Associative learning, decision-making, and addiction

National Institute On Drug Abuse

Investigators

Linked publications, trials & patents

Abstract

Substance use disorder (SUD) and other neuropsychiatric diseases involve altered learning and decision-making. The mechanism of these alterations, the circuits affected, and their significance to behavior are not well understood. We are using an animal models of SUD, associative learning and decision making to study the neural circuits mediating simple associative learning and decision-making and to identify critical changes that may underlie SUD and other co-morbid neuropsychiatric disorders. It is hoped that by better understanding these changes, we will have better models for intervention. In the current year, we have published papers and abstracts and presented data at meetings, which broadly address how neural circuits mediate associative learning and decision-making and how these processes are altered by use of addictive drugs. Several studies have focused on the role of dopamine in learning. Dopamine is linked closely to the mechanism of drugs of abuse and is also closely related to learning functions, thus understanding its role in the brain is likely central to understanding SUD. In key work this year, we have published data showing that dopaminergic teaching signals reflect prediction errors outside the realm of value. That dopamine can signal multidimensional prediction errors dramatically expands the scope of learning dopamine may impact. This work involved close collaboration with computational modelers, which we also engaged in several other projects. In collaborative work we also have published on signaling of identity prediction errors by dopamine systems in humans and on the dependence of this signal on predictions from the orbitofrontal cortex. In other work focused on understanding cortical processing, we published new data elaborating on the role of the orbitofrontal cortex in learning conflicting rule sets. Ongoing work is extending this to the dopamine system. Additional work in collaboration with the Kahnt lab has extended these results to humans, linking our rodent finding to human neural systems. Additionally the lab has done well on a practical basis - two postdocs left to start their own labs, one postbacc entered medical school.