Computational models of addiction often rely on a model-free reinforcement learning (RL) formulation, owing to the close associations between model-free RL, habitual behavior and the dopaminergic system. However, such formulations typically do not capture key ... ...

请注册登录后继续浏览