Philosophy - Index > Testing - Index

Beyond Animal Research
By Jonathan Balcombe, Ph.D.

September 2005

Anorexia Nervosa and Animal Experiments

A childhood family friend developed anorexia nervosa as a teenager while studying at an elite ballet school. She nearly died. Anorexia is a terrible illness that afflicts some two million Americans, mostly young women.

Not surprisingly, there is a concerted effort to understand and address anorexia and other eating disorders. A search for "anorexia nervosa" on the NIH's PubMed online database yielded more than 8,600 studies. Most are human clinical studies--but not all. Many researchers are being funded to perform experiments on animals, usually (male) rats. But because rodents don't spontaneously develop eating disorders, experimenters must create "animal models" of the condition.

One such model is the "activity-based anorexia" (ABA) or "semi-starvation-induced hyperactivity" (SIH) model, which combines starvation with exercise. Experimenters place rats on a starvation diet and place an activity wheel in their shoebox cage. Perhaps because they are desperate to find food, starved rats show excessive use of the wheel, which accelerates weight loss. This mimics the excessive physical activity common among self-starving human patients.

Here are some recent examples of ABA model studies:

      In a study from the University Medical Center Utrecht, starved/exercised rats exercised less and ate less after leptin (a blood protein) was injected into their brains.1

      At Florida State University, rats were starved by restricting food access to two hours per day. Rats given running wheels developed ABA. During recovery (resumption of normal food access), food consumption rose and exercise dropped.2

      At the University of British Columbia, 200 rats were forced into a tight-fitting cylinder tube for one of three periods of time--20 minutes, two hours, or two hours per day for five days--to investigate the role of restraint stress on uptake of dietary fat, carbohydrate, and protein.3

      At Memorial University, Newfoundland, rats starved on the ABA model were confined to running wheels except during a 90-minute meal per day period. A control group (animals not confined to a running wheel) didn't lose weight, whereas the wheel group consumed less food, lost weight, and ran increasingly farther during the study.4

      In another Memorial University study, starved rats were kept in one of three different environments--a flat, circular alley; a running wheel; or a shoebox cage--to see if the alley also produces ABA (it didn't).5

These studies subject rats to the grinding misery of starvation while frustrating their frenetic efforts to seek and find food. And to what end? Anorexia is a complex syndrome, unique to humans, of primarily psychological origin. Trying to understand it by forcing rats to starve in their cages is rather like trying to understand suicide by giving a gun to a depressed guinea pig. My next column will explore more applicable, human clinical approaches to anorexia nervosa.

Jonathan Balcombe, Ph.D., is a PCRM Research Scientist with a background in ethology. He is the author of The Use of Animals in Higher Education, as well as many scientific papers on humane life science education and animal behavior. His recent scientific review showing that animal experiments are more stressful than previously understood was published in Contemporary Topics in Laboratory Animal Science, and a forthcoming review in Laboratory Animals, reveals how laboratory housing thwarts rodent's behavioral needs. His new book, Pleasurable Kingdom: Animals and the Nature of Feeling Good, is due out in May 2006.


1.Hillebrand JJ, Koeners MP, de Rijke CE, et al. Leptin treatment in activity-based anorexia. Biol Psychiatry. 2005 Jul 15;58(2):165-71.

2.Dixon DP, Ackert AM, Eckel LA. Development of, and recovery from, activity-based anorexia in female rats. Physiol Behav. 2003 Nov;80(2-3):273-9.

3.Wang SW. Effects of restraint stress and serotonin on macronutrient selection: a rat model of stress-induced anorexia. Eat Weight Disord. 2002 Mar;7(1):23-31.

4.Lett BT, Grant VL, Smith JF, et al. Preadaptation to the feeding schedule does not eliminate activity-based anorexia in rats. Q J Exp Psychol B. 2001 Aug;54(3):193-9.

5.Koh MT, Lett BT, Grant VL. Activity in the circular alley does not produce the activity anorexia syndrome in rats. Appetite. 2000 Apr;34(2):153-9.

Fair Use Notice and Disclaimer
Send questions or comments about this web site to Ann Berlin,