Michael S. Brainard - US grants

[unreadable] DESCRIPTION (provided by applicant): Vocal learning by songbirds provides a model for studying general mechanisms of sensorimotor learning with particular relevance to human speech learning. For both songbirds and humans, hearing the sounds of others, and auditory feedback of oneself, plays a central role in vocal learning. Moreover, both song and speech learning are subject to critical periods: appropriate experience in early life is necessary for normal vocal learning and lack of that experience can lead to permanent deficits in nervous system function. Our previous work suggests that a basal ganglia-forebrain pathway participates in processing auditory feedback and in driving experience-dependent changes to vocalizations. Here, we propose to use a new approach to directly investigate at a behavioral and neural level how auditory feedback contributes to learning and production of songs. We will use systematic manipulations of auditory feedback to characterize more precisely how song behavior depends on hearing (Aim 1). We will couple feedback manipulations with chronic neural recordings from awake behaving birds to determine the nature of signals elicited by altered feedback in vocal control structures (Aim 2). We will also use song-triggered microstimulation of these same structures to test their functional influence on song production and song plasticity (Aim 3). Songbirds provide a system where the influence of performance-based feedback on a well-defined and quantifiable behavior potentially can be understood at a mechanistic level. Such an understanding will provide basic insight into normal learning processes and contribute to our ability to prevent and correct disabilities that arise from dysfunction of these processes. [unreadable] [unreadable]

stimulus /response

DESCRIPTION (provided by applicant): While great advances have been made in understanding the mechanisms of learning in the single synapse or cell, a large gap remains between this understanding and our knowledge of learning at the behavioral level. We know that the activity of large-scale neuronal circuits gives rise to behavior, yet we have little knowledge of what changes in those circuits during learning or how sensory feedback drives these changes. The biggest impediment to answering these questions is the inability to quantitatively measure large-scale circuit properties (e.g. connectivity between brain areas) or to precisely manipulate the activity patterns across these circuits. Optogenetics offers the potential to bridge this gap by allowing the direct control of neural activation in targeted cell types on the millisecond timescale. The development of these tools is progressing most rapidly in mouse, due to the relative ease of genetic manipulations in that species. In contrast, behavioral and circuit-level studies of learning are most practical and have been most successful in "non-genetic" species. Within our team, we have expertise in studying both the behavioral and neural bases of learning in rat, songbird, and nonhuman primate. We propose to develop the optogenetic tools and experimental techniques required to study the circuit-level mechanisms of learning in these species and to apply these to two specific scientific aims: Aim 1: Determine the functional connectivity of learning-related circuitry and how it is altered by experience. It is widely presumed that learning relies on the ability of instructive signals to drive functional modifications of connectivity in the circuits that underlie behavior. However, the tools for measuring functional connectivity in vivo have been limited. We will overcome this limitation using temporally and/or spatially precise optical activation of neurons within a circuit. Functional connectivity will be measured by recording optical-stimulation-triggered changes in activity in downstream neurons. We will assess how functional connectivity is dynamically altered by learning and by factors that may contribute crucially to learning. Connectivity changes will serve as a mechanistic index of the nature and sites of the plasticity that give rise to behavioral change. Aim 2: Test the causal contributions of patterned activity to learning in vivo. Prior research has generated specific and testable hypotheses about how and where patterned activity drives learning. Yet support for these hypotheses has derived primarily from correlative observations of activity during learning rather than causal tests of the proposed mechanisms. We will use optogenetics to causally test the contributions of patterned activity to learning. We will test the sufficiency of instructive signals by imposing precisely controlled patterns of activity at defined loci in a circuit and test their necessity by eliminating the putative signals for learning. PROJECT NARRATIVE This project is aimed at revolutionizing the study of the mechanisms of learning within large neural circuits in the brain by directly measuring large-scale properties of these circuits and precisely manipulating circuit activity. To accomplish this, we will make use of, and continue to develop, advanced new techniques that permit the control of specific population of neurons using optical stimulation (light). The knowledge and tools that we gain from these studies are likely to find broad application in the search for treatments of neurological disorders.

The proposed research will initiate a cross-institutional collaboration between an electrophysiologist and a computational neuroscientist to examine fundamental questions about how the brain produces complex sequences of behaviors. Research will be conducted using songbirds, animals that produce a richly structured sequence of highly reproducible song syllables. Previous studies have recorded the activity individual brain cells while birds are singing, and have shown that neural activity is locked to song production at a temporal scale of several thousands of a second, a level of precision rarely achieved in the study of complex natural behaviors. This proposal will exploit this precise relationship between brain activity and sequential behavior by recording from individual birds as they sing many song renditions. Sophisticated analysis tools will then be used to examine both brain activity and song output in great detail. Subtle variations in syllable features and syllable sequencing will be used as ?natural experiments? to determine how the precise activity of individual nerve cells are grouped together to form behavioral units (the syllables), and how these groupings of neural activity are strung together into syllable sequences. Guided by these data, computer models will be constructed to better understand the underlying biological mechanisms that orchestrate complex sequences of brain activity. As part of the proposed interdisciplinary research, novel algorithms for the fine-grained analysis of vocal behavior will be developed, and these may find application in related fields ranging from motor control to robotics to human speech. The proposal also supports a cross-institutional student exchange between the University of Texas at San Antonio, a Hispanic-Serving Institution, and the University of California San Francisco, a world-renowned center for biomedical research.

Neural responses are noisy. Yet behavior is remarkably accurate and precise. How is this accomplished? Is variation in neural responses simply noise that needs to be reduced? Or is it an important signal that is used to control behavior and learning? The seven projects in this application for a Conte Center for Neuroscience Research (CCNR) will ask how neural variation is used to drive behavior, determining where noise falls on the continuum from being a detriment for brain function, neutral, or an asset. The responses of a group of neurons for a given stimulus or action is called a "central representation", and these exist throughout the brain for processing complex sensory stimuli, or planning and executing complex movements. The cerebral cortex has particularly variable neural codes, yet has been widely implicated in guiding normal performance and learning for both sensation and action. This CCNR has the long-term goal of understanding the role of variation as a component of the neural code in the cortex and related areas of the basal ganglia. A related long-term goal is to understand how abnormal variation contributes to failures of adaptation, reductions in the precision and accuracy of behavior, and ultimately to the symptoms of neuro-behavioral and neuro- psychiatric disorders. The seven projects have three center-wide specific aims. First, they will characterize the variation in specific neural codes and behaviors and divide the variation into the component that is related to behavior and one that comprises unrelated, or "residual", variation. Second, they will ask whether natural modes of neural modulation such as attention and reward, or chemical modulatory systems, alter neural variation, and thereby behavior. Third, they will investigate the relationship between changes in neural variation and behavioral learning, seeking cause-and-effect relationships. The seven projects will ask these questions on six different behaviors: smooth pursuit eye movements, reaching arm movements, bird song, spatial behavior, auditory processing, and categorization of speech sounds. They will use species including rodents, songbirds, non-human primates, and humans. The ultimate goal of the CCNR will be to understand the relationship between neural variation and normal behavior to enable creation of behavioral and chemical therapies for treating and curing neuro-behavioral disorders.

Cortical-basal ganglia (CBG) circuits are critical for normal motor and reinforcement learning, and accordingly are a major site of psychiatric and neurological pathology, including schizophrenia and addictions. However, because of their cellular and circuit complexity, and the broad suite of behaviors they control, many basic questions about the link between neural signals in these circuits and behavior remain unanswered. Songbirds provide a powerful model in this regard, because they have a specialized CBG loop, the anterior forebrain pathway (AFP), devoted to a well-defined learned behavior, song. Recent studies in songbirds, including our own work from the last grant cycle, have suggested an important new function for CBG circuits, the active generation of behavioral variability important for learning. Consistent with this, LMAN, the 'cortical' outflow nucleus of the AFP, carries both a temporally structured signal related to song, and trial-by-trial variability around that signal. Moreover, both neural and behavioral variability can markedly decrease when birds are in a social, 'performance' context, possibly in response to midbrain dopamine (DA) release. We will now ask how and where neural variability and its underlying pattern emerge in the song network, and how the different stages of the AFP contribute to these signals and to song in both juveniles and adults, using recordings of multiple neurons in combination with behavioral and pharmacological manipulations of the AFP. Our first aim will study the first nucleus in the circuit, the striato-pallidal Area X , during both alone ('undirected') and female-directed singing, and will test possible sources of context-dependent variation by acutely blocking LMAN recurrent inputs or DA receptors. Our second aim will further test the importance of different stages of the AFP by examining LMAN activity in response to manipulations of Area X, and of its thalamic target DLM. Finally, we will test how both variability and patterned signals emerge and evolve in LMAN during learning in juvenile birds, and how they respond to social cues. The systematic dissection of circuit function possible in this system should shed light not only on normal learning, but on the many diseases of these circuits, whose symptoms often include too little or too much variability.

DESCRIPTION (provided by applicant): Processing and perception of communication sounds such as speech are enormously important to humans, and hearing is a critical component of both. Yet we know surprisingly little about how the brain transforms auditory signals to accomplish these complex tasks. Our long-term objective is to further the development of a unified framework for understanding high-level auditory processing and how it changes, for good or ill, with experience and learning. To this end, we propose to study a stage in a central auditory circuit where there is a remarkable change from simple, primary-like response properties to complex vocalization-sensitivity, in the auditory forebrain of songbirds. These animals provide an excellent model for extracting general principles of higher-level sound processing, because they learn auditory tasks of similar difficulty to humans, and possess a hierarchical network of auditory areas that sub serve these tasks, including the avian equivalent of primary auditory cortex, field L, and several secondary areas that are the likely equivalent of belt cortex. Moreover, we have access to a rich set of auditory stimuli of behavioral relevance, songs. We recently found that within field L there is a strikingly orderly organization of receptiv fields, along spectral and temporal axes. We will ask whether and how this organization propagates to the next level, by mapping the response selectivity of these secondary areas to batteries of songs, using an information theory-based technique called maximally informative dimensions (MID). We will also record responses to songs learned earlier (from father, mate) to examine the additional hypothesis that sound memories learned early in life have a special representation in such areas. In parallel, we will test the anatomical and physiological contribution of the inputs from field L to these secondary areas, by selectively labeling or turnin off subsets of these inputs. Finally, armed with the knowledge of these circuits' organization, we will ask whether and how their receptive field organization and song representations change after birds learn a behavioral discrimination task that strongly focuses their attention on either spectral or temporal aspects of song, two key parameters mapped in field L, and critical both in normal and impaired auditory processing.