Joseph E. LeDoux - US grants

DESCRIPTION (provided by applicant): Anxiety disorders depend, in part, on alterations in the mechanisms by which fear is learned, stored and expressed in the brain. Much has been learned about the neural basis of fear through studies of Pavlovian fear conditioning, a procedure in which an emotionally neutral conditioned stimulus (CS), such as a tone, is paired with an aversive unconditioned stimulus (US), typically electric shock. Following pairing, the CS acquires the capacity to elicit conditioned responses (CRs), such as freezing behavior and changes in autonomic and endocrine activity, that are part of the organism's defensive repertoire. The neural pathways involved include transmission of CS information from the auditory system to the lateral amygdala (LA), where the CS-US association is formed and stored. The LA then communicates with the central amygdala (CE) via intra-amygdala circuits to control the expression of the conditioned fear reactions. In addition to eliciting fear reactions, an aversive CS also alters instrumental behaviors (actions). For example, patients with anxiety disorders often exhibit behaviors that reduce exposure to fear-eliciting stimuli or situations. So-called avoidance responses can be effective coping strategies, but become problematic when they interfere with daily life. Less is known about the neural basis of fear-based instrumental actions than is known about fear reactions like freezing. In this proposal we use tasks that provide means of assessing how a Pavlovian aversive CS contributes to aversive instrumental action. The traditional task used for this is avoidance conditioning. However, because the Pavlovian and instrumental learning components of the task are intermixed in avoidance conditioning, this procedure is not ideal for examining Pavlovian influences on instrumental behavior. We therefore focus on two tasks that isolate the Pavlovian and instrumental learning phases: escape from fear (EFF) and Pavlovian-to-instrumental transfer (PIT). EFF is particularly useful for exploring how a Pavlovian CS contributes to the acquisition of new instrumental responses. In this task, the CS-US association is first conditioned. Then, in a different situation, the animal learns to perform responses that terminate the CS. CS termination is believed to function as a negative conditioned reinforcer, an event whose termination reinforces behavior. PIT, on the other hand, is useful for exploring how a Pavlovian CS alters the motivation to perform a previously acquired instrumental response. In this task, the CS-US association is also conditioned first and then used to alter the vigor with which the subject performs a previously trained instrumental response. We will use the same CS and US (a tone paired with shock) that we and others have used to identify the brain mechanisms of Pavlovian conditioning. A combination of lesion and inactivation techniques and measurement of physiological activity will be used to examine the contribution of distinct amygdala nuclei and extra-amygdala targets to these tasks. Given existing findings with these and related tasks, we hypothesize distinct roles of three amygdala regions: LA, CE and the basal nucleus (B). In the previous submission, we hypothesized that LA and B would be required in both the EFF and PIT tasks: LA because it stores critical aspects of the CS-US association used in the acquisition of new instrumental responses, and B because it allows the CS to function as a negative conditioned reinforcer in the learning of the instrumental response, and also allows the CS to function as a conditioned incentive in the motivation of performance. Although we proposed that B would be involved in both negative conditioned reinforcement and conditioned motivation, we also proposed that different cells in B would contribute to these processes. New preliminary findings lead to a revised hypothesis. We now propose two different forms of aversive conditioned motivation involving two different amygdala circuits-connections between LA and B are involved in one form and connections from LA to CE in the other. Building on similar (though not identical) findings in appetitive conditioning, we suggest that the LA-B connection is important for using specific information about the CS-US association to motivate behavior, while the LA-CE connection is more important for using the emotional arousal triggered by the CS to motivate behavior. The involvement of B and CE in these forms of motivation is consistent with appetitive findings, but the idea that B and CE contribute to conditioned motivation as distinct outputs of LA diverges with appetitive findings. In addition, we will pursue the output connections of the amygdala in these tasks. We hypothesize that the nucleus accumbens is a potential site where conditioned reinforcement and the two forms of conditioned motivation, processed by distinct outputs from the amygdala, are integrated in the acquisition and performance of aversive instrumental behavior. The studies proposed will provide new information about a how a fear-arousing CS influences instrumental actions by contributing to aversive negative conditioned reinforcement and conditioned motivation. Given that anxiety disorders involve both pathological reactions and actions, the results of this work should greatly extend our understanding of brain circuits relevant to this common class of psychiatric disorders. Better appreciation of aversive instrumental behavior may also provide insight into the relationship between aversive and appetitive processes, and thus may also be relevant to understanding how fear and anxiety contribute to addiction and other disorders that ar traditionally viewed as primarily involving appetitive motivation.

DESCRIPTION: (Adapted from applicant's abstract): Considerable evidence points to the amygdala, and specifically its lateral nucleus (LA), as a key site of the plasticity underlying fear conditioning, a behavioral procedure useful for studying how fear, including psychopathological fear, is learned and remembered. In this proposal past findings are built upon in an attempt to gain a deeper understanding of the nature of plasticity that occurs at synapses linking neurons in the auditory system (thalamus and cortex) with postsynaptic neurons in LA, and the relation of this plasticity to conditioned fear behavior. To do this, four broad projects, all involving rats, are proposed. The first examines synaptic plasticity in LA in vitro, as this level is useful for pursuing cellular/molecular mechanisms that might underlie plasticity. Mechanisms uncovered in vitro are then examined for relevance to living brains by performing similar in vivo studies of LA plasticity using intracellular recordings in anesthetized animals. The findings from anesthesia are then evaluated for their relevance to awake animals by performing behavioral and physiological studies in behaving rats. The final project attempts to take a step forward from the mechanisms of plasticity at the input synapses in LA and begins to look at the contribution of local circuits within the amygdala. Much of the work at all levels is aimed at determining the relative contribution the L-type voltage gated calcium channels and the NMDA class of glutamate receptors to synaptic plasticity in fear conditioning circuits, and to fear conditioning itself. Together, these studies should reveal new information about the cellular basis of fear learning, and may, in the long run, lead to better understanding of the genesis and maintenance, and hopefully the treatment, of pathological fear.

When exposed to a painful or threatening stimulus, our brains form neural associations that link the painful or threatening stimulus to other innocuous environmental cues that are present and that might later serve as warning signals. The process of forming such associations is referred to as fear conditioning. Some of the innocuous cues that get conditioned are in the foreground and are closely associated with the pain or threat, whereas others are in the background. The neural system underlying the conditioning of fear responses to specific foreground cues, such as a tone paired with footshock, is well understood and involve the amygdala. However, until recently, little was known about how fear reactions come to be coupled to background or contextual stimuli. In recent studies Dr. LeDoux found that lesions of hippocampus or amygdala interfered with contextual conditioning. The purpose of the proposed studies is to use state-of-the-art anatomical, physiological, and behavioral methods to understand how neural interactions between amygdala and hippocampus might contribute to contextual fear conditioning. Specifically, Dr. LeDoux will explore the anatomical connections (afferents and efferents) underlying the contribution of the hippocampal formation to contextual fear conditioning. These studies are likely to be important in understanding a host of complex emotional learning phenomena.*** //

[unreadable] DESCRIPTION (provided by applicant): Considerable evidence points to the amygdala, and specifically its lateral nucleus (LA), as a key site of the plasticity underlying the learning and storage of information about threatening or harmful life events. Most of what is known about this form of learning and memory has come from studies of classical or Pavlovian fear conditioning. The candidate for this K05 Award has contributed significantly to this body of work on the role of the amygdala in fear conditioning. His career plans include additional training in several areas that will allow him to pursue new lines of work on this topic. Specifically, he seeks training in molecular neuroscience and functional imaging of the human brain, and also hopes to achieve a deeper understanding of fear/anxiety disorders. He has assembled an international team of collaborators who will guide his training in these areas. Four sets of studies will be performed, each of which represents a funded area of research. The first project attempts to understand in greater detail the neural system underlying fear conditioning, and to determine how this basic circuitry interacts with systems involved in cognitive control over mental and behavioral functions. The second project addresses questions about the cellular and molecular mechanisms that underlie fear conditioning and attempts to learn more about the receptor mechanisms, signaling pathways, genes, and proteins involved. The next two projects use functional imaging (fMRI) to examine the mechanisms of fear in normal humans (project 3) pathological fear in patients with anxiety disorders (project 4). Together, these studies should reveal new information about the neural basis of fear learning and memory at the systems, cellular and molecular levels, and should provide a better understanding of how fear mechanisms in experimental animals relate to brain mechanisms in normal humans and patients with anxiety disorders.

Considerable progress has been made in elucidating the neural pathways underlying conditioned fear in animals. This work has implicated circuits centered around the amygdala, and interactions between the amygdala, hippocampus and medial frontal cortex (mPFC), in the acquisition and/or expression of different aspects of conditioned fear. Research in humans has confirmed essential aspects of the animal work. Further, the same trio of brain regions (amygdala, hippocampus, mPFC) are also implicated in the regulation of stress hormones and are also often altered in patients with fear/anxiety disorders. The overall hypothesis guiding research in the Center for the Neuroscience of Fear and Anxiety (CNFA) is that the effects of stress on fear circuits in animals mimics changes that occur in the brains of patients with fear-related disorders relative to healthy controls. To test and explore the implications of this hypothesis, we use the same behavioral paradigm, fear conditioning, to study rats, normal humans, and patients with fear disorders. The aim of the animal work is to examine the effects of stress on the behavioral functions, physiology, and morphology of fear circuits. The aim of the studies of normal humans is to use behavioral methods and fMRI to extend our understanding of fear mechanisms in the human brain and to develop new probes for testing patients with fear disorders. And the aim of the studies of patients with fear disorders is to determine whether the patterns of functional brain activation during fear in healthy humans are altered in patients with fear disorders, and whether these alterations are consistent with the effects of stress on fear circuits, as determined in the animal work. In pursuing these goals, several new themes will also be pursued across the various projects, including an emphasis on questions about circuit interactions in fear, the role of individual and sex differences, and treatment related issues. Through this translational research we hope to reveal brain mechanisms that are altered in patients with fear/anxiety disorders and, in the long run, to identify new hypotheses about how to treat or prevent such disorders.

DESCRIPTION (Adapted from applicant's abstract): The Quantitative Morphology Core (QMC), directed by Dr. Patrick Hof, provides expertise with quantitative neuroanatomical methods and image analysis, as well as help in study design to the Center's investigators. An important feature of the QMC is its capability to cross levels of morphological resolution, from macroscopic features to subcellular characteristics, to provide a comprehensive understanding of neurochemical morphology, not only in the context of functional circuits, but also in relation to the behavioral and electrophysiological paradigms employed by the Center's investigators. The QMC is equipped with two computer-assisted photomicroscopes, one confocal laser scanning microscope and has full access to an electron microscope facility. Quantitative morphology is performed principally using NeuroZoom, a software application designed for full-scale mapping, cell reconstruction, image analysis and stereology. In view of the very high level of interaction in terms of quantitative anatomic analyses among the different members of the Center, a key role of the QMC is to generate a uniform approach to quantitative neuroanatomy in order to ensure a high level of consistency in analysis, statistics, and interpretation of the morphologic data. Specifically, the QMC will participate in collaborative studies of glucocorticoid and glutamate receptor localization in projection neurons in the medial prefrontal cortex and amygdala as well as in intra-amygdala circuits (with Projects 1 and 3), stereologic analysis of neuron numbers and structure volumes of select subfields in the hippocampus (with Project 4), and quantitative cellular reconstructions of physiologically characterized and dye-filled neurons to assess the degree of stress-induced dendritic atrophy in all three brain regions (with Projects 1 and 3). In addition, the QMC will continue to develop new methods to optimize sampling on confocal images and to permit stereology on electron microscopy materials. It will collaborate with Projects 2 and 4 to provide neuroanatomical expertise to interpret MRI scans from experimental animals and human subjects, particularly with respect to delineation of regional boundaries in the hippocampal formation, and will assist with statistical analyses of regional volumes estimates calculated from the MRI.

DESCRIPTION: (Adapted from applicant's abstract): The overall hypothesis being explored is that the amygdala plays a significant role in normal and pathological fear, and that interactions between the amygdala, hippocampus and mPFC play key roles in the normal regulation of fear and the modulation of fear by stress. In particular, we propose that stress alters the function and/or structure of fear circuits (involving areas within and connections between the amygdala, hippocampus, and medial prefrontal cortex) and thereby gives rise to exacerbated fear reactions that are unconstrained by the context in which they are learned. The studies proposed examine humans and experimental animals. Study 1, performed together with Liz Phelps, will use fMRI to examine the normal patterns of functional activation occurring during fear conditioning. These studies will involve collaborations with Silbersweig, Stern, and Gorman and will serve to establish techniques and basal conditions under which to evaluate functional changes seen in patients with fear disorders, as explored in Project 2 by Silbersweig, Stem and Gorman. The facilities and expertise of the Functional Neuroimaging Core are used throughout Study 1. The remaining studies turn to experimental animals (rats) and explore various issues related to fear and stress. Study 2 maps the pathways through which fear stimuli processed by the amygdala control HPA function. Study 3 uses amygdala and hippocampal-dependent aspects of fear conditioning to test the hypothesis that the amygdala and hippocampus are affected in opposite ways by stress. Study 4 examines the effects of drugs with clinical efficacy in treating anxiety disorders (especially the SSRI class of drugs) on fear conditioning and on the effects of stress on fear conditioning. The studies draw upon the expertise of the LeDoux lab in physiological and behavioral approaches to the fear system, and involve collaborations with McEwen's lab (Project 4) on various studies involving stress, and with Morrison (Project 3) in studies of morphological changes in glutamate receptors induced by fear conditioning and stress. The morphological studies also rely on the Quantitative Morphology Core. Project 1 thus involves collaborations with all aspects of the Center and serves as the conceptual launch pad for several of the other investigations.

Lay abstract

Memory is generally believed to involve alterations in the ease with which neurons communicate with one another across synapses. These alterations in turn are made possible by the production of proteins under the guidance of genes. A major impediment to the identification of the relevant genes is the lack of information about which neurons and synapses are involved. In the case of memory formed by fear conditioning, the crucial synapses are believed to be in the lateral nucleus of the amygdala. Armed with this information, it should be possible to use the recently developed microarray technique to screen genes that are activated during memory formation. This approach, when combined with another recently developed tool, laser captured microdissection, should provide a powerful means for probing gene expression in specific cells in the amygdala that change during memory formation. The findings may also be relevant to other forms of memory subserved by other brain regions since existing data indicate that common molecular pathways are involved in different forms of learning and memory.

GRANT=P50MH58911-06-0005 The focus of Project 1 is on the normal mechanisms of fear, and the effects of stress on these in both rats and humans. All studies involve interactions with other members of the Center, as described below. Six rat studies are proposed. Study 1 examines the effects of stress on conditioned fear, including an analysis of individual and sex differences. The role of glucocorticoids in mediating effects of stress on fear will be determined. Study 2 assesses whether the effects of stress on fear are due to actions of stress hormones in the amygdala or other regions of the conditioned fear circuitry (medial prefrontal cortex, hippocampus, bed nucleus of the stria terminalis). Study 3 evaluates the effects of stress and stress hormones on the physiology of the amygdala cells and their plasticity. Study 4 examines the extent to which stress and/or fear conditioning produce structural changes in dendritic morphology or spine organization in the amygdala, and examines the molecular mechanisms mediating structural changes. Study 5 continues on-going projects on the mechanisms by which mediations like SSRIs regulate conditioned fear. Study 6 explores reconsolidation of fear in an effort to identify, though animal studies, pharmacological agents that are safe and practical to use during memory retrieval in an effort to block the later survival and/or access to traumatic memories. The human studies are led by Elizabeth Phelps of NYU. Results of these will feed into the design of new patient studies for Project 2. Study 1 builds upon the ideas from the rat studies above and examines whether individual or sex differences in trait anxiety, HPA response, or in the response to acute brain activation patterns different between high and low anxious individuals during fear conditioning and active coping. Study 3 examines active coping in patients with anxiety disorders. Studies 4 and 5 explore the possibility that pharmacological manipulation of memory during retrieval might alter the survival or accessibility of memory and thus might be useful in treating traumatic memory symptoms in PTSD.

0340607
LeDoux

Fear is an emotional response characterized by both the subjective state (the feeling of fear) and bodily responses elicited by threatening stimuli. The detection of threats and the production of protective bodily responses (behavioral, autonomic, and endocrine changes) are a vital part of biological survival. The brain has circuits that appear pre-wired to respond to some stimuli, such as ancestral threats, and novel stimuli associated with pre-wired dangers. Individuals can use information from past situations to avoid potentially dangerous organisms, objects, and situations. However, the circuits, network properties and computations performed by the brain's fear systems to encode, store and retrieve conditioned fear memories are not yet known.

In this three-year cooperative research project, Joseph LeDoux, Luke Johnson of New York University's Center for Neural Science and Valerie Doyere of the Universite Paris-Sud will identify the underlying neural circuitry and networks in lateral amygdala of the brain responsible for fear behaviors. The project combines complementary expertise of the U.S. and French investigators and will advance understanding of how the mammalian brain encodes, stores and retrieves fear memory.

The National Science Foundation (NSF) and the Centre National de la Recherche Scientifique (CNRS) will jointly support this project. NSF will cover the costs of the U.S. investigator's visits to France. The CNRS will provide funding to the French investigator for travel to the United States.

All organisms use environmental stimuli to guide their behavior in adaptive ways. One important class of adaptive behaviors involves responding to threats, stimuli that cause or predict harm. The overall goal of this project is to address some fundamental issues about the neural circuitry through which learned stimuli trigger responses to dangerous or threatening stimuli. To study this these investigators will use behavioral, physiological and anatomical approaches to understand how neutral stimulus that has acquired threat-arousing properties is processed in the amygdala, which is known to be crucially involved in responding to threats. A major controversy has been whether a fearful (threat-arousing) stimulus must enter the amygdala via the lateral nucleus, as much research has suggested, or whether it may also enter the amygdala via the central nucleus. These are called the serial and parallel models of the amygdala, respectively. The studies proposed will test which model is most viable. This project predicts that the serial model is correct. This work will expand our understanding of how organisms adapt to dangerous environmental conditions and the results will also further our understanding of the circuits underlying the learning and memory storage of emotional information. The project will offer the opportunity for training postdoctoral, doctoral, and undergraduate students in basic approaches to the neuroscience of emotion, learning, and memory.

? DESCRIPTION (provided by applicant): In spite of massive amounts of work, the neural basis of addiction remains only partly understood. Much progress has been made in recent years in understanding the motivational role of drugs as positive incentives and rewards. Although it has long known that aversive motivation also plays a role in addiction, this is less clearly understood. External stimuli associated with environmental stress or drug withdrawal are negative reinforces that contribute to instrumental drug seeking and consumption responses by strengthening behaviors that allow escape from and/or avoidance of the aversive states elicited by these stimuli. Because active avoidance conditioning is based on negative reinforcement and involves brain circuits that overlap with addiction, we argue that a detailed understanding of the neural basis of escape/avoidance behavior will provide important information that may allow a deeper understanding of the role of aversive states in substance abuse. While much research was conducted on the neural basis on avoidance in the 1950s and 60s, this work fell out of favor, in part because the results did not lead to a clear understanding of the circuitry. However, in the intervening years, the neural basis of the first phase of avoidance, Pavlovian fear conditioning, has been elucidated in detail. This information makes it possible to revisit the neural basis of avoidance in a new light. In particular, given that we now understand in detail the neural mechanisms through which a neutral environmental stimulus associated with an aversive unconditioned stimulus (US) becomes a Pavlovian conditioned stimulus (CS) that elicits aversive states, we can now build on this information to understand the neural basis of avoidance conditioning, especially if the same stimuli used as CSs and USs (tone and shock) to reveal the neural mechanisms of Pavlovian conditioning are also used in avoidance conditioning. The previously funded grant examined the contribution of the amygdala, a key structure for Pavlovian aversive conditioning, to avoidance. In the present proposal we continue to pursue the role of the amygdala, but in addition also attempt to begin to reveal the broader circuitry involved. Specifically, we examine the role of connections between the subareas of the amygdala and accumbens, in the transition from Pavlovian conditioned reactions to negatively reinforced avoidant actions.

DESCRIPTION (provided by applicant): Anxiety disorders often develop following exposure to stress or trauma. These frequently have a significant fear component and are sometimes called fear/anxiety disorders. Examples include post- traumatic stress disorder (PTSD) and simple phobias. Because only a minority of persons with comparable levels of exposure to stress or trauma develops these disorders, the affected individuals must possess some innate genetic or epigenetic vulnerability. The identification of such vulnerabilities is a major goal of biological psychiatry, and is the main focus of our studies. Most existing knowledge about the brain mechanisms of fear, and hence about the basic biological mechanisms that may be altered in fear/anxiety disorders has been obtained in animal studies of fear conditioning. These studies have pinpointed particular circuits in the amygdala in the acquisition, storage, expression, and regulation of fear responses. Most of these studies have focused on the mechanisms of fear in random populations of animals. However, because most humans with fear/anxiety disorders are believed to represent extremes in the degree of underlying vulnerability, studies in random populations of animals may not be ideal for determining these vulnerabilities. A more promising alternative is to use animals expressing extremes of fear phenotypes. One popular view is that, in particular individuals, vulnerability to the development of fear/anxiety disorders after trauma exposure may, at least in part, involve exaggerated fear learning/memory in response to trauma and/or failure to recover following trauma cessation. To capture these biological phenotypes, we propose to identify animals with (1) extreme fear reactivity behavior (i.e., exhibiting the highest and the lowest levels of conditioned fear responses) and (2) extreme fear extinction behavior (i.e., exhibiting the slowest and the fastestrate of fear memory extinction)as well as animals that exhibit the intermediate levels of fear reactivity and fear extinction, and thus represent average (normal) individuals. Using fear conditioning paradigms developed in our laboratory, animals exhibiting extreme fear reactivity, extreme fear extinction, and intermediate phenotypes will be separated from within a population of outbred rats. Next, using state-of-the art technology [whole transcriptome sequencing (RNA-Seq)]and weighted gene co-expression network analysis, we will study gene expression differences in particular nuclei of the amygdala among the highest, the intermediate, and the lowest fear reactivity phenotypes as well as among the slowest, the intermediate, and the fastest fear extinction phenotypes. The identified differences will help to pinpoint genes and pathways that characterize individuals with high or low liability t fear/anxiety disorders as well as individuals who are resistant to these disorders. These comprehensive studies will determine specific molecular targets for future focused research of fear-related behaviors and illnesses. PUBLIC HEALTH RELEVANCE: Over the past two decades, key aspects of the neural basis of fear have been elucidated through studies of Pavlovian fear conditioning, which have been mostly performed in random populations of animals. However, because most humans with fear/anxiety disorders (e.g., post dramatic stress disorder) are believed to represent extremes in the degree of underlying vulnerability, studies in random populations of animals may not be ideal for determining these vulnerabilities. To pinpoint molecular networks that characterize individuals with high or low liability to fear/anxiety disorders, in the present proposal we will focus on behaviorally identified phenotypes that represent extremes of fear-related behaviors in rats.

Project Summary/Abstract In spite of massive amounts of work, the neural basis of compulsive behavior in anxiety and especially addiction remains poorly understood. Much progress has been made in recent years in understanding the motivational role of drugs as positive incentives and rewards. Although it has long been known that aversive motivation also plays a role in addiction, this is less clearly understood. External stimuli associated with environmental stress or drug withdrawal are negative reinforcers that contribute to instrumental drug seeking and consumption responses by strengthening behaviors that allow escape from and/or avoidance of the aversive states elicited by these stimuli. Because active avoidance conditioning is based on negative reinforcement and involves brain circuits that overlap with addiction, we argue that a detailed understanding of the neural basis of escape/avoidance behavior will provide important information that may allow a deeper understanding of the role of aversive states in substance abuse. While much research was conducted on the neural basis on avoidance in the 1950s and 60s, this work fell out of favor, in part because the results did not lead to a clear understanding of the circuitry. However, in the intervening years, the neural basis of the first phase of avoidance, Pavlovian fear conditioning, has been elucidated in detail. This information makes it possible to revisit the neural basis of avoidance in a new light. In particular, given that we now understand in detail the neural mechanisms through which a neutral environmental stimulus associated with an aversive unconditioned stimulus (US) becomes a Pavlovian conditioned stimulus (CS) that elicits aversive states, we can now build on this information to understand the neural basis of avoidance conditioning. This is especially true if the same stimuli used as CSs and USs (tone and shock) are used to reveal the neural mechanisms of Pavlovian conditioning are also used in avoidance conditioning. The previously funded grant examined the contribution of the amygdala, a key structure for Pavlovian aversive conditioning, to avoidance. In this proposal we continue to pursue the role of the amygdala, but in addition also begin to dissect the broader circuitry involved. Specifically, we examine the role of connections between the subareas of the amygdala and nucleus accumbens, in the transition from Pavlovian conditioned reactions to negatively reinforced avoidant actions. Optogenetic techniques will be used to relate activity in specific amygdalostriatal pathways to discrete stages of avoidance learning and behavior, including precise negative reinforcement events (i.e. CS-termination, US- omission or both). Lastly, because dysfunction in nucleus accumbens endocannabinoid signaling may promote negative reinforcement and compulsions, we will use biochemistry, pharmacology and receptor knockdowns to examine the contribution that endocannabinoid signaling makes to negatively-reinforced avoidance responses.