News
Addiction as Learning Disorder
*In an article that has stimulated fevered Internet crosstalk among addiction professionals, Dr. Marc Lewis in the New England Journal of Medicine argues the case for a learning-theory-based model of addiction. Given the number of variables within this proposal, beginning with the definition of addiction, and proceeding through interpretation of the research in the context of who has or does not have addiction, this was a gutsy endeavor. It should be rewarded with reading the full article, which unfortunately is not yet open source; and as a review article, does not contain an abstract.
It is, however, consistent with his earlier works (e.g., The Biology of Desire: Why Addiction is Not a Disease 2015), and is thus not for him a new notion, but builds upon previous meditation and current neuroimaging research. Dr. Lewis is professor emeritus of psychology at the University of Toronto; as he acknowledges elsewhere**, his familiarity with addiction is also quite personal, and goes beyond bench or populational research. In the course of his review, he tabulates the biological and behavioral evidence in support of both disease and learning models. Some suppositions are less well-supported than others; for example the contention that most substance use disorders will ultimately remit (2 citations) is not quite the same as saying that they will remit without intervention. It is also the case that other diseases do remit over time; failure to remit is not a criterion for characterization of addiction as a disease.
The greatest risk when reading this carefully-reasoned, detailed overview of the topic is being lured into a realm of dichotomy. Many patterns of substance use conform to a learning model; one example is the reinforcement experienced by those with hangovers, whose symptoms are relieved by re-ingestion of the substance. For them, the message is elementary: “Don’t stop drinking,” or its variant, “The pain occurs when you aren’t drinking.” Conversely, it is difficult to accept operant conditioning alone as the foundation of the pattern experienced by the adolescent scion of alcoholics, whose dependence accelerates precipitously after first exposure. Both examples seem to call for a mixed model to account both for trajectories and predicting interventional success. This causes me to worry that much of any difficulty in conciliating the two views is definitional; how much of a disease must be biological for it to be a disease? At what point can panic disorder be said to reside solely within the locus ceruleus; and then, the follow-up question, do we always mean the same thing about the term, “panic disorder”? The article achieves one apparent intent, which is to challenge devotion to any single model of addiction development and progression.
* Marc Lewis, Ph.D., NEJM 18 OCT 2018: https://www.nejm.org/doi/full/10.1056/NEJMra1602872