Circuit Diagrams

Project Summary The brain uses sensory representations to assess risk and predict reward in order to adjust behavior. Per­ ception is a multisensory process. To make reliable predictions, it is advantageous for the brain to combine more than one sensory modality to represent the world. In humans, as in many species, there is evidence for sophisticated forms of learning, such as crossmodal enhancement, where the integration of multiple stimuli from different modalities facilitates memory formation and/or improves discrimination.

Assessing social signals, such as vocalizations, figures prominently in the outcome of aggressive encounters, including the potential to win a fight or prevent escalation resulting in physical injury. Among vertebrates, the neural circuitry integrating sensation, for example what an individual hears, with modulation of aggressive output, is distributed throughout the forebrain and conserved across vertebrate lineages. Social behavior is thought to be an emergent property of activity across this network, and audition plays a prominent role in modulating aggression in primates and birds.

Despite the fact that visual perception represents such a fundamental aspect of our everyday life, our knowledge of the underlying neural coding of natural stimuli is woefully lacking. One major limitation preventing our understanding of the neural underpinnings of natural vision is the lack of viable methodologies for recording and synchronizing eye movements and incoming visual stimuli from freely-moving monkeys during unrestrained exploratory behavior.

The neuromodulators acetylcholine (ACh) and norepinephrine (NE) are associated with an activated cortical brain state characterized by an increase in the reliability of cortical responses to external stimuli and enhanced performance on behavioral tasks. A key unresolved question is the spatial and temporal scale at which these neuromodulators exert their effects. New methods to directly record the local availability of ACh and NE simultaneously with the activity of neural populations in mice opens up the possibility to answer this question.

The goal of this project is to provide the building blocks for an independent research program focused on the mechanisms by which neural networks incorporate multisensory cues into episodic memories. Discrimination of different contexts composed of distinct constellations of multisensory cues is a hallmark of both episodic memory and spatial navigation, two functions ascribed to the mammalian hippocampus.

ABSTRACT Perceptual decision-making is a fundamental cognitive ability that is vital to healthy, daily functioning and is impaired in many diseases. Although many brain regions are known to be involved, there is no clear brain-wide model of how perceptual decisions are formed and executed and the underlying circuit mechanisms are still largely unknown.

Classical conditioning has been studied in many different animal models, and even in humans. However, the larval zebrafish with its transparent brain offers a unique opportunity to observe large scale changes in synaptic structure that accompany this form of learning. Accordingly, we have developed a novel paradigm for visualizing synaptic changes that occur during classical conditioning in larval zebrafish. Using this paradigm we have observed striking region-specific changes in the distributions of synapses that drive the rewiring of neural circuits that mediate threat responses.

Abstract: Attachment powerfully shapes our development and remains a primary driver of health and well-being in adulthood; disruption of attachments is highly traumatic. While affiliation, defined as general positive social interactions, is shared widely among mammals, attachment, or selective affiliation as a result of a bond, is far rarer and of primary relevance to humans. While affiliation has been studied in a number of contexts, how the neural circuitry that underlies affiliation ultimately contributes to adult attachment remains largely unknown.

Project Summary This proposal explores an emergent computational framework for understanding the neural population codes that support flexible, context-dependent behavior. The current state of the field is based on two competing views. According to the circuits view, fixed behaviors arise from specific anatomically or genetically defined cell populations that serve specific functions. Alternatively, the network computation view instead holds that neural activity provides mixed representations of task variables and can be understood only based on the joint activation of many neurons.

Abstract Central to human and animal cognition is the idea of internal models: an internal repository of knowledge about the structure of the world and its affordances that enables prediction and planning. The existence of such models is fundamental to experience. As we move through the world, the raw instantaneous sensory information that we receive is highly impoverished and dynamic relative to the rich, organized, stable and detailed nature of experience.