The role of the lateral hypothalamus in the balance of learning and behavior towards relevant stimuli

PROJECT SUMMARY Growing evidence suggests addiction results in part from a dominance that drug-paired cues exert over behavior. That is, when trying to remain abstinent, people with substance-use disorders (SUDs) struggle to focus on anything that is not related to drug use when reminders of drug use are present, exacerbating craving and

producing relapse. We have recently uncovered a novel neural locus that appears to bias learning and behavior towards cues that are immediately related to rewards. Specifically, we have found that GABAergic neurons in the lateral hypothalamus (LH) drive learning and behavior directed to information related to rewards, while

actively opposing learning about other relationships in the environment that are not immediately relevant to rewards. This is the first time one region has been found to contribute to this dissociation. We can see how a change in this balance in learning and behavior mediated by LH could contribute to addiction; increases in LH

function seen following drug exposure could potentiate cue-elicited drug seeking while trying to remain abstinent, hijacking the adaptive function of LH to focus on cues that help predict rewards that likely promote survival under normal circumstances. This work will provide insight into how the pathophysiological state of addiction can

contribute to relapse even during voluntary abstinence, and suggest a neural target that could drive pre-clinical work to develop treatments for SUDs. This proposal will test a novel theoretical framework to account for the influence LH has over learning. Specifically, we hypothesize that LH GABAergic neurons bias learning towards cues proximal to rewards, and

away from those more distal to rewards. To test this, we have developed an innovative task taken from the cognitive neuroscience literature based on human participants, allowing us to isolate behavior and neural activity during both proximal and distal predictors of rewards in the same animal. Further, we will uncover the nature of

learning about proximal cues supported by LH function. On the basis of preliminary data, we hypothesize this will involve a sensory-specific representation of reward, which allows LH to influence behavior in flexible and specific ways. Finally, our preliminary work also demonstrates that methamphetamine self-administration

enhances the control that cues proximal to rewards have over behavior, and that this is also dependent on a specific reward representation. This expands our knowledge of the behavioral change that results from SUDs, standing as a counterpoint to many theories of addiction that focus on and enhancement of habitual behavior,

which is without voluntary control and devoid of reward representations. On the basis of preliminary work, we hypothesize that enhancements of LH GABAergic neuronal function after drug exposure produces this effect. Accordingly, we will directly test our theory that LH GABA neurons promote a bias towards proximal cues in a

specific manner, and that this function is enhanced in following drug exposure.