“A Theory of Activity-Based Anorexia”, W. Frank Epling, W. David Pierce, Larry Stefan1983-09 ()⁠:

The present paper documents the etiological importance of physical activity to self-starvation in animals and suggests similarities between this research area and the literature concerned with some self-starvation in humans.

An activity anorexia is proposed that may account for 38% to 75% of anorexia nervosa. An account of excessive locomotor activity is made in terms of schedule-induced behavior. A reciprocally interactive effect of activity and food ingestion is taken to explain self-starvation for animals and activity anorexia in humans.

Literature is reviewed which demonstrates that rats and mice self-starve when they are given access to a running wheel and placed on food restriction. In this paradigm, these animals become excessively active and paradoxically reduce food consumption when compared with control subjects. This evidence and related findings are shown to be consistent with a phylogenetically based model of anorexia.

Sociocultural factors are hypothesized to set and maintain the conditions that produce activity anorexia in humans.