John P. Donoghue - US grants

This proposal requests support (1) to add drug ejection techniques to the methods by which we examine the function of the cerebral neocortex and (2) to carry out preliminary studies with this methodology that begin to examine the functional role of cortical cholinergic input. We will combine drug ejection with neural recording to test the hypothesis that the widespread cholinergic input to cortex modulates the effectiveness of sensory inputs. First, we will confirm previous studies that have demonstrated a selective effect of acetylcholine on specific subsets of cortical neurons, and next we will identify changes in the response of neurons in the somatic sensorimotor cortex to peripheral stimulation when cholinergic agonists or antagonists are applied. The results will provide evidence relating to the ability of the cholinergic system to modify the way cortical neurons respond to specific sensory inputs and will be the basis for a major grant designed to examine the role of cortical cholinergic systems in memory, learning, and in memory disorders such as Alzheimer's disease.

The proposed experiments have two goals: First, to investigate the normal functional organization of motor cortex in adult primates and second, to explore the capacity of motor cortex representations to reorganize following peripheral nerve injury or subsequent to the acquisition of a skilled movement. MI representation pattern will be evaluated by combining intracortical microstimulation mapping with EMG recording techniques and by using extracellular and intracellular tracing techniques. Normal "static" MI organization pattern: Unlike previous studies that have described the movements evoked by MI stimulation, our studies will demonstrate the representation of functional groups of forelimb muscles in MI by recording EMG activity from each of 16 muscles that are activated from individual cortical sites. Muscle representation will be mapped at 100-200 MI sites using several different current intensities. These data will be used to determine the topological relationship, extent of overlap, and size of representation of each of the 16 muscles. Our preliminary data suggest that some muscles are multiply represented in spatially separate foci that overlap the representation of several other muscles. These foci may represent a pattern of functional grouping that has been described in MI. Intracellular recording/dye injection and anatomical pathway tracing techniques will be used after mapping to examine the connectional relationship between these individual or grouped representations in MI. Dynamic organization: A second goal of these studies is to determine the extent to which the relationship of the motor cortex with the muscles is changed (a) under normal conditions (i.e., is there a continual reshaping of MI representations?), (b) following nerve injury that prevents the normal use of the distal forelimb muscles, (c) following the acquisition of skilled digit movements. Microstimulation mapping will be used to study MI representation patterns at different intervals in normal animals, before and after anterior interosseus nerve section and before and after learning of a precision finger grip task. These experiments will provide important new information concerning the efferent topography of primate motor cortex as well as the flexibility of the relationship between MI cortex and muscles. The results of these studies may suggest new strategies for enhancing functional recovery and suppressing spasticity subsequent to injury of the peripheral and central nervous system.

The major goal of the proposed experiments is to test the general hypothesis that individual frontal motor areas function in motor learning as well as motor performance. The experiments will test three hypotheses that primary and non-primary frontal motor areas to learning. The first hypothesis is that there is a component of discharge modulation in motor cortex uniquely related to the motor learning process itself. The second hypothesis is that motor learning occurs in association with a functional reorganization of motor cortex that supports newly learned motor behaviors. The third hypothesis that individual frontal motor areas subserve different roles in motor learning. In the proposed experiments, neural activity, EMG, and movement kinematics will be recorded in monkeys during two types of visuomotor learning tasks that involve planar arm reaching movements. The first task will require the monkey to adapt either the extent or direction of the reaching movement when the compatibility between a visual position feedback signal and the amplitude or direction of arm movement is altered. Activity during adaptation will be compared to that obtained before and after adaptation and to trials that control for changes in kinematics and target location. The second task will investigate motor learning that requires the establishment of new spatiotemporal patterns of muscle activity when monkeys learn new sequences of reaching movements. In both tasks, recordings will be made using both single microelectrodes and chronically implanted microwires, which will permit the simultaneous and continued evaluation of multiple motor sites in identical behavioral conditions. Reorganization of the functional relationships of motor cortex neurons with each other, with their target muscles or with encoded kinematic variables will be examined using cross correlation techniques. Electrical stimulation methods will also be used to evaluate changes in cortical output. The proposed studies will demonstrate the potential for learning-related reorganization in each of three major subdivisions of motor cortex and the circumstances under which they change. In addition they will clarify the role of motor cortex in controlling the direction and extent of simple and sequential movements. These results have direct relevance to understanding the role of the motor cortex in the development of movement disorders, in functional recovery after damage and may assist in developing neurorehabilitation strategies. Finally, they may help to identify common cortical processes related to learning.

This project is being funded through the Learning & Intelligent Systems Initiative, and is supported in part by the NSF Office of Multidisciplinary Activities in the Directorate for Mathematical & Physical Sciences. Learning in many cognitive domains, including language and vision, involves recognition of complex hierarchical structure that is hidden or only indirectly reflected in the input data. In this project a multi-disciplinary group of applied mathematicians, cognitive scientists, computer scientists, linguists, and neuroscientists will study the learning of compositional structure in language, vision, and planning, and will also probe the neural mechanisms for identifying and exploiting such structure. The research involves three interacting lines of work. The first refines and extends statistical learning models; the second applies these models to language, vision, and planning; and the third develops and applies new experimental and analysis techniques for probing the neural mechanisms that learn and exploit compositional structure. The results of the project should significantly increase our understanding of complex learning, and should have implications for a wide range of topics in education (e.g., learning of complex knowledge structures in science and math) and technology (e.g., automated speech recognition, computer vision, robotics).

DESCRIPTION (provided by applicant): The long-term goals of my research are (1) to understand how cortical networks flexibly compute actions from sensory and cognitive signals and (2) to use this knowledge to develop brain computer interfaces (BCIs) that restore the ability to perform useful actions to humans with paralysis. The specific objective of this project is to determine how interconnected networks of neurons in the premotor and motor cortex compute reach and grasp actions from different types of signals. Dorsal premotor (PMd) and ventral premotor (PMv) areas of frontal cortex appear to represent separate processing streams that converge on primary motor cortex (MI) where reach and grasp are unified. This interacting circuit appears to generate commands for actions by dynamically linking various cues to specific arm and hand actions, but studies conflict on the nature of the signals processed, the types of sensorimotor transforms performed, or their roles in arm and hand movement. The present study introduces several new approaches to understand how the collective dynamics of these cortical networks provide a flexible substrate to link different signals and actions. Chronically implanted multi-electrode arrays implanted in PMv, PMd and MI and full upper limb motion capture will be used to simultaneously record neuronal spiking (192 channels) and measure the actions of the arm, wrist and fingers while monkeys perform tasks in different contexts. Context will be varied by training monkeys to perform four tasks that differ in the signals used for to guide arm actions: (1) using ongoing sensory feedback to capture a swinging object, (2) planning different grasp types from visual cues, (3) planning different goals for reach (either point or grasp), (3) making a perceptual judgments to determine grasp-type. Context-dependent changes will be measured in single cell responses and population representation (SA1) and in network functional connectivity (SA2) in PMv, PMd and MI simultaneously recorded for each task. Recording two or more contexts within single sessions will allow direct comparison of the same ensembles to determine how context modifies coding in single neurons, how population models generalize across contexts, and how networks combine various signals to achieve coordinated reach and grasp. We will address the hypothesis that PMv and PMd form separate information processing channels by comparing neural coding features and context sensitivity between areas. Finally, SA3 will test the requirement for specific sensorimotor computations by disrupting processing in these networks using emerging, highly-selective optogenetic methods or electrical stimulation. These studies will reveal mechanisms by which cortical circuits perform sensorimotor transformations and flexibly link groups of neurons to generate voluntary behavior. In addition, they will help to determine whether signals available in premotor areas, as well as MI, can provide useful sources for BCI command signals. Such signals could allow people with paralysis to regain complex functions like reach and grasp through robotic limbs that could operate usefully across a range of contexts. PUBLIC HEALTH RELEVANCE: This project examines how the brain computes movements from a range of sensory and cognitive signals. It specifically seeks to understand how interacting networks of neurons can flexibly link different types of information to commands used for voluntary reaching and grasping movements. Beyond revealing how the brain computes, the information from this project is also important to identify the best sources of brain signals for brain computer interfaces, which are technologies that can help those with paralysis regain independence and control.

DESCRIPTION (Applicant's Abstract): Funds are requested to help support a meeting that is designed to promote integrated research in brain science. The conference will include prominent researchers from various fields of brain research to discuss major issues in brain function from an interdisciplinary perspective. The meeting will cover (Session 1) Plasticity of Sensory Systems; (Session 2) Temporal Dynamics from the Synaptic to the Systems Level; (Session 3) Memory, and finally (Session 4) the Relationships between Neurons and Cognition. A novel aspect of the meeting is a sincere attempt to bridge issues of brain function from the fundamental cellular and molecular mechanisms to their operation at the whole-organ level. Facilitators in every session will be charged with pursuing the challenges to solving the fundamental issues raised by the speakers. It is hoped that the initiatives identified by this meeting will catalyze novel interdisciplinary studies that will accelerate our understanding of normal and abnormal brain function. In addition, we hope to promote the idea of using interdisciplinary approaches to the study of the most challenging aspects of brain function, both to the current research community and to those who will become leaders in the future. Finally, by inviting and supporting minority students to the conference, we hope to increase their representation in the field.

EIA- 0113679
Black, Michael
Brown University

ITR/SY: The Computer Science of Biologically-Embedded Systems

Biologically-embedded systems that directly couple artificial computational devices with neural systems are emerging as a new area of information technology research. The physical structure and adaptability of the human brain make these biologically-embedded systems quite different from computational systems typically studied in Computer Science.

Fundamentally, biologically-embedded systems must make inferences about the behavior of a biological system based on measurements of neural activity that are indirect, ambiguous, and uncertain. Moreover these systems must adapt to short- and long-term changes in neural activity of the brain. These problems are addressed by a multi-disciplinary team in the context of developing a robot arm that is controlled by simultaneous recordings from neurons in the motor cortex of awake behaving monkeys. The goal is to probabilistically model the behavior of these neurons as a function of arm motion and then reconstruct continuous arm trajectories based on the neural activity. To do so, the project will exploit mathematical and computational techniques from computer vision, image processing, and machine learning.

This work will enhance scientific knowledge about how to design and build new types of hybrid human/computer systems, will explore new devices to assist the severely disabled, will address the information technology questions raised by these biologically-embedded systems, and will contribute to the understanding of neural coding.

DESCRIPTION (provided by applicant): This BRP grant application describes a pioneering, integrated lightweight neuromotor prosthetic microsystem (NMP) for paralyzed humans: NMPs use neural activity as a direct output to machines that run assistive devices. To meet this goal we have assembled an interdisciplinary team that combines leaders in neuroscience (Donoghue) and engineering (Nurmikko), with support from computer science (Black) at Brown University and neurology/neurosurgery at Brown and at Massachusetts General Hospital (Hochberg);experts in NMP design, development, manufacturing and commercialization from Cybernetics, Inc. (CKI), a neurotechnology company;and experts in neural prosthesis development and human application at the Cleveland FES center/Case Western University (H. Peckham/R. Kirsch). We will develop an integrated and implantable microelectronic neurosensor system that features on-chip signal processing and wide bandwidth transcutaneous wireless transmission capabilities. The high performance microsystem incorporates cutting-edge ultralow power microelectronics and is designed to be technologically flexible and modular, to enable scaling to increasingly complex neural signal extraction and manipulation. Its components imbed adaptive processors for automated calibration and set-up which exploit neural decoding algorithms, currently being developed at Brown, to provide a stable, multipurpose output signal. The microsystem implant will be first tested in animal models (motor cortex of monkeys) to establish efficacy, biocompatibility and biostability. We will input learning acquired from ongoing and future human trial patients with the present percutaneous cabling system to establish principles of human NMP operation for adapting the implantable microsystem design, the role of learning in NMP use, and the limits of human NMP control. We will pursue the federal regulatory pathways to gain approval for the NMP microsystem while developing the assistive technology for human patients. The end goal of this BRP grant is to achieve the implantation and testing of the new microsystem chronically in a human patient. t 'The ultimate vision and guiding light for this BRP proposal is to develop the technological infrastructure to achieve a longer term goal, namely system is the restoration of semi-autonomous, closed-loop, distributed- feedback control of a limb that has lost spinal cord connection to the motor cortex.

DESCRIPTION (provided by applicant): This application addresses broad Challenge Area (01) Behavior, Behavioral Change, and Prevention and specific Challenge Topics: Enabling Technologies 06-HD-101* Improved interfaces for prostheses to improve rehabilitation outcomes;06-NS-107 Sensors to monitor neurologic function and 06-NS-104 Developing and validating assistive neurotechnologies. The overall goal of this RC1 is to demonstrate the ability for humans with tetraplegia to drink a cup of water using a neurally controlled robot arm. The aims directly related to three challenge areas related to rehabilitation, sensor development, and enabling those with disabilities: 06-HD101, NS 104 and 107. This project capitalizes on an exceptional opportunity for persons with tetraplegia involved in pilot clinical trial of a neural interface system, 'BrainGate', to participate in research to develop new means to restore independence and control. Specifically, the research will establish the ability for BrainGate trial participants to use neural signals from their motor cortex to perform useful reach and grasp actions with a robotic arm. This enabling neurotechnology research is made possible by state of the art robots, designed and tested for safe human interactions, capable of human-like reach and grasp movements. The robots will be provided by the robotics group of the German Aerospace Agency DLR, who have developed and tested this robot. This unique opportunity is also made possible by an experienced clinical, research and engineering academic team who are running a new IDE BrainGate2 clinical trial. The work will extend already demonstrated abilities for persons with longstanding severe paralysis to perform 'point and click'computer mouse actions and control simple robots using BrainGate as part of an earlier FDA and IRB approved IDE pilot trial. The first aim is to determine the number of dimensions that can be independently controlled by neural signals and the means to learn to control these dimensions, using simulations of robot arm function and with the physical robot at a distance. The research will establish optimal decoding and training methods for humans to achieve the highest degree of freedom control. The second aim will advance algorithms to improve reliability and stability of performance over time. The third aim is to create the communication link to the LWRIII robot arm. For the fourth aim, physical system use will be evaluated using optimal training and decoding approaches. The ability for a person with tetraplegia to reach out and grasp a cup of water and drink, using the robot under neural control will be demonstrated. This research will advance assistive technologies that would permit substantially greater independence and control for persons with severe movement disabilities. This Challenge Grant aims to develop assistive technology that will allow persons with severe paralysis to be able to reach and grasp objects using their own brain signals. The experiments will test the ability for persons unable to move their arms or legs, resulting from spinal cord injury, stroke, or Lou Gehrig's disease, to control a robotic arm and hand that can safely interact with people. We will demonstrate the ability for a person with paralysis who is part of an ongoing pilot human clinical trial on neural interfaces to pick up and drink a cup of water using only their own brain signals. This technology could lead to a set of new devices that markedly enhance quality of life and independence of people with severe disabilities.

DESCRIPTION (provided by applicant): Computational requirements of contemporary brain science research often exceed financial and resource management limits of individual investigator laboratories. Many contemporary neuroscience research projects require analysis of large data sets with advanced statistical methods and anatomical reconstruction techniques. These methods require high speed computational and graphics engines operating in a multiple processor environments equipped with large capacity, high-speed storage devices. A limitation in the Brown brain science effort at understanding neural processing is the lack of a readily accessible high-speed computational resource. A central computational resource based on a unified cluster of contemporary Linux CPUs and GPUs will serve the computational needs of a core group of brain science investigators at Brown without compromising individual access common to stand-alone workstations. The requested computer cluster has system software that automatically balances CPU and GPU usage, thereby ensuring maximum access to the computational resource for all users. Intensive 3D graphics are off-loaded either to GPUs or to client workstations, thereby further reducing the central computational load. Commercial or open-source software with an open operating environment will be used for analysis using standard and novel statistical and machine learning approaches to assess significance of large data sets. This proposal details the architecture and benefits of a contemporary computational resource for the major and minor users, and more generally the Brown brain science community. The resource was designed to fill immediate and near-term computational and storage needs of a core group of Brown brain scientists. The system can be readily expansion as needs, either computational, storage, or new users, arise. Expansion of the existing core investigators group can occur easily since the computational power or storage capacity of the system can be readily enhanced at relatively low cost. The flexible nature of the system will serve a variety of research needs of the Brown brain science community. The computational resource is expected to bring together researchers at Brown working on the common problem of neural processing.