This study compared 40 female participants in a behavioral weight loss program who frequently reported cravings sweets to 40 who rarely reported craving sweets using 2-week behavioral eating diaries. The two groups were compared on physiological, demographic, and questionnaire measures and no significant differences were found. There were no significant differences in macronutrient intake either overall or in wide range of specific situations. The relative proportions of carbohydrate, protein, and fat consumed in association with craving sweets differed only slightly from the composition of other meals and snacks. Carbohydrate and protein intake when craving sweets was similar to breakfasts while the relative amount of fat consumed when craving sweets comparable to episodes of overeating. The two groups differed in their reporting of moods with the high-craving group reporting more boredom and less stress than the low-craving group. The relationship between situational andmood variables and reports of craving sweets did not differ between the two groups. Craving sweets was negatively associated with hunger and was not associated with meal skipping. A sequential analysis demonstrated that eating in response to craving sweets triggers an abstinence violation effect. These data are not consistent with the hypothesis that sweet craver's consume high-carbohydrate, low- protein meals and snacks in order to self-medicate depression caused by serotonin depletion. Instead, the data suggests that we should further explore the role of food palatability and food- related cognitions in order to understand cravings sweets. © 1993.