Previously, studies have explored the relationship between dietary behavior and salivary reactivity to food. Despite this, it remains unclear which behaviors are associated with enhanced reactivity. One problem is that measures of behavior have not been compared directly. In particular, it is unclear whether elevated reactivity is associated with measures of dietary restraint or with measures of failed dietary control and a tendency to overeat. To address this problem, we compared the association between salivary reactivity and scores on the subscales of the Three-Factor Eating Questionnaire (restraint, disinhibition, and hunger). Estimates of reactivity were derived front the difference between a baseline saliva measure and a similar measure taken in close proximity to hot pizza. Our second aim was to explore how salivary reactivity changes after a meal. Female participants (N=40) were tested before and after a lunch (cheese sandwiches). All tended to show reactivity to pizza before but not after lunch. No significant differences were associated with the disinhibition or hunger subscales. However, prelunch reactivity was significantly greater in those participants with high scores on the restraint scale. This does not appear to be related to reported levels of hunger before lunch. Rather, it may reveal an intrinsic difference between the reaction of restrained and unrestrained eaters to food. (C) 2004 Elsevier Inc. All tights reserved.