Joaquin M. Fuster - US grants

The long-term objective of this research is to gain better understanding of the neural foundation of perception and memory. It is designed to provide new data concerning the processes by which the brain of the primate encodes, retains, and utilizes spatial information in behavior. The rationale is based on suggestive evidence that the posterior parietal cortex, in man and monkey, is the cortical substrate for spatial representation, a kind of dynamic neural map of extrapersonal space. The work will be conducted on macaques (Macaca mulatta). Some of the animals will be trained to distinguish objects by touch and to remember their shape for later recognition. The posterior parietal cortex will be cooled with implanted probes to test the effects of the reversible functional depression of that cortex on tactile (haptic) and crossmodal (haptic/visual) recognition performance. Nerve-cell activity will be investigated in the posterior parietal cortex while the monkey is engaged in those forms of performance; the principal aim is to determine the organization of cortical neurons representing stereognostic information. Other animals will be trained to distinguish and remember colored stimuli indicating the position of reward in the immediate space and future. The functional interactions between parietal and prefrontal cortex will be explored in these animals with a combination of cryogenic and microelectrode recording methods. The effects of cooling one cortical region on the neuronal activity in the other region will be analyzed, as will be the effects of cortical cooling on behavioral performance; the aim is to clarify the role of the normal interactions between the two cortices, presumably mediated by direct connections, in the spatial organization of behavior. Computer methds will be used for analyzing the effects of cryogenic depression as well as the patterns of neuronal discharge in task performance. The results of these experiments may help to elucidate the pathogenesis of disorders of the cerebral cortex in the human and to develop remedial measures for such disorders.

This action is to allow the investigator, Dr. Joaquin Fuster, to purchase computer equipment to support an innovative focus of his research. The overall objective of his research is to gain understanding of the neural foundation of perception and memory. It is designed to provide new data concerning the processes by which the brain of the primate encodes, retains, and utilizes information acquired by active touch. The principal and most innovative aspects of this research is the in-depth computational analysis of temporal patterns of neuron discharge during haptic discrimination and short-term memory. This work is important because it builds on previous models of memory and pushes it to the cutting edge by the addition of computational methodology.

9308905 Fuster When one wants to remember a piece of information like a telephone number for 20 seconds or so, one uses a memory process called active or working memory. This working memory resembles a mental scratchpad. When the information has been used to dial the phone, the scratchpad is erased and the information is gone. If the number was dialed incorrectly, it has to be looked up again because no trace remains. Thus this memory process differs from that used in long term memory where the trace remains permanently. Neurons in the cerebral cortex fire actively during the interval that the information is "kept in mind." With this NSF grant, Dr. Fuster a pioneer in the study of biological memory, will use multielectrode arrays to record from active memory cells and correlate the activity of neighboring cells when memories are being kept in mind and when they are erased. He will compare these correlations to those simulated by an artificial neural network, in order to elucidate the cortical architecture and connectivity that is necessary to provide such a memory capability. The results will improve our understanding of biological memory and our ability to design artificial circuits that remember and adapt.***

The overall objective is to gain better understanding of the memory systems and functions of the brain. This project is designed to clarify the mechanisms of temporary active memory in neuronal networks of the cerebral cortex. The question is important because active memory is essential for effective control of sequential behaviors and higher cognitive functions which are commonly deranged in certain mental disorders, notably schizophrenia. The proposed research is basically computational and has two closely intertwined components: (a) the analysis of electrical discharge of cortical nerve cells in memory; and (b) the development of computer simulation models of cortical memory. The role of recurrent activation of neurons in memory will be investigated by analysis of the patterns of spike discharge in single units and in multiple simultaneously recorded units. The role of prefrontal cortex in memory will be examined by studying the effects of cooling that cortex on parietal cells and on tactile memory performance. Resulting empirical models will be used as the source of empirically testable hypotheses of cortical network dynamics in active memory.

The long-term objective of this research is to gain better understanding of the neural mechanisms of perception and memory. It is designed to provide new knowledge concerning the processes by which the brain of the primate encodes, retains, and utilizes sensory information in behavior. The rationale is based on suggestive evidence that the posterior parietal cortex, in man and monkey, is the neural substrate for representation of information acquired by active touch (haptically). The work will be conducted on macaques (Macaca mulatta) trained to distinguish objects by touch and to remember their features for later recognition. Nerve-cell activity will be recorded from prefrontal and parietal cortex while the monkey is engaged in haptic performance. The functional interactions between parietal and prefrontal cortex will be explored with a combination of cryogenic and microelectrode recording methods in the behavioral setting. An important and innovative aspect of this research is the in-depth computational analysis of temporal patterns of neuron discharge during haptic discrimination and shorterm memory. Present knowledge of cortical organization raises the possibility that the firing patterns of cortical nerve cells reflect the patterns of reentry of impulses from other parts of the representational networks to which they belong. Thus, the effects of cooling one cortical region will be analyzed on the neuronal discharge patterns of another, as will be the effects of cortical cooling on behavioral performance. The aim is to clarify the functional interactions between components of cortical networks in sensory perception and short-term memory. The results of these experiments may help us elucidate the pathogenesis of disorders of the cerebral cortex in the human and to develop remedial measures for them.

Active short-term memory plays a critical role in the temporal organization of behavior, reasoning, and language. It is impaired in a variety of mental disorders, notably in the psychoses, major affective disorders and syndromes resulting from pathological aging of the brain. Recent experimental evidence indicates that active short-term memory consists in the sustained activation of an extensive network of interconnected neuronal assemblies of the cerebral cortex. Memory networks are widely distributed, extending beyond the boundaries of anatomically defined cortical areas. The mechanisms of active memory are believed to include the sustained circulation of neuronal impulses within one such network. This research will attempt to substantiate the distributed nature of active short-term memory and the reentry of neural impulses presumed to be the basis of short-term memory retention. Experiments with that objective will be conducted in nonhuman primates trained to perform auditory, tactile, and cross-modal memory tasks. Fields potentials and cell discharge will be recorded from frontal and parietal regions of the cortex during performance-and also to some extent during learning-of those tasks. Frequency and pattern of cell discharge will be analyzed for evidence of neuronal interactions within and between those cortical regions during the short-term memorization of sensory information. Special computational methods of time-series analysis will be used to explore those interactions. Some of the interactions will be further explored by reversible functional depression (by local cooling) of frontal cortex and the study of its effects on parietal cell discharge and the animal's performance of haptic memory tasks. The results of these studies are expected to shed light on the functional architecture of cortical memory networks and on the mechanisms of encoding, retention, and retrieval of memory. A better understanding of the neural dynamics of memory may help us better understand the pathogenesis of memory disorders in the mentally ill and thus lead to better treatments.

Part of the brain called the prefrontal cortex plays an important role in active memory of sensory experience, but short-term memory apparently involves cortical memory networks that are more widely distributed. The general objective of this study is to clarify how the cerebral cortex retains information in the short term for goal-directed behavior. This project uses two methods of recording cortical activity in monkeys performing visual short-term memory tasks: the recording of surface field potentials and the recording of optical intrinsic signals. This combination will allow the study of neural network activity within and across cortical areas with high spatial and temporal resolution. Correlation of location and time course will be examined between electrical and optical measurements during task performance. Two specific hypotheses will be tested: (a) neuronal activation develops over prefrontal and parietal cortex during retention of spatial and nonspatial memoranda; and (b) in parietal cortex, that activation is more prominent during retention of spatial memoranda than during nonspatial memoranda. Because of the phenomenological relation between active memory and consciousness, an understanding of the dynamics of active memory may help elucidate the role of the cerebral cortex in conscious states. Analysis of electrical and optical signals should reveal underlying topographic specificity in the retention of memoranda within the cortical areas explored. In addition, by contrasting the optical data against those from an established electrical method, this study should help substantiate the biophysical basis of optical imaging and its use in cognitive neuroscience.
Results will have an impact beyond cortical physiology by leading to a better understanding of memory and consciousness, and will be important to technological developments in optical imaging.

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. This project has two major objectives. The first is to identify structural and functional properties of cognitive neuronal networks in cortex of association (prefrontal and posterior parietal) during working memory. The second objective is largely methodological: to substantiate the coupling between neural activity and hemodynamic changes in working memory. Both objectives will be pursued in the monkey by the combined use of four minimally invasive and behavior-compatible recording methods: near-infrared spectroscopy (NIRS), surface field-potential (FP) recording, local field-potential (LFP) recording, and unit-activity recording. NIRS signals and surface FPs will be recorded simultaneously with epidural probes. Unit activity and local field potentials (LFPs) will also be recorded simultaneously by means of transdural microelectrodes. Based on certain assumptions of cognitive network architecture and the spatial resolution of each method, the four methods will be used in combination to test three specific hypotheses of neural activation and hemodynamic change in the regions of interest during the performance of two working-memory tasks, spatial delayed response (DR) and non-spatial delayed matching to sample (DMS). The analysis will focus on the neural and hemodynamic activity during the retention of a sensory stimulus in working memory. NIRS, FP, and unit data will be correlated with each of the variables most relevant to the specific hypotheses to be tested: cortical location, stimulus or memorandum, task, time of trial, and level of correct performance. In the study of neural-hemodynamic coupling in cognitive function, special emphasis will be placed on the correlations between NIRS signals and electrical manifestations of cell discharge. These correlations are expected to provide crucial information on the neuronal basis of functional imaging signals, such as those obtained by BOLD fMRI, in human cognition.