Today, I am pleased to have a guest post from the fantastic Dr. Becca, author of the blog 'Fumbling Toward Tenure'. Enjoy!

Last week, the New York Times’ “Well” section ran a piece titled, “How Exercise Can Prime the Brain for Addiction.” Scary, right? One minute you’re cruising along on the treadmill, and next thing you know, you’re ADDICTED TO COCAINE. Hovering over the web page tab header, however, reveals what may have been the original title—the more qualified, but less provocative “How Exercise May Make Addictions Better, or Worse.”


Ironically, it’s the cutting-room-floor version of the title that more accurately (but only marginally so) reflects the findings of Mustroph et al (2012), an Illinois-based research group who studied the influence of exercise on the learning processes associated with drug use. In a nutshell, the researchers showed that the timing of exercise and drug exposure mattered: animals that exercised after getting a few injections of cocaine had an easier time “letting go” of their drug-associated cues than animals that exercised before cocaine exposure did. What Mustroph et al were not studying, though, was addiction—and this is only the beginning of where NYT writer Gretchen Reynolds does a disappointingly poor job of science reporting.

This paper is about learning. With every experience we have, we learn something about the circumstances in which that experience occurred, and experience with drugs is no different. If you always do drugs in a certain room of your house, or at one particular club, you’re going to start associating those places with the drug, and, in all likelihood, with the way the drug makes you feel. You might even enjoy hanging out in those places when you’re not using the drug, because of the positive associations you’ve formed.

This is the idea behind Conditioned Place Preference (CPP), a common test of context-drug associations in animals. An animal that experiences cocaine in one environment will choose to spend time in that environment later on, even if its system is drug free at the time. But as Reynolds describes it, “If a rodent returns to and stubbornly plants itself in a particular place where it has received a drug or other pleasurable experience, then the researchers conclude that the animal has become habituated.” She goes on, “All of the mice had, essentially, become addicts.”

A couple of things are wrong here. 1) Why are we calling the mice “stubborn?” This anthropomorphizing is not actually reflective of cocaine-treated animals’ behavior in CPP, and is totally unnecessary for the purposes of the article. 2) “Habituated?” I do not think it means what you think it means. In behavioral neuroscience, “habituation” usually refers to the way in which an animal can get used to a certain environment—much like how you sleep better after a couple of nights in a new home. Here Reynolds appears to be confusing the process of acclimating the mice to the testing environment before the experiment begins with that of the actual testing. 3) What a preference for the location previously associated with cocaine demonstrates is exactly that—the mouse understands the association. That mouse is, in no possible definition of the word, an “addict.”

So then, what does it mean to be addicted to a drug? The DSM-IV criteria are easy enough to consult, and I can’t anywhere in there find the part where having four injections of cocaine makes you an addict. And yet, Reynolds refers to “cocaine-addicted mice” or the animals’ “addiction” over and over again throughout the article, when she just as easily could have said “cocaine-treated mice” or something similarly more accurate.

There are other inaccuracies, too; most egregious in my mind is a reference to the process of extinction as “forgetting.” In extinction, repeated exposure to the cues or contexts in the absence of the drug teaches the animal that it can’t expect drug in that environment anymore, and the animal will stop showing place preference. But as any recovering drug abuser will tell you (and as can be demonstrated in animals as well), the memories for a drug and its related cues are never forgotten—this is what makes relapses, or “falling off the wagon” so common. Instead, the animal learns that the predictive value of the environment has changed, and its behavior changes accordingly. This is a critical point of the study itself—the primary difference between experimental groups is in the way they to extinguish—and yet the significance of this finding gets lost in translation.

Now, why am I getting my panties in a twist over a couple of misappropriated scientific terms? Because people are smarter than this! Without interviewing both Ms. Reynolds and the study’s primary authors, it’s hard to know whether Reynolds did a bad job interpreting the science, or if the authors or public relations office “dumbed it down” for the journalist in a way that misrepresented their actual work (or a muddy combination of the two). But neither scientists nor science writers should shy away from telling the public what science is really about—and what the true implications of a given study may be, however less exciting they seem.

Do researchers want the public to find their work interesting? Yes! Do writers want people to read their articles? Naturally! But the people who read this article may have come away with the impression that exercising could make them more likely to become a cocaine addict, which is not at all the case. When scientific findings are overstated, oversimplified, or misinterpreted, neither the public nor scientists benefit.

Again, this paper was not about addiction, but learning—a distinction I’m willing to bet the average newspaper reader is able to make. So why was that so hard for the NYT to convey?

Mustroph, M., Stobaugh, D., Miller, D., DeYoung, E., & Rhodes, J. (2011). Wheel running can accelerate or delay extinction of conditioned place preference for cocaine in male C57BL/6J mice, depending on timing of wheel access European Journal of Neuroscience, 34 (7), 1161-1169 DOI: 10.1111/j.1460-9568.2011.07828.x