INAUGURAL LECTURE

WHAT NEUROLOGY REVEALS OF HUMAN NATURE LESSONS FROM PHANTOM LIMBS, CAPGRAS SYNDROME, AUTISM, AND ANOSOGNOSIA

V. S. Ramachandran, M.D., Ph.D., Psychology & Neuroscience, UCSD

Neurological syndromes that result from stroke, such as phantom limbs, Capgras delusions, anosognosia, etc. are often regarded as curiosities in clinical practice. We show that very important lessons can be learned from investigating them systematically in the laboratory. The relevance of these studies to stroke rehabilitation will be considered.  After arm amputation, touching a patients face will evoke sensations from the phantom fingers, evidence that massive reorganization of sensory maps has occurred in the adult braincontrary to dogma. Second, phantoms can initially be moved voluntarily by the patient but after a few months he/she loses the ability (a phenomenon we have dubbed "learned paralysis"). But by providing visual feedback with a mirror it is possible to "unlearn " this learned paralysis. We now have preliminary evidence from our lab that even the hemiparesis following stroke might have a "learned" componenta sort of temporary "block" or inhibition that can be partially overcome with a mirror in at least some patients. Double blind trials are underway (Altschuler & Ramachandran, in preparation). 

We will then discuss Capgras syndrome and autism and suggest that these two disorders are caused by distortions in connections between IT and amygdala causing a distorted "salience landscape" followed by strange, unpredictable responses to sensory inputs. In autism, for example, Bill Hirstein and I find that the autonomic responses (SCR) are chaotic and unrelated to perceptual and social salience.

Finally we discuss visual neglect and anosognosia (the "denial" syndrome) and describe experiments we have conducted on these patients.  Thus even the most enigmatic and bizarre syndromes become at least partially comprehensible in light of these experiments and the results might provide key insights into human nature. We conclude with some speculative remarks on the evolutionary origins and neural basis of art and aesthetics.

MAPPING HUMAN BRAIN FUNCTION IN HEALTH AND DISEASE LESSONS LEARNED FROM PATIENTS AND NORMAL SUBJECTS

John C. Mazziotta, M.D., Ph.D., Neurology, UCLA School of Medicine

Brain Mapping techniques in the clinical arena serve a number of important functions. Using multiple imaging modalities, it is possible to characterize abnormalities in patients as well as in the normal brain, in which the composite characterization of the information gathered is greater than the sum of its parts. We utilize structural and functional magnetic resonance imaging (MRI), positron emission tomography (PET), optical intrinsic signal (OIS) imaging, and behavioral responses during "awake" surgery" to identify regions of the brain that are normal versus pathologic and, as such, appropriate for surgical resection. This is particularly important in patients with lesions (e.g., brain tumors, sites in the brain that produce epilepsy, vascular abnormalities) that are situated close to or within critical motor, visual or language areas of the brain. In such patients, preoperative imaging using PET or MRI studies is validated with intraoperative evaluations involving OIS and "awake" stimulation of the exposed brain. Such studies have allowed for surgery in patients that would otherwise have been denied because of the proximity of their abnormalities to critical motor, visual or language areas. In addition, such studies make more routine surgeries safer and more efficient. It is hoped that by validating the preoperative PET and functional MRI studies, the current intraoperative battery of tests, that awakening patients from surgery will, in the future, no longer be necessary thereby avoiding the trauma and prolonged operative time required for such diagnostic testing in the operating room. Evidence already points to the fact that this may be, in fact, the case. Examples of each of these types of studies will be demonstrated during the presentation as well as examples of individual patients who have benefited from these procedures.

AN APPROACH TO THE PHYSIOLOGY OF CONSCIOUSNESS

Joseph E. Bogen, M.D., Neurosurgery, University of South California

An anatomico-physiologic approach to consciousness is facilitated by recognizing that the various meanings of consciousness have in common a crucial core C characterized by subjectivity including the conviction of volition. A sharp distinction is made between the property C and the contents of consciousness, partial loss of which is typical of cerebro-cortical lesions. The neuronal mechanism producing subjectivity also acts as an attention-action coordinator, and hence must have specific connectivity requirements. These requirements are best met by the thalamic intralaminar nuclei (ILN). Whereas large lesions elsewhere leave C undisturbed, quite small bilateral lesions in the ILN engender immediate unresponsiveness. This combination of anatomic and neurologic evidence is bolstered by a variety of physiologic evidence leading to the conclusion that further investigations of the ILN, and their interaction with lower centers as well as cerebral cortex, are most apt to yield a better understanding of consciousness.  C of nausea, thirst, fatigue and the like are at least as typical of being conscious as C of time, place, person, meaning, memories and expectations.  Excessive emphasis on cognition by current students of consciousness may derive, as do so many attitudes, from Descartes who claimed I think, therefore I am. A neurobiological understanding of consciousness might come more quickly were we to aver I feel, therefore I am.

CORTICOFUGAL INFLUENCES ON BRAIN PLASTICITY

T. P. Pons, Ph.D., Neurosurgery, Wake Forest University

School of Medicine & E.R. Ergenzinger, Ph.D.

After denervation of the hand and arm by dorsal rhizotomies in adult monkeys, the face representation in somatosensory cortex expands into the silenced representation of the hand. Several mechanisms at cortical and subcortical levels have been put forth to explain how such massive reorganization occurs. We originally suggested that reorganization at the thalamic level might contribute to the cortical changes, but this suggestion has received little attention. Recently, we were able to study monkeys with limbs that had been deafferented for 12 to 20 years and electrophysiologically record from the thalamus and anatomically study the brainstem and thalamus (Jones and Pons '98). In these monkeys we determined that both the brainstem cuneate, and the thalamic ventral posterior lateral nucleus (VPL), had undergone severe transneuronal atrophy. Interestingly, electrophysiological recording in the thalamus demonstrated that the face representation had expanded so as to be adjacent to the representation of the trunk.  In particular the normally relatively small representation of the lateral face had expanded in a manner comparable to the expansion in cortex that we previously demonstrated (Pons et al., '91). Ongoing transneuronal atrophy in the VPL of the thalamus was evident, even 20 years after the dorsal rhizotomies.  Such anatomical changes appear to be a progressive process and are likely to be related to the electrophysiological changes we observed in the thalamus

We have recently demonstrated that top-down influences from cortex can exert a substantial influence over plastic changes in the thalamus (Ergenzinger et al., '98). Reducing neuronal activity in somatosensory cortex results in a substantial enlargement of receptive field (RF) size in VPL of primates. We determined that suppression of cortical activity, when paired with severing peripheral nerves to the hand, enhances reorganization in VPL but not in the brainstem cuneate nucleus. This process occurs acutely, and suggests different roles for the cortico-thalamic and cortico-cuneate projection. Furthermore, our findings indicate that cortex plays a major role in regulating the magnitude of subcortical plastic changes which occur following peripheral injury. Such acute changes include the emergence of RFs in VPL that are responsive to stimulation of the back of the hand and the lower face. This finding indicates that pre-existing inputs from the face and hand onto individual neurons in VPL can be unmasked under certain conditions. The presence of these connections may provide the substrate through which immediate phantom sensations emerge following deafferentation, and along which greater expansion and reorganization can occur over time.

THE ZOMBIE IN THE BRAIN:  VISUAL ROUTES TO KNOWLEDGE AND ACTION

Melvyn A. Goodale, Ph.D., Psychology, University of Western Ontario

Sight is our pre-eminent sense and makes the most important contribution to our conscious experience of the world of objects and events beyond our bodies. We do not simply respond to visual stimuli; we experience them as integral components of a visual world that has depth, substance, and most important of all, an existence separate from ourselves. But even though the perceptual representation of objects and events in the world is an important function of vision, it should not be forgotten that vision evolved, not to provide perception of the world per se, but to provide distal sensory control of the many different movements that organisms make. Work with monkeys and with neurological patients is making it increasingly clear that the control of skilled actions depends on visual processes that are quite independent from those that lead to perception-based knowledge of the world. Indeed, the distinction between vision for perception and vision for action is reflected in the organization of the visual pathways in primate brain. In the cerebral cortex, for example, two broad "streams" of projections from primary visual cortex have been identified a ventral stream projecting eventually to the inferotemporal cortex and a dorsal stream projecting to the posterior parietal cortex. Both streams process information about the structure of objects and about their spatial locations - and both are subject to the modulatory influences of attention. Each stream, however, uses this visual information in different ways. The ventral stream transforms the visual information into perceptual representations that embody the enduring characteristics of objects and their relations. Such representations play an essential role in the identification of objects and enable us to classify objects and events, attach meaning and significance to them, and establish their causal relations. These operations are essential for accumulating knowledge about the world. In contrast, the transformations carried out by the dorsal stream deal with the moment-to-moment information about the location and disposition of objects with respect to the effector being used and thereby mediate the visual control of skilled actions, such as manual prehension, directed at those objects.

Additional evidence for this duplex account of visual processing comes from studies with normal observers. For example, the visual control of a number of different actions has been shown to be quite insensitive to pictorial illusions that can have profound effects on conscious perceptual judgements. These and other demonstrations suggest that the visual control of many of our movements, from grasping objects to walking through cluttered environments, uses metrics that are quite different from those employed by visual perception.  In summary, the dorsal action system and ventral perception system play complementary roles in the production of adaptive behavior. Thus, even though the execution of a goal-directed action might depend on dedicated on-line control systems in the dorsal stream, the selection of appropriate goal objects and the action to be performed depends on the perceptual machinery of the ventral stream. What remains a central and outstanding issue is how the two streams communicate in the production of adaptive behavior.

TWO APPROACHES TO CONSCIOUSNESS

Paul Churchland, Ph.D., Philosophy, UCSD

WHAT CAN WE EXPECT FROM A THEORY OF CONSCIOUSNESS?

Patricia Churchland, Ph.D., Philosophy, UCSD

IS INTEGER ARITHMETIC FUNDAMENTAL TO MENTAL PROCESSING?  THE MIND'S SECRET ARITHMETIC

Allan Snyder, Ph.D., Centre for the Mind, The Australian National University

Unlike the ability to acquire our native language, we struggle to learn multiplication and division. It may then come as a surprise that the mental machinery for performing lightning fast integer arithmetic calculations could be within us all even though it can not be readily accessed, nor do we have any idea of its primary function. We are led to this provocative hypothesis by analyzing the extraordinary skills of autistic savants. In our view such individuals have privileged access to lower levels of information not normally available through introspection.

THE SALIENCE LANDSCAPE THEORY AUTONOMIC DYSREGULATION AS A MAJOR CAUSE OF CHILDHOOD AUTISM

Bill Hirstein, Ph.D. & V. S. Ramachandran, M.D., Ph.D.,

Center for Brain and Cognition, UCSD

Input from cortical areas concerned with perception (e.g., IT) is gauged by the amygdala for perceptual salience, resulting in autonomic arousal as evidenced by SCR. Normal brains have a "perceptual salience landscape" that is determined by the strength of connections between IT and amygdala and modulated by connections to the cingulate and ventromedial frontal structures. We postulate that in autism these connections have become deranged (e.g., due to kindling, as occurs in temporal lobe epilepsy.) resulting in a randomly skewed salience landscape. We tested this theory in twenty autistic children, a majority of whom were "high functioning".  We observed severe autonomic dysregulation in the form of an inability of the sympathetic nervous system to establish an electrodermal baseline in 9 out of 11 autistic children. In general, the autonomic responses of the autistic children were chaotic and not sensitive to the significance of the stimulus, especially with social stimuli (Figure 1). In addition, these limbic-autonomic responses have consequences for higher-level cognition (Hirstein and Ramachandran, 1997, Proc Roy Soc, autonomic response produces 'mpostor' delusion). The function of this system in perception is to produce a 'salience landscape', in the sense in which the eyes of a face are the most salient parts for normal people. Failure of this system can in some autistic children lead to an inability to be guided by incoming perception. The system can alternate between extremely high, chaotic levels of arousal, in which every outside event is given annoyingly high significance, and low non-responsive levels, in which the child is relaxed, but unreachable. Bachevalier has argued that the central problem in the autistic brain is amygdala damage/malfunction (Bachevalier, 1994). The central route to the amygdala in visual perception occurs along the temporal lobe, in the 'ventral' visual stream.  The autistic brain may 'prefer' activity in the other, 'dorsal' stream (generated by tactile or somesthetic activity) to activity in the ventral stream and its concomitant sympathetic overexcitation. Repetitive, motor stereotypies, or 'stimming' behavior causes steep reductions in sympathetic activity, which may explain why autistic children seek to engage in such behavior they are themselves attempting to regulate a system which should be self-regulating. A therapy has been developed in which autistic children can self-regulate their autonomic nervous systems, by engaging in a tactile stimulating behavior while they learn. Perhaps this can "damp" the autonomic dysregulation.

EXPORTING VISION TO THE MIND

P. Cavanagh, Ph.D, Psychology, Harvard University

Attention is the gateway to visual awareness and it imposes the final limit on what we can consciously experience in our visual world. We have evaluated the temporal and spatial grain of selection for awareness and discovered that the smallest regions and briefest intervals which can be isolated by attention are surprisingly coarse. Objects spaced more finely than the limit of attentional resolution can not be individuated for further processing and can only be perceived as a grouped spatial or temporal texture. As a result, only part of the information registered by the early sensory systems is available to conscious perception. The properties of attentional resolution suggest that the locus of attentional selection is at a stage beyond primary visual cortex.

ATTENTION SERIAL AND PARALLEL PROCESSING IN THE MIND/BRAIN

Hal Pashler, Ph.D., Psychology, UCSD

The brain is capable of massive parallel processing at the level of individual processing units (i.e., neurons). At the level of mental operations, ordinary intuition suggests that we often carry out different tasks in parallel, so long as the tasks are not too cognitively demanding.  Laboratory research on "divided attention" provides a rather different picture of people's ability to carry out different tasks at the same time, however. Fine-grained measurements of behavior reveal that even seemingly trivial mental operations sometimes interfere profoundly with other mental operations. While parallel processing occurs with some mental operations, other operations seem to generate processing "bottlenecks" that require serial processing. I will describe recent research in our laboratory that begins to characterize the nature of these processing limitations and provides clues about their possible neural basis.

THE BIOCHEMICAL BASIS OF SCHIZOPHRENIA

John Smythies, M.D., Ph.D., UCSD

Schizophrenia is today the most important problem in neuropsychiatry. It affects 1% of the population and current treatment leaves much to be desired. This talk will review the present state of our knowledge about the biochemical factors relating to schizophrenia and current work in this Center on this topic. There are two different types of schizophrenia-type 1 and type 2. The risk factors of these will be stated together with an account of the research strategies used in this research. There are a number of different biochemical hypotheses, of which the most prominent are the transmethylation, the adrenochrome, the one-carbon cycle, the serotonin, the dopamine and (particularly) the glutamate. The background and current status of these will be reviewed. The most significant finding so far in the pathobiology of schizophrenia has been the fact that certain neurons in the cortex and striatum have lost some 50% of their dendritic spines, which form the main locus for their excitatory glutamate synapses.  These spines are normally highly dynamic structures and are continually being formed and destroyed. I have recently presented a new hypothesis-the Redox Hypothesis-of the basic biochemical mechanism involved in spine plasticity. This involves the balance between neurodestructive pro-oxidant molecules produced at the glutamate synapse (such as the superoxide anion and hydrogen peroxide) and neuroprotective antioxidants (such as vitamin C and dopamine). The oxidative pathway of dopamine via orthoquinones is also involved. The function of this system in the normal brain is reviewed as well as the evidence of its malfunction in schizophrenia, and other brain diseases such as Parkinson's disease and Alzheimer's disease.

WHY DO THINGS LOOK AS THEY DO?

Dr. Tom Albright, Ph.D., UCSD, Salk Institute

HOW THE BRAIN GETS ORGANIZED FOR LANGUAGE AND OTHER COMPLICATED THINGS

Elizabeth Bates, Ph.D., Cognitive Science, UCSD

Aphasia is the oldest research domain in cognitive neuroscience, since the first observations about the link between brain injury and language impairment approximately 5000 years ago. For much of the modern history of the field, theories of brain organization for language have vacillated between two extremes the phrenological view, with its emphasis on innate and compactly localized "mental organs" for language and other complicated things, and the tabula rasa view in which an equipotential brain is determined entirely by experience. I will present evidence from lesion studies of both children and adults showing that both of these views are wrong. Brain organization for language emerges gradually across the course of development, and experience plays a major role its specification; however, this organization is overlaid on richly articulated sensorimotor coordinates that existed long before language evolved, providing strong constraints on the class of possible, workable "brain plans" for language and other human cognitive skills.