Computational Models: Applications to Drug Abuse

United States

The National Institute on Drug Abuse (NIDA) is interested in developing and fostering a computational modeling program that can use new perspectives, models and paradigms to better understand drug abuse and addiction. NIDA sponsored a workshop held May 30-31, 2000 on "Computational Models: Applications to Drug Abuse," to explore how computational and theoretical modeling could be used in drug abuse research. Workshop participants included researchers studying the neurobiology and behavior of drug abuse and researchers using computational modeling techniques in a variety of other applications. The purpose of this meeting was to explore with experts in drug abuse research and in computational and theoretical modeling the needs, gaps and promising areas of research, and to begin a discussion of possible approaches NIDA could take to foster computational and theoretical approaches to drug abuse. In the course of the two-day meeting, the participants considered these broad issues and, at the same time, discussed some specific challenges and questions facing this potential and emerging field in their presentations and discussions. These questions included:

1. Can we model our current knowledge related to substance abuse and addiction to gain additional insight into underlying neurobiological, cognitive and behavioral processes?
2. What are the most promising areas of research that could benefit from a modeling approach in the near term?
3. What additional cognitive, neurobiological and behavioral data would be useful for developing models of drug abuse and addiction?
4. What existing models and computational techniques can be applied to substance abuse and addiction, and is there a need for substantial development of new techniques? For example, do we need new reinforcement models, and do we need additional dynamic models to accommodate drug effects on cellular homeostasis?

Conclusions

1. The participants saw a great potential for collaborations that would apply computational methodology to problems of importance for a scientific understanding of drug abuse and addiction, such as neural modulation, cellular homeostasis, learning and memory, decision making and economics, and behavioral flexibility.
2. Participants saw value, and application to drug abuse and addiction, in developing computational models at all levels of analysis from macro level economic and behavioral models through micro-level cellular regulation and homeostasis models.
3. Both theoretical modeling approaches as well as those more integrated with experimental data were thought to be equally important for drug abuse research.
4. The participants saw a particular need for pre- and post-doctoral training in computational approaches and the need to strongly support newly independent faculty trained to use and develop computational methods.