Background: Markers of chronic cocaine exposure on neural mechanisms in animals and humans is of great interest. The probabilistic reversal-learning task may be an effective way to examine dysfunction associated with cocaine addiction. However the exact nature of the performance deficits observed in cocaine users has yet to be disambiguated. Method: Data from a probabilistic reversal-learning task performed by 45 cocaine users and 41 controls was compared and fit to a Bayesian hidden Markov model (HMM). Results: Cocaine users demonstrated the predicted performance deficit in achieving the reversal criterion relative to controls. The deficit appeared to be due to excessive switching behavior as evidenced by responsivity to false feedback and spontaneous switching. This decision-making behavior could be captured by a single parameter in an HMM and did not require an additional parameter to represent perseverative errors. Conclusions: Cocaine users are characterized by excessive switching behavior on the reversal-learning task. While there may be a compulsive component to behavior on this task, impulsive decision-making may be more relevant to observed impairment. This is important in building diagnostic tools to quantify the degree to which each type of dysfunction is present in individuals, and may play a role in developing treatments for those dysfunctions.
Bibliographical noteFunding Information:
Funding for this study was provided by the National Institute of Drug Abuse (NIDA) grant P20DA024196 . NIDA had no further role in study design; in the collection; analysis and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication.
- Bayesian hidden Markov model
- Decision making
- Reversal learning
- State switching