Abstract
PURPOSE: This study was motivated by Ahmed's (2024) observation that subjects in animal models of addiction lack agency. Animals have been given increasing control over drug-taking in addiction models, but not the opportunity to change their circumstances in ways that favor drug self-administration vs. abstinence. The present study aimed to develop a model wherein rats could intervene to change the choice contingencies available to them. METHODS: Rats were trained to choose between self-administered heroin or sucrose pellets in two choice conditions. In one condition, heroin was cheap and sucrose was expensive. In the other condition, the relative prices of these reinforcers were reversed. Then, rats were allowed to choose between these choice conditions. MAIN FINDINGS: Rats frequently switched between the conditions when the cost of switching was low, and obtained many of the cheap reinforcers available in each condition. As the cost of switching increased, rats switched less often and increased consumption of the cheap reinforcer available in their current condition. When rats could only make one choice between conditions per session, large individual differences were observed, with some rats preferring the cheap heroin condition and others preferring the cheap sucrose condition. CONCLUSIONS: This study found that when rats are given the ability to choose between choices they will take advantage of this additional level of control. The initial environmental contingencies were, however, an important determinant of heroin-taking behavior. This finding is a step towards refining animal models of addiction towards a stronger analogy to human behavior.