Transition to drug addiction: a negative reinforcement model based on an allostatic decrease in reward function

Rationale: The transition from initial drug use to drug addiction has been proposed to result from an allostatic decrease in reward function driven by an over-activation of brain antireward processes. Objectives: How decreased reward function explains compulsive drug use is not entirely clear at present, and is still a subject for debate. Methods: We present a quantitative model of cocaine self-administration that integrates pharmacokinetic, pharmacodynamic, and motivational factors to address this question. The model assumes that reward system responsivity is a homeostatically regulated process where the desired level of responsivity ( called the reward set point) is initially different from the baseline level. The reduction or correction of this difference or error in reward function would drive cocaine self-administration. Results: Theoretical data obtained by computer simulation fit the experimental data obtained in animals self-administering cocaine (i.e.,the within-session pattern of self-injections, the shape and curvature of the dose-injection function, the nonlinear relationship between drug intake and regulated drug effects). Importantly, simulation of an allostatic decrease in reward system responsivity exacerbates the initial error that drives self-administration, thereby increasing both the intake of, and the motivation for, the drug. This allostatic change manifests as a vertical shift in the dose-injection function similar to that seen in animals with escalating cocaine self-administration. Conclusions: The present model provides a satisfactory explanation of escalated drug intake and suggests a novel negative reinforcement view of addiction based on an allostatic decrease in reward function.