Mixed Signals: Exploring the role of the Lateral Hypothalamus and Anterior Insula in reward and aversive learning, and its possible implications for addiction.

Addiction can be understood as a disorder of motivation and associative learning. Every day we encounter contexts and cues that have acquired meaning through the outcomes we have associated with them or with actions performed in response to them. In this sense, we can see our everyday behaviour as guided by stimulus-response-outcome processes, the aim of which is to maximise rewards, while minimizing aversive consequences, leading to survival. However, what happens when the rewarding properties of an outcome highjack possible negative consequences? In disorders like addiction, maladaptive positive and negative cue-outcome associations can lead to persistent use despite aversive consequences and relapse, thus opposing survival. In this thesis, I am to identify brain regions and neuronal subpopulations involved in different aspects of the stimulus-response-outcome process that may ultimately lead to addiction. I use the rat as a model for the following reasons: 1) The brain structures investigated are well conserved across species. 2) The genetic tools employed to target specific brain regions and neuronal subpopulations are only possible in rodents. In Chapter 1, I provide an introduction to the cognitive processes underlying addiction, as well as the principles of classical and operant conditioning. I also describe the various behavioral paradigms used throughout this thesis to investigate these processes. Furthermore, I introduce the brain regions examined in this thesis and explain the methodologies employed to target them, offering the necessary context for the experimental chapters that follow. In Chapter 2, we use a novel cue-induced relapse after punishment paradigm to test whether context-induced reinstatement after punishment extends to the use of discriminative cues. Then, using a motivational conflict operant conditioning paradigm, we show that NAcS, PVT, and LH are recruited when faced with conflicting outcomes. In Chapter 3, we use fiber photometry calcium imaging to monitor LH-GABA neurons during alcohol pavlovian conditioning, extinction, and reinstatement. We show that these neurons track learning of cues associated with alcohol, and that this activity is causal, as inactivating these neurons with optogenetics abolished learning. We also show LH-GABA are involved in extinction and reinstatement. In Chapter 4, we show that the role of LH-GABA might extend to encoding of aversive cue-outcome associations by monitoring activity during punishment of a previously associated alcohol predictive cue, as well as during Pavlovian motivational conflict using alcohol, neutral, or punishment associated cues with fiber photometry. In Chapter 5, we show that the aIC might track outcome processing in nicotine self-administration, punishment, and context induced relapse after punishment. We also show that aIC increases activity prior responding and we show that relapse is prevented by inactivating aIC, and that these processes might be mediated by Basolateral Amygdala projections to aIC, as well as interhemispheric projections between contralateral aIC. Together, these experiments show that both the LH and aIC are important nodes for stimulus-response-outcome associations that might play a role in addictive behaviours