Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)




Herman & Mack (1975) theorized that "restrained eaters," conflicted between social pressure to be thin and biological pressure to be fat, tend to alternately eat very little or a great deal, as they respond to one or the other constraint. According to restraint theory, restrained eaters' chronic dieting induces physiological and psychological states that make them highly susceptible to external disruption of eating controls. Research has shown that when led to believe they have already overeaten (i.e., when they are "preloaded"), restrained eaters will loosen restraints and "counterregulate" (i.e., binge-eat). In contrast, "unrestrained eaters" (i.e., normal eaters) will compensate by subsequently eating less under such conditions. However, while this effect has been shown in normal weight restrained subjects, overweight restrained subjects have not reliably counterregulated. Consequently, questions can be raised as to restraint theory's ability to predict eating behavior of overweight individuals. One study utilizing a private setting found counterregulation in preloaded normal weight and overweight restrained eaters. However, because a no-preload group was not included in this study, it could not be determined whether the preload or the private setting was responsible for the counterregulatory eating.

In the present study, 113 female subjects were told they were participating in a sensory experiment. Normal weight and overweight subjects, who were low restraint or high restraint, either consumed a "high calorie" milkshake as a preload or received no preload. Subjects were subsequently asked to "taste-test" ice cream flavors. Using subtle situational cues, subjects were led to believe the amount of their ice cream consumption would not be easily detected by experimenters. As expected, low restraint-normal weight subjects compensated for a preload by eating less ice cream; and low restraint-overweight subjects ate the same amount regardless of preloading. However, both normal weight- and overweight-high restraint subjects failed to counterregulate after preloading.

The author discusses implications of this and previous findings. It is argued that the restraint dimension may reflect more a cognitive style than a behavioral style. It is suggested that disordered eating may be more productively studied under naturalistic conditions, or longitudinally, than by taking a single measurement in a laboratory situation.

Included in

Psychology Commons