Justine Sergent’s research while affiliated with McGill University and other places

Little is understood about the brain, the mind and their relationships. However, rapid technical advances in brain-imaging devices such as positron emission tomography (PET), functional magnetic resonance imaging, EEG and EMG have increased the capabilities for visualizing the working brain, and uncovering the cerebral areas participating in the realization of cognitive tasks, and progress in cognitive science has led to a better understanding of the functional architecture of mental abilities. There is, therefore, considerable potential for achieving a greater understanding of the relationships between cognition and cerebral structures through brain-imaging studies of mental functions. However, these studies are confronted with a series of difficulties related to the assumptions that govern their application, the constraints imposed by these techniques on the design of cognitive experiments, the complexities inherent in establishing relations between cognition and anatomy through physiology, and to the interpretation of patterns of cerebral activation. In this article, potential difficulties are described drawing essentially on examples from PET studies of cognitive functions. Whereas a bright future lies ahead for the study of human brain mapping, many problems still have to be overcome and solved in order to exploit the full potential of new brain-imaging techniques.

The brain is organized into segregated areas of relative functional autonomy and specialization. This basic principle of cerebral organization is well documented for cognitive functions that differ drastically from one another, but less so for functions that belong to the same domain, such as face processing. Yet several sources of evidence point to a functional and structural dissociation of various aspects of face processing, as suggested by (1) an analysis of the perceptual and cognitive demands made by the processing of diverse properties conveyed by facial configurations, (2) selective impairment of aspects of face processing in brain-damaged patients, and (3) different localizations of face cells responsive to properties conveyed by faces such as identity and emotion in the monkey's brain. This study used positron emission tomography (PET) and magnetic resonance imaging (MRI) to delineate better the neurofunctional organization of face processing in the human brain, by measuring cerebral blood flow while subjects performed tasks involving the recognition of faces or the recognition of emotions expressed by faces. The results showed segregated processing of facial identity and facial emotion, with the former being performed predominantly in the ventro-mesial region of the right hemisphere including the limbic system, whereas the latter was carried out predominantly in the latter part of the right hemisphere and the dorsal region of the limbic system. This structural organization allows the parallel processing of different information contained in physiognomies and underlies the high efficiency with which humans process faces.

This chapter presents findings that may help to understand the anatomical and functional architecture underlying the processing of faces and their dissociation from the processing of objects. The recognition of faces requires the involvement of the ventral areas of the right hemisphere and; as presented by the PET and the radiological data from the patients, the right hemisphere seems to be both necessary and sufficient to sustain face recognition. Three cortical areas appear to be essential to this function, and they subserve specific operations. The right lingual and fusiform gyri perform the perceptual operations by which the physiognomic invariants are extracted and the uniqueness of the face can be assessed. The right parahippocampal gyrus, but not the hippocampus itself, seems to play a crucial role in the reactivation of pertinent memories associated with a given facial representation. The anterior temporal lobes of both hemispheres seem to contain the biographical information that has to be reactivated for a given face to become meaningful and, thus, be identified.

Understanding the functional organization of the cerebral structures underlying receptive and expressive musical processes is confronted with a wide variety of difficulties inherent in the artistic and subjective nature of the musical experience. Yet clarifying the relationships between music and the brain is a legitimate goal of neuroscientific research. One approach toward this goal is based on new developments of brain imaging techniques, and recent investigations indicate that the realization of musical abilities such as sight-reading and piano performance relies on a distributed neural network comprising locally specialized cortical areas. Another approach is concerned with the study of musicians, like Maurice Ravel, who have been affected by brain damage. An analysis of their deficits helps to uncover some properties of music-brain relationships, to identify the essential questions raised by these deficits, and to clarify the neurofunctional anatomy of musical abilities. The understanding of the neurocognitive bases of musical functions is still at an early stage, but recent progress in cognitive and neurofunctional research opens the way to more systematic studies than had so far been possible.

The understanding of the relationships between music and the brain is a legitimate goal of neuroscientific research. In spite of an already large body of experimental investigation, however, the outcome has not been as satisfactory as expected, to the point that some have voiced their incredulity at the feasibility of mapping musical functions onto cerebral structures. There are indeed considerable inconsistencies in the literature bearing on the neurobiological substrates of musical functions. Before giving in to skepticism, though, it is appropriate to examine whether such inconsistencies are unavoidable and a necessary outcome of neuroscientific investigation into musical functions. This paper, therefore, examines some of the reasons that may be responsible for the inconsistencies. Beyond the superficial similarities between language and music, these two domains are functionally distinct, yet neuropsychological research in music and its neurobiological substrates has been modeled after that in verbal language, borrowing concepts and methods that were not entirely suited to the study of musical functions. An examination of the diverse factors inherent in any study of music-brain relationships reveals a variety of potential sources of difficulties, which, if properly controlled, would guarantee higher consistency and enhance the reliability of experimental findings. It is thus suggested that progress in understanding music-brain relationships may result from (1) explicitly outlining the structural cognitive architecture of musical functions; (2) formulating the problems in terms of underlying operations rather than in terms of vague and unspecified dichotomic views of hemisphere processing; (3) designing experiments that are musically valid and coherent, and that do not lend themselves to multiple strategies; (4) setting up appropriate controls to ensure that subjects are performing as expected; (5) restricting the investigation to musically literate individuals; (6) selecting, among such individuals, a homogeneous set of subjects sharing common musical abilities; and (7) using investigatory techniques that yield reliable evidence of local cerebral involvement in realizing the tasks under study. © 1993 Wiley-Liss, Inc.

A commissurotomized subject, L.B., was shown asterisks flashed at random locations, up to four in each field, and attempted either to compare the numbers in the two fields or to report the total number. The main results were: (a) Report was more accurate with unilateral than with bilateral presentation, suggesting that the diffuculty integrating across fields was partly attentional; (b) in integrating across fields, attention was focused on one field, with only crude ‘one-or-many’ information from the other; (c) in cross-field comparisons, the focus was on the LVF, but in reporting the number it was on the RVF when report was oral or right-handed, and on the LVF when report was left-handed; (d) cross-field comparisons were improved when the locations were mirrored across the midline.

Music, like other forms of expression, requires specific skills for its production, and the organization and representation of these skills in the human brain are not well understood. With the use of positron emission tomography and magnetic resonance imaging, the functional neuroanatomy of musical sight-reading and keyboard performance was studied in ten professional pianists. Reading musical notations and translating these notations into movement patterns on a keyboard resulted in activation of cortical areas distinct from, but adjacent to, those underlying similar verbal operations. These findings help explain why brain damage in musicians may or may not affect both verbal and musical functions depending on the size and location of the damaged area.

Prosopagnosia is a neurologically based deficit characterized by the inability to recognize faces of known individuals in the absence of severe intellectual, perceptual, and memory impairments. The nature of the underlying disturbance was investigated in three patients in an attempt to identify the structural and functional levels at which the processing of faces breaks down, the relation between prosopagnosia and associated deficits, and the specificity of the prosopagnosic disturbance. The breakdown of face processing resulted from unilateral damage in different cerebral structures of the right hemisphere in the three patients, and it involved different functional levels of face processing, but all three patients displayed perceptual impairments of unequal severity. In one patient (R.M.), the deficit encompassed all perceptual operations on faces, including matching identical views of the same faces, but it did not extend to all categories of objects characterized by a close similarity among their instances; the second patient (P.M.) exhibited a less severe perceptual impairment but was unable to derive the configurational properties from a facial representation and to extract its physiognomic invariants; the third patient (P.C.) had not lost the capacity to differentiate faces on the basis of their configurations but could not associate a facial representation with its pertinent memories. Associated deficits were present in each patient but differed depending on the anatomofunctional locus of the breakdown, although all patients were impaired at recognizing noncanonical views of objects that they readily recognized when shown from a conventional viewpoint. However, performance dissociation within patients and double dissociation between patients suggest that these associated deficits are not necessary concomitants of prosopagnosia.

Prosopagnosia is an acquired neurological impairment characterized by an inability to experience a feeling of familiarity at the view of faces of known individuals and to identify these individuals. The inability of prosopagnosic patients to recognize faces does not entail that perceived faces go unprocessed in their brains, and there are indications that cognitive operations are still performed whereby a perceived face reactivates pertinent memories, but either these operations cannot be completed or their outcome fails to reach consciousness. A series of experiments were conducted on three severe prosopagnosic patients in an attempt to understand better this phenomenon known as covert face recognition, the conditions for its occurrence, and its functional locus. The capacity of the patients implicitly to access pertinent knowledge related to overtly unrecognized faces was inversely related to the severity of their perceptual deficit, suggesting that some preserved ability to extract the physiognomic invariants of a face is a necessary condition for the occurrence of the phenomenon. However, the results indicated that covert recognition does not take place at a perceptual or structural level although it is initiated at such a level, that it may have more than one underlying mechanism, and that it is achieved through the reactivation of specific information related to the individual's identity. Under special conditions that restricted the relevant knowledge that needed to be activated, transient overt recognition of faces was experienced by one of the prosopagnosic patients.

The study of face-selective neurons in the monkey temporal lobe, and face recognition deficits in humans after brain damage have both become very active fields of investigation. Face-selective neurons appear to be members of ensembles for coding faces rather than individual face detectors or grandmother cells. They reflect the more general role of temporal cortex in pattern recognition. In humans there are a variety of face-processing impairments that result from damage to different areas, and which reflect interference at different levels of processing of the facial image.