1000 ms inter-stimulus interval).
Analysis. The Z-score is the ratio of the difference between P.S.’s score and the normal controls average score by the SD of the normals. A Z-score >2 means that P.S.’s performance is above or below 2SDs of the normals. A' is a nonparametric measure of sensitivity used in previous investigations of prosopagnosic patients (Gauthier et al., 1999). Chance performance yields a score of 0.5 and positive values indicate better than chance sensitivity.
MRI acquisition
Stimuli. Twenty full front photographs of unknown faces (http://faces.kyb.tuebingen.mpg.de/) and 20 pictures of common objects (object databank at http://www.cog.brown.edu/tarr/stimuli.html) were used. Stimuli were presented to the subjects via a video projector, a front-projection screen and a system of mirrors fastened to the head coil. The visual field spanned by this set-up was approximately 15° x 11°.
Scanning procedures. For anatomical scans, experiments were performed on a 1.5 T whole-body ECLIPSE system (Marconi Medical Systems, Cleveland, OH, USA) using the standard head coil configuration (body coil for excitation, head coil for detection). Acquired multi-slice volume was positioned on sagittal scout images. A first GRE (gradient-recalled-echo) T1-weighted sequence [TR (repetition time) = 162 ms; TE (echo time) = 4.47 ms; flip = 80°; FOV (field of view) = 250 mm; matrix = 256 x 256; slice thickness = 5 mm] was performed to acquire the same volume as in the functional session. Anatomical reference images consisted of a 3D GRE T1-weighted sequence (TE = 15 ms; FOV = 250 mm; matrix = 256 x 256; slice thickness = 1.25 mm).
For the fMRI localizer experiment, functional imaging consisted of an echo planar imaging (EPI) GRE sequence (TR = 3000 ms; TE = 40 ms; flip = 80°; FOV = 250 mm; matrix = 128 x 128; 30 contiguous 5 mm axial slices). Epochs of face and object presentations (18 s, 6 x TR) were counterbalanced and separated by baseline epochs (fixation cross, 9 s; 3 TR). Two runs were performed, with six epochs of each experimental condition (faces or objects) in each run. In each face/object epoch, 24 stimuli were presented for 750 ms, without any offset, but a small shift of position (20 pixels) between consecutive images. Twenty different images were used in each epoch, with four images repeated twice consecutively, and subjects were required to detect the repetitions (one-back task) by pressing a response key. In addition, patient P.S. was scanned twice, with the same paradigms, in two different sessions at a 5 month interval in order to ensure reproducibility of our findings.
Data analysis. Collected data was processed with cross-correlation analysis (Bandettini et al., 1993) after motion correction (Woods et al., 1992) using MEDX software (Sensor Systems, Sterling, VA, USA) (Gold et al., 1998). Data were smoothed spatially by convolution with a Gaussian [full width half maximum (FWHM) = 6 x 6 x 6] kernel. The cross-correlation, expressed in terms of Z-values (Le Bihan et al., 1993) was calculated pixelwise between a delayed box-car function and the set of measurements, without temporal auto-correlation correction (Friston et al., 1995).
Individual maps were then normalized to the Talairach space (Talairach and Tournoux, 1988). Afterwards, the statistical distribution of the Z-values for each subject was calculated and a probability value for each Z-value was attributed. Clusters size of >0.08 cm3 (>10 voxels; voxel size after normalization was 2 x 2 x 2 mm3) showing statistically significant Z-score (typically Z-threshold = 3.0, at P < 0.002, uncorrected) in the faces–objects comparison were considered (Forman et al., 1995). In addition, clusters with signal intensity <50% of the mean intensity of the functional images were ruled out, in particular in the vicinity of P.S. lesions.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Neuropsychological investigations
Computer tests of face processing
The results of the face processing tests in P.S. and normal controls are given in Table 2. P.S. was impaired at a sex decision task, both in accuracy and sensitivity (percentage: Z = 7.94, P < 0.001; A': Z = 8.30; P < 0.001) and response times (Z = 7.77, P < 0.001). Although she could discriminate faces of different expressions, P.S. was not as good as normal subjects at these tasks (first block: percentage: Z = 21, P < 0.001; reaction times: Z = 9.66, P < 0.001; second block: percentage: Z = 0.79, NS; reaction times: Z = 3.96, P < 0.001). As for judging the age of faces, P.S. had a normal performance at this task (accuracy: 90%; Z = 1.282, P = 0.10), but was significantly slower than control subjects (Z = 8.84, P < 0.001). Finally, P.S. was strongly impaired at a face matching task, both in accuracy and sensitivity (percentage: Z = 7.74, P < 0.001; A': Z = 11; P < 0.001) and she was also much longer than control subjects (reaction times: Z = 4.23, P < 0.001).
Discrimination of non-face objects from visually homogenous categories
As indicated in Table 3, P.S.’s performance was good, although slightly lower than the age-matched controls (sensitivity: Z = 2.23; P < 0.05) due to her higher rate of false alarms, but she was not significantly slower then controls (Z = 1.67, not significant). In the first two blocks, objects were presented simultaneously in the same viewpoint. For set A, P.S.’s sensitivity was very good; she made only three mistakes but she was slower than the controls (Z = 4.19, P < 0.001). A similar pattern was found for set B, which was more difficult overall, objects being more similar to each other. P.S.’s sensitivity was in the normal range (see Table 3), but she was also slower than the controls (Z = 4.48, P < 0.001). Finally, we ran a block of trials with the different and same trials presenting objects under different viewpoints. Again, P.S.’s performance was within the normal range, but she was slightly slower than normal controls (Z = 2.45, P < 0.01) and hits (Z = 2.19, P < 0.05).
|
Neuroimaging study
A significant activation for the comparison between faces and objects was found in the right middle fusiform gyrus for all the normal subjects (P < 0.002 uncorrected, 10 voxels minimum, see Table 4 and Figs 3 and 5). This is the region that has been termed the FFA (Kanwisher et al., 1997
) and will be referred to in this paper as the right fusiform face area (rFFA). The mean coordinates of activation for this region (Table 4) in the Talairach coordinates correspond to the coordinates found in previous neuroimaging studies (e.g. Kanwisher et al., 1997; Gauthier et al., 2000; Rossion et al., 2000). Compared with the Talairach-transformed individual brain of our prosopagnosic patient P.S., this average activation corresponds to an area of cortex completely spared by the lesions (Figs 1 and 6). Most importantly, when P.S. was tested in the same experiment, the rFFA presents a significant differential response between faces and objects just like normal subjects (Figs 1, 4 and 6; Table 5). Furthermore, the extent and height of the rFFA activation in P.S. do not differ from normal controls (see Figs 3–6; Table 5). Structurally and functionally, her rFFA thus appears to be perfectly normal, and these observations were replicated after an interval of several months (Tables 5 and 6, Fig. 4). These observations were made despite the fact that, as expected, P.S. performed quite well in the one-back object discrimination task (37 out of 48 correctly responded <750 ms; mean reaction times for correct trials: 499 ms), but her performance was very low for the face stimuli (12 out of 48; 526 ms).
|
|
|
|
|
|
In most normal subjects (n = 9 out of 11), there was also a significantly larger activation for faces compared with objects in the right inferior occipital cortex [Table 4; Fig. 5; ‘right occipital face area’ (rOFA)]. The coordinates of the maxima of activation in the group analysis correspond to a region that has been damaged by the lesion in P.S.’s brain (Figs 1 and 6). Furthermore, the individual Talairach coordinates for the maximum of activation in the rOFA of each subject also correspond spatially to a structurally damaged area of cortex in P.S.’s brain (Fig. 6).
|
In the left hemisphere, a number of subjects (n = 8 out of 11; see Table 4) showed a significant increase in the left middle fusiform gyrus for faces (–objects). Strikingly, these areas of activation were all localized in a region that is completely damaged in P.S.’s brain (Fig. 6). Few (five out of 11) subjects also showed a significant differential response to faces in an area of the left inferior occipital cortex (Fig. 6). For P.S., there was a small but significant extent of activation in the left inferior occipital cortex only in the first session, which was posterior to the damaged area. However, the mean signal intensity of this activation was reduced by 60% compared with the mean intensity of the functional images, such that this activation in the vicinity of the lesion may reflect an artefactual combination of partial volume and head motion effects.
Finally, as described previously (e.g. Epstein and Kanwisher, 1998), there was a significant increase of activation for the opposite contrast (objects–faces) in a more medial and anterior region of the temporal lobe, in the parahippocampal gyrus, for both P.S. (P < 0.002; size >10 voxels; see Table 6; Fig. 3) and the normal controls (11 out of 11 subjects at the same threshold).
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
A selective deficit at recognizing faces
On the basis of the patient’s complaints and her results at classical object recognition tests (Table 1), P.S.’s case can arguably be defined as one of the most selective deficits for face recognition that has been described in the neuropsychological literature. She is perfect at object recognition and naming. In this respect, she represents a clearer case of prosopagnosia than most previously well-described cases of acquired prosopagnosia, who presented evident deficits and complaints at object recognition, such as patients L.H. (e.g. Levine and Calvanio, 1989
; Farah et al., 1995), S.M. and C.R. (Gauthier et al., 1999), or E.L.M. (Dixon et al., 1998) for instance. Although the distinction between apperceptive and associative (prosop)agnosia is certainly not clear-cut (see Farah, 1990), P.S. is definitely of the former type, as shown by her impairment at simultaneous face matching tasks.
P.S. is also slightly impaired at other face processing tasks such as sex categorization and facial expression analysis, whereas a number of previous cases of apperceptive prosospagnosia have been reported to be intact at these functions (e.g. Tranel et al., 1988). Evidence from functional imaging in humans (e.g. Sergent et al., 1992; Morris et al., 1998) and neurophysiological recordings in monkeys (e.g. Hasselmo et al., 1989) indicates that facial expression processing and face recognition depend on separate brain substrates, but that they probably share several primitive perceptual mechanisms that could be impaired in P.S.’s case due to the posterior localization of her lesions. The fact that there was no activation of her amygdala (faces compared with objects P > 0.05, uncorrected), a region that is involved in emotional aspects of face processing (Haxby et al., 2000) is inconclusive because there was not any activation in this region for any of the control subjects for the face–object comparison.
P.S. has no problem with object perception, matching, recognition and naming. She also performed at a normal level and she was not significantly slower than normal controls (one control was actually slower than P.S.) at a within category discrimination task of control objects, namely cars. When tested with novel objects sharing identical surface properties but differing slightly by their overall shape and the shape of their parts (not their organization), P.S.’s performance was again in the normal range. She was slightly slowed down overall, which should not be surprising given her general slowing down typical of brain-damaged subjects (e.g. Benton, 1986) and the fact that even normal control subjects were also particularly slow at this task (>2000 ms on average). Most importantly, however, increasing the visual homogeneity of the distractors or presenting the objects under different viewpoints caused a slight drop of performance and an increase of response times that was similar for P.S. and for normal subjects (Table 3). Other evidence reported elsewhere (Rossion et al., 2002) also indicates that she is not more sensitive than controls to manipulations of the level of categorization in object discrimination tasks as used previously in other patients (Gauthier et al., 1999).
How can P.S.’s object recognition performance be so remarkably preserved despite presenting such extensive lesions of the extrastriate visual cortex (Figs 1 and 6)? One possibility may be that non-face objects are recognized using a single well-localized system, different than that for faces and located in the medial ventral temporal cortex where the larger activation for objects was found to be normal for P.S. (see Figs 1 and 6). However, recent evidence suggests that the discrimination of object categories, including faces, can be carried out through a distinct pattern of response across a wide expanse of cortex in the occipital and ventral part of the temporal lobe (Ishai et al., 1999; Haxby et al., 2001). Because large but also small amplitude responses in the ventral temporal cortex carry information about the object appearance, only very large bilateral lesions of the ventral lateral and medial visual pathway would lead to impairments of basic level object recognition according to this view. The fact that P.S., despite extensive lesions of the ventral temporal cortex, is still able to easily perform such basic level discriminations, can be taken as supporting the view that object recognition (including faces) at the basic level is widely distributed (Ishai et al., 1999; Haxby et al., 2001): there is still enough information carried out in the visual cortex to perform basic level object recognition efficiently. Yet, contrary to most objects, faces pose a special challenge to the visual system because they belong to a highly visually homogeneous category (Damasio et al., 1982) and need to be recognized at the individual level for efficient social interactions (Tarr and Gauthier, 2000; Tanaka, 2001). The development of a visual expertize at recognizing faces at the individual level appears to require the fast extraction of configural relationships (Maurer et al., 2002), a mechanism that may extend to the discrimination of non-face categories from visually homogeneous categories following expertize training (Gauthier and Tarr, 1997) and seems to be impaired in P.S.’s case (Rossion et al., 2002).
The necessary integrity of the right hemisphere face-sensitive areas in face processing
The lesions of prosopagnosic patients usually concern the ventral part of the occipital and temporal cortex, bilaterally (e.g. Damasio et al., 1982; Sergent and Signoret, 1992; Clarke et al., 1997), with an area of maximum overlap between lesions including the right posterior and middle fusiform gyrus (Barton et al., 2002). Three recent case descriptions of prosopagnosia with unilateral damage have also identified the right middle fusiform gyrus as a critical area for normal face perception. In one case, the lesion concerned both the right fusiform gyrus and the lateral occipital cortex (Wada and Yamamoto, 2001), whereas in two other cases the lesion also extends posteriorly but possibly spares the right lateral occipital cortex, at least structurally (Uttner et al., 2002). These observations are in agreement with the proposal that the right middle fusiform gyrus or rFFA is a dedicated area that is critical for face perception, or a ‘face module’ (Kanwisher et al., 1997; McCarthy et al., 1997). However, in the present study, we observed a prosopagnosic patient presenting a structurally intact right middle fusiform gyrus or rFFA and, most importantly, a significantly larger response to faces–objects at this level (Figs 1 and 6). This result suggests that this region was also intact functionally, at least as far as basic level face categorization is concerned. These neuroimaging findings are in agreement with a recent fMRI study that disclosed a rFFA activation in a prosopagnosic patient (S.M.; Marotta et al., 2001), although this patient also presented large deficits at recognizing common objects (Gauthier et al., 1999) and his lesions were mostly anterior (temporal lobe), sparing the entire occipital lobe and the fusiform gyrus bilaterally. Together, these findings contradict the view that the rFFA represents a module for face perception and supports a critical role of other high-level visual regions for normal face processing (Tovée, 1998; Haxby et al., 2000, 2001).
What is the localization of these high-level visual regions that appear to be critical for normal face processing? Although the region of the left FFA is damaged in P.S., it is unlikely that the areas of the left occipital and temporal cortex that respond more to faces in normal subjects are also critical. Indeed, several cases of prosopagnosia have been observed previously with unilateral right hemisphere lesions (Landis et al., 1988; Wada and Yamamoto, 2001; Uttler et al., 2002). The functional integrity of the left face-sensitive areas was unknown in these cases, although there was no evidence for a decrease of blood flow in the left hemisphere in the case described by Wada and Yamamoto (2001). There are also a number of cases with extensive lesions of the ventral occipital and temporal left hemisphere without prosopagnosia that have been reported in the literature (see Farah, 1990). Yet, it has also been shown that a lesion to the left hemisphere, even in right-handed humans, may be necessary at least in some cases (Ettlin et al., 1992) to produce a prosopagnosic deficit, or even sufficient (Mattson et al., 2000), thus pointing to the individual-specific degrees of hemispheric dominance for face processing.
If the necessary role of the left hemispheric regions remains unclear, what our study indicates, however, is that the rOFA is not only involved during face processing in normals, but that it is a necessary component for such normally functioning face processes to take place.
The face-sensitive response in the inferior occipital cortex (rOFA), within the lateral occipital (LO) complex involved in shape perception (Malach et al., 1995), is observed in many neuroimaging studies (e.g. Halgren et al., 1999; Haxby et al., 1999; Gauthier et al., 2000; Ishai et al., 2000; Rossion et al., 2000). Yet, despite the robust co-occurrence of this region with the FFA, authors have been quick to reject the possibility that the OFA presents a particular sensitivity to faces (Kanwisher et al., 1997), probably because of its relatively early localization in the ventral pathway, and its somewhat weaker face-sensitive response compared with the FFA (Gauthier et al., 2000). On the other hand, Haxby et al. (2000) included this OFA in their neuro-functional architecture of face processing, suggesting that the rFFA receives most of its input information for face processing directly from this region (see also Halgren et al., 1999). Although our data suggest rather a role of feedback connections in leading to the face-sensitive response of the OFA (see below), they definitely support a critical role of the rOFA for normal face processing, and thus a distributed processing model in the ventral visual pathway for this function rather than a sparse coding (Young and Yamane, 1992).
What is the neuro-functional relationship between the OFA and FFA?
In their neuro-functional model of face processing, Haxby et al. (2000) proposed that the rFFA receives most of its input information for face processing directly from the posterior rOFA, albeit suggesting the presence of feedback mechanisms between the two areas (see also Ishai et al., 2000). According to this view, the rOFA is involved in the extraction of individual facial features and project in turn to the rFFA, which would respond more to the whole face configuration. This hypothesis is consistent with a classical hierarchical and feedforward view of the visual system (Hubel and Wiesel, 1977), according to which the flow of visual information goes from more posterior visual areas to more anterior areas, with an increase of the receptive field and the complexity of the visual information represented. At first glance, this proposal is also consistent with the observation that the rFFA, but not the rOFA, is more activated when the subject’s attention is focused on the whole face versus single face features (Rossion et al., 2000). The less robust differential activation between faces and objects usually found at the level of the rOFA is also considered as supporting the view of a progressively increasing face-selectivity along the ventral visual pathway (Grill-Spector et al., 1999; Halgren et al., 1999).
However, the data reported here do not support this hypothesis of a main feedforward processing of information related to faces from the rOFA to the rFFA since there was a normal activation of the FFA in the right hemisphere of P.S., despite a structurally and functionally damaged rOFA. In addition, to our knowledge, there has not been any empirical evidence in normal subjects supporting such a feedforward and dependent processing of information from the OFA to the FFA, whose differential function in face processing remain largely unclear. Moreover, behavioural evidence clearly suggests that the processing of faces as integrated wholes has precedence over the processing of single face features (Tanaka and Farah, 1993)—an observation that does not fit with a feedforward hierarchical model of face processing that would go from processing features to whole faces. Electrophysiological recordings in humans also show that the peak latency of the N170 face-sensitive potential is delayed for isolated face features, compared with the response to whole face stimuli (e.g. Bentin et al., 1996). For these reasons and given the data reported here, we suggest that the larger response observed for faces compared with objects as early as in the inferior occipital cortex (OFA) rather arise from feedback connections from the rFFA—an hypothesis perfectly compatible with the concurrent and distributed processing of information (Felleman and Van Essen, 1991; De Yoe et al., 1994) as well as the presence of feedback (Lamme and Roelfsema, 2000; Bullier, 2001) and re-entrant phasic signalling (Edelman, 1993) in the visual cortex. Recent studies using cooling techniques in monkeys indicate that such feedback signals from higher order visual areas may contribute to the emergence of functional responses in early visual areas (Galuske et al., 2002; see also Bullier, 2001). If such mechanisms were in place within the ventral visual stream, the response to faces observed at the level of the inferior occipital cortex could arise as a consequence of higher-level face-sensitivity processes in the FFA, possibly to guide the fine-grained visual analysis of faces for individual discrimination (Gauthier et al., 2000) and old/new recognition (Rossion et al., 2003).
Although the findings reported here suggest strongly that the two right-sided face-sensitive regions and their possible re-entrant interactions are critical for normal within-category discrimination and recognition of human faces, the limited temporal resolution of fMRI does not allow us to clarify the nature of the dynamic interactions between these two functional areas at this stage. Future neuroimaging investigations of P.S. and other brain-damaged prosopagnosic patients may be directed towards this goal of understanding functional connectivity between visual areas (Friston, 2002). As far as this prosopagnosic patient is concerned, future neuroimaging studies will also have to clarify the exact localization of the lesions with respect to the retinotopic visual cortex and other extrastriate visual areas involved in object recognition in the ventral temporal and occipital cortices (Ishai et al., 1999; Haxby et al., 2001), as well as to test the sensitivity of the rFFA to within category discrimination and recognition of faces.
Conclusions
Combining a neuropsychological and neuroimaging investigation on a case of prosopagnosia, we showed that besides the right middle fusiform gyrus (FFA), the right inferior occipital cortex (OFA) is critical for normal face perception and that the face-sensitive responses observed at the level of the occipital cortex in normal subjects may arise from feedback connections from the right FFA rather than from a feedforward mode of processing. We suggest that a re-entrant mechanism between these two regions in the right hemisphere subtends within category discrimination and recognition of faces.
|
Acknowledgements |
|---|
We are immensely indebted to P.S. for her patience and motivation during the experiments carried out in this study. We also wish to thank Isabel Gauthier and Christine Schiltz for their comments on a previous version of this manuscript and Michael J. Tarr for his support and for making available the novel stimuli used to test P.S. behaviourally. B.R. is supported by the Belgian National Fund of Scientific Research. This study was supported by a grant ARC 01/06–267 (Communauté Française de Belgique – Actions de Recherche Concertées).
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Bandettini PA, Jesmanowicz A, Wong, EC, Hyde, JS. Processing strategies for time-course data sets in functional MRI of the human brain. Magn Reson Med 1993; 130: 161–73.
Barton JJS, Press DZ, Keenan JP, O‘Connor M. Lesions of the fusiform face area impair perception of facial configuration in prosopagnosia. Neurology 2002; 58: 71–8.
Bentin S, Allison T, Puce A, Perez A, McCarthy G. Electrophysiological studies of face perception in humans. J Cogn Neurosci 1996; 8: 551–65.[Web of Science]
Benton AL. Reaction time and brain disease: some reflections. Cortex 1986; 22: 129–40.[Web of Science][Medline]
Bodamer J. Die-Prosop-agnosie. Arch Psychiat Nerv 1947; 179: 6–54. English translation by Ellis HD, Florence M. Cogn Neuropsychol 1990; 7: 81–105.[CrossRef][Web of Science]
Brodtmann A, Puce A, Syngeniotis A, Darby D, Donnan GA. Hemodynamic response to faces remain stable until the ninth decade. Neuroimage Hum Brain Mapp 2002; 648.
Bullier J. Integrated model of visual processing. Brain Res Brain Res Rev 2001; 36: 96–107.[CrossRef][Medline]
Clarke S, Lindemann A, Maeder P, Borruat FX, Assal G. Face recognition and postero-inferior hemispheric lesions. Neuropsychologia 1997; 35: 1555–63.[CrossRef][Web of Science][Medline]
Damasio AR. Disorders of complex visual processing: agnosias, achromatopsia, Balint’s syndrome, and related difficulties in orientation and construction. In: Mesulam M-M, editor. Principles of behavioural neurology. Philadelphia: F.A. Davis; 1985. p. 259–88.
Damasio AR, Damasio H, Van Hoesen GW. Prosopagnosia: anatomic basis and behavioural mechanisms. Neurology 1982; 32: 331–41.
Damasio AR, Tranel D, Damasio H. Face agnosia and the neural substrates of memory. Annu Rev Neurosci 1990; 13: 89–109.[CrossRef][Web of Science][Medline]
De Yoe EA, Felleman DJ, Van Essen DC, McClendon E. Multiple processing streams in occipitotemporal visual cortex. Nature 1994; 371: 151–4.[CrossRef][Medline]
Dixon MJ, Bub DN, Arguin M. Semantic and visual determinants of face recognition in a prosopagnosic patient. J Cogn Neurosci 1998; 10: 362–76.[CrossRef][Web of Science][Medline]
Edelman G.M. Neural Darwinism: selection and reentrant signalling in higher brain function. Neuron 1993; 10: 115–25.[CrossRef][Web of Science][Medline]
Epstein R, Kanwisher N. A cortical representation of the local visual environment. Nature 1998; 392: 598–601.[CrossRef][Medline]
Ettlin TM, Beckson M, Benson DF, Langfitt JT, Amos EC, Pineda GS. Prosopagnosia: a bihemispheric disorder. Cortex 1992; 28: 129–34.[Web of Science][Medline]
Farah MJ. Visual Agnosia: disorders of object recognition and what they tell us about normal vision. Cambridge (MA): MIT Press; 1990.
Farah MJ, Levinson KL, Klein KL. Face perception and within-category discrimination in prosopagnosia. Neuropsychologia 1995; 33: 661–74.[CrossRef][Web of Science][Medline]
Felleman DJ, Van Essen DC. Distributed hierarchical processing in the primate cerebral cortex. Cereb Cortex 1991; 1: 1–47.[Medline]
Forman SD, Cohen JD, Fitzgerald M, Eddy WF, Mintun MA, Noll DC. Improved assessment of significant activation in functional magnetic resonance imaging (fMRI): use of a cluster-size threshold. Magn Reson Med 1995; 33: 636–47.[Web of Science][Medline]
Friston KJ. Beyond phrenology: what can neuroimaging tell us about distributed circuitry? Annu Rev Neurosci 2002; 25: 221–50.[CrossRef][Web of Science][Medline]
Friston KJ, Holmes AP, Poline JB, Grasby P, Williams SC, Frackowiak RS, et al. Analysis of fMRI time-series revisited. Neuroimage 1995; 2: 45–53.[CrossRef][Web of Science][Medline]
Galuske RAW, Schmidt KE, Goebel R, Lomber SG, Payne BR. The role of feedback in shaping neural representations in cat visual cortex. Proc Natl Acad Sci USA 2002; 99: 17083–8.
Gauthier I, Tarr MJ. Becoming a ‘Greeble’ expert: exploring mechanisms for face recognition. Vision Res 1997; 37: 1673–82.[CrossRef][Web of Science][Medline]
Gauthier I, Behrmann M, Tarr MJ. Can face recognition really be dissociated from object recognition? J Cogn Neurosci 1999; 11: 349–70.[CrossRef][Web of Science][Medline]
Gauthier I, Tarr MJ, Moylan J, Skudlarski P, Gore JC., Anderson AW. The fusiform ‘face area’ is part of a network that processes faces at the individual level. J Cogn Neurosci 2000; 12: 495–504.[CrossRef][Web of Science][Medline]
Gold S, Christian B, Arndt S, Zeien G, Cizadlo T, Johnson DL, et al. Functional MRI statistical software packages: a comparative analysis. Hum Brain Mapp 1998; 6: 73–84.[CrossRef][Web of Science][Medline]
Grill-Spector K, Malach R. fMR-adaptation: a tool for studying the functional properties of human cortical neurons. Acta Psychol (Amst) 2001; 107: 293–321.[CrossRef][Medline]
Gross CG. Representation of visual stimuli in inferior temporal cortex. Philos Trans R Soc Lond B Biol Sci 1992; 335: 3–10.
Halgren E, Dale AM, Sereno MI, Tootell RBH, Marinkovic K, Rosen B. Location of human face-selective cortex with respect to retinotopic areas. Hum Brain Mapp 1999; 7: 29–37.[CrossRef][Web of Science][Medline]
Hasselmo ME, Rolls ET, Baylis GC. The role of expression and identity in the face-selective responses of neurons in the temporal visual cortex of the monkey. Behav Brain Res 1989; 32: 203–18.[Web of Science][Medline]
Haxby JV, Ungerleider LG, Clark VP, Schouten JL, Hoffman EA, Martin A. The effect of face inversion on activity in human neural systems for face and object perception. Neuron 1999; 22: 189–99.[CrossRef][Web of Science][Medline]
Haxby JV, Hoffman EA, Gobbini MI. The distributed human neural system for face perception. Trends Cogn Sci 2000; 4: 223–33.[CrossRef][Web of Science][Medline]
Haxby JV, Gobbini MI, Furey ML, Ishai A, Schouten JL, Pietrini P. Distributed and overlapping representations of faces and objects in ventral temporal cortex. Science 2001; 293: 2425–30.
Heywood CA, Cowey A. The role of the ‘face-cell’ area in the discrimination and recognition of faces by monkeys. Philos Trans R Soc Lond B Biol Sci 1992; 335: 31–8.
Hubel D, Wiesel T. Functional architecture of macaque monkey visual cortex. Proc R Soc Lond B Biol Sci 1977; 198: 1–59.
Ishai A, Ungerleider LG, Martin A, Schouten JL, Haxby JV. Distributed representation of objects in the human ventral visual pathway. Proc Natl Acad Sci USA 1999; 96: 9379–84.
Ishai A, Ungerleider LG, Martin A, Haxby JV. The representation of objects in the human occipital and temporal cortex. J Cogn Neurosci 2000; 12 Suppl 2: 35–51.
Kanwisher N, McDermott J, Chun M. The fusiform face area: a module in human extrastriate cortex specialized for face perception. J Neurosci 1997; 17: 4302–4311.
Kroll JF, Potter MC. Recognizing words, pictures and concepts: a comparison of lexical, object and reality decisions. J Verb Learn Verb Behav 1984; 23: 39–66.[CrossRef]
Laeng B, Caviness, VS. Prosopagnosia as a deficit in encoding curved surface. J Cogn Neurosci 2001; 13: 556–76.[CrossRef][Web of Science][Medline]
Lamme VAF, Roelfsema PR. The distinct modes of vision offered by feedforward and recurrent processing. Trends Neurosci 2000; 23: 571–9.[CrossRef][Web of Science][Medline]
Landis T, Regard M, Bliestle A, Kleihues P. Prosopagnosia and agnosia for noncanonical views. An autopsied case. Brain 1988; 111: 1287–97.
Le Bihan D, Jezzard P, Allison T, Cuenode C, Ziffo T. Analysis of functional MR images with z-maps. J Magn Reson Med 1993; 3: 141–8.
Levine DN, Calvanio R. Prosopagnosia: a defect in visual configural processing. Brain Cogn 1989; 10: 149–70.[CrossRef][Web of Science][Medline]
Malach R, Reppas JB, Benson RR, Kwong KK, Jiang H, Kennedy WA, et al. Object-related activity revealed by functional magnetic resonance imaging in human occipital cortex. Proc Natl Acad Sci USA 1995: 92: 8135–9.
Marotta JJ, Genovese CR, Behrmann M. A functional MRI study of face recognition in patients with prosopagnosia. Neuroreport 2001; 12: 1581–7.[CrossRef][Web of Science][Medline]
Mattson AJ, Levin HS, Grafman J. A case of prosopagnosia following moderate closed head injury with left hemisphere focal lesion. Cortex 2000; 36: 125–37.[Web of Science][Medline]
Maurer D, Grand RL, Mondloch CJ. The many faces of configural processing. Trends Cogn Sci 2002; 6: 255–60.[CrossRef][Web of Science][Medline]
Mayer E, Fistarol P, Valenza N. Prise en charge neuropsychologique d’une patiente prosopagnosique. In Azouvi P, Perrier D, Van der Linden M (editors) La reeducation en neuropsychologie: etudes de cas. Marseille: Solal; 1999. p.
McCarthy G, Puce A, Gore JC, Allison T. Face-specific processing in the human fusiform gyrus. J Cogn Neurosci 1997; 9: 605–10.[Web of Science]
Morris JS, Friston KJ, Buchel C, Frith CD, Young AW, Calder AJ, et al. A neuromodulatory vote for the human amygdala in processing emotional facial expressions. Brain 1998; 121: 47–57.
Price CJ, Friston KJ. Functional imaging studies of neuropsychological patients: applications and limitations. Neurocase 2001; 8: 345–54.[CrossRef]
Puce A, Allison T, Bentin S, Gore JC, McCarthy G. Temporal cortex activation in humans viewing eye and mouth movements. J Neurosci 1998; 18: 2188–99.
Riddoch MJ, Humphreys GW. BORB: the Birmingham Object Recognition Battery. Hove (UK): Lawrence Erlbaum; 1993.
Rossion B, Pourtois G. Revisiting Snodgrass and Vanderwart’s object database: colour and texture improve object recognition [abstract]. J Vision 2001; 1: 413.
Rossion B, Dricot L, Devolder A, Bodart J-M, Crommelinck M, de Gelder B, et al. Hemispheric asymmetries for whole-based and parts-based face processing in the human fusiform gyrus. J Cogn Neurosci 2000; 12: 793–802.[CrossRef][Web of Science][Medline]
Rossion B, Schiltz C, Crommelinck M. The functionally defined right occipital and fusiform ‘face areas’ discriminate novel from long-term familiar faces. Neuroimage. In press 2003.
Rossion B, Gauthier I, Caldara R, Schuller AM, Tarr M, Mayer E. Are ‘face-selective areas’ necessary to acquire expertise with non-face objects? Evidence from a pure case of prosopagnosia with a bilateral lesion of the fusiform gyrus [abstract]. J Cogn Neurosci 2002; Suppl C104.
Sergent J, Signoret J-L. Varieties of functional deficits in prosopagnosia. Cereb Cortex 1992; 2: 375–88.
Sergent J, Ohta S, MacDonald B. Functional neuroanatomy of face and object processing. A positron emission tomography study. Brain 1992; 115: 15–36.
Snodgrass JG, Vanderwart M. A standardized set of 260 pictures: norms for name agreement, image agreement, familiarity, and visual complexity. J Exp Psychol Hum Learn 1980; 6: 174–215.
Talairach J, Tournoux P. Co-planar stereotaxic atlas of the human brain. Stuttgart: Thieme; 1988.
Tanaka JW. The entry point of face recognition: evidence for face expertise. J Exp Psychol Gen 2001; 130: 534–43.[CrossRef][Web of Science][Medline]
Tanaka J, Farah MJ. Parts and wholes in face recognition. Q J Exp Psychol A 1993; 46: 225–45.[Medline]
Tarr MJ, Gauthier I. FFA: a flexible fusiform area for subordinate-level visual processing automatized by expertise. Nat Neurosci 2000; 3: 764–769.[CrossRef][Web of Science][Medline]
Tovée MJ. Face processing: getting by with a little help from its friends. Curr Biol 1998; 8: R317–20.[CrossRef]
Tranel D, Damasio AR, Damasio H. Intact recognition of facial expression, gender, and age in patients with impaired recognition of face identity. Neurology 1988; 38: 690–6.
Uttner I, Bliem H, Danek A. Prosopagnosia after unilateral right cerebral infarction. J Neurol 2002; 249: 933–5.[CrossRef][Web of Science][Medline]
Wada Y, Yamamoto T. Selective impairment of facial recognition due to a haematoma restricted to the right fusiform and lateral occipital region. J Neurol Neurosurg Psychiatry 2001; 71: 254–7.
Woods RP, Cherry SR, Mazziotta JC. Rapid automated algorithm for aligning and reslicing PET images. J Comput Assist Tomogr 1992; 16: 620–33.[Web of Science][Medline]
Young MP and Yamane S. Sparse population coding of faces in the inferotemporal cortex. Science 1992; 256: 1327–31.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. V. Peelen, N. Lucas, E. Mayer, and P. Vuilleumier Emotional attention in acquired prosopagnosia Soc Cogn Affect Neurosci, April 28, 2009; (2009) nsp014v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Farivar, O. Blanke, and A. Chaudhuri Dorsal-Ventral Integration in the Recognition of Motion-Defined Unfamiliar Faces J. Neurosci., April 22, 2009; 29(16): 5336 - 5342. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Brancucci, G. Lucci, A. Mazzatenta, and L. Tommasi Asymmetries of the human social brain in the visual, auditory and chemical modalities Phil Trans R Soc B, April 12, 2009; 364(1519): 895 - 914. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. Bell, F. Hadj-Bouziane, J. B. Frihauf, R. B. H. Tootell, and L. G. Ungerleider Object Representations in the Temporal Cortex of Monkeys and Humans as Revealed by Functional Magnetic Resonance Imaging J Neurophysiol, February 1, 2009; 101(2): 688 - 700. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kouider, E. Eger, R. Dolan, and R. N. Henson Activity in Face-Responsive Brain Regions is Modulated by Invisible, Attended Faces: Evidence from Masked Priming Cereb Cortex, January 1, 2009; 19(1): 13 - 23. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Pitcher, L. Garrido, V. Walsh, and B. C. Duchaine Transcranial Magnetic Stimulation Disrupts the Perception and Embodiment of Facial Expressions J. Neurosci., September 3, 2008; 28(36): 8929 - 8933. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Fairhall and A. Ishai Effective Connectivity within the Distributed Cortical Network for Face Perception Cereb Cortex, October 1, 2007; 17(10): 2400 - 2406. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-C. Chen, K.-L. C. Kao, and C. W. Tyler Face Configuration Processing in the Human Brain: The Role of Symmetry Cereb Cortex, June 1, 2007; 17(6): 1423 - 1432. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Kanwisher and G. Yovel The fusiform face area: a cortical region specialized for the perception of faces Phil Trans R Soc B, December 29, 2006; 361(1476): 2109 - 2128. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. von Kriegstein, A. Kleinschmidt, and A.-L. Giraud Voice Recognition and Cross-Modal Responses to Familiar Speakers' Voices in Prosopagnosia Cereb Cortex, September 1, 2006; 16(9): 1314 - 1322. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Schiltz, B. Sorger, R. Caldara, F. Ahmed, E. Mayer, R. Goebel, and B. Rossion Impaired Face Discrimination in Acquired Prosopagnosia Is Associated with Abnormal Response to Individual Faces in the Right Middle Fusiform Gyrus Cereb Cortex, April 1, 2006; 16(4): 574 - 586. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. E. Bouvier and S. A. Engel Behavioral Deficits and Cortical Damage Loci in Cerebral Achromatopsia Cereb Cortex, February 1, 2006; 16(2): 183 - 191. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Winston, R.N.A. Henson, M. R. Fine-Goulden, and R. J. Dolan fMRI-Adaptation Reveals Dissociable Neural Representations of Identity and Expression in Face Perception J Neurophysiol, September 1, 2004; 92(3): 1830 - 1839. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||