UCLA

Our decisions often involve consideration of prospective outcomes, which allow us to assess the consequences of our possible responses before deciding on a course of action. This process requires a detailed representation of how events are related so we can recall this information when appropriate. While normally adaptive, disruptions in this learning process can give rise to maladaptive behaviors underpinning neuropsychiatric disorders, such as with substance abuse. This dissertation investigates the neural substrates for the formation of associative maps between rewarding cues and outcomes, and how these neural circuits are changed with drug experience to contribute to maladaptive decision-making. Inhibition of GABAergic neurons in the lateral hypothalamus (LH) revealed the LH to be important for helping to learn detailed associations between cues and rewards that can be used to influence behavior. Targeted inhibition and stimulation of dopaminergic projections from the ventral tegmental area (VTA) to the LH were identified as facilitating learning about detailed cue-reward associations in LH. A history of methamphetamine experience was shown to increase the control that reward cues have over decision-making and to strengthen LH-VTA circuits. To examine endogenous dopamine activity in the LH and how this might change following drug exposure, we measured dopamine release across learning of cue-reward associations in rats with or without a prior history of methamphetamine self-administration. Dopamine release in the LH was shown to increase to cues and rewards across learning, which constitutes a unique profile of dopamine activity. Importantly, prior methamphetamine self-administration was found to amplify dopamine release to reward cues that emerges across learning. Altogether, these data characterize a circuit between the hypothalamus and the midbrain that supports the acquisition of detailed cue-reward associations that are strengthened with prior drug experiences to heighten the control that reward cues have over behavior.