Stress and emotions predict naturalistic variations in food craving: the role of interindividual difference measures in N=764 participants

The validity of the concept of emotional eating and the related concept of stress eating - understood as the general tendency to eat in response to intense affective experience – is under intense debate. For one, theory and evidence point to interindividual differences in this tendency, typically assessed with self-report questionnaires. Furthermore, overt eating underlies manifold contextual conditions, pointing to the value of studying how emotions relate to food craving, an important precursor to food intake that can be triggered rapidly and involuntarily. Last, temporal ordering of affect on the one hand and craving on the other under natural conditions seems key to its understanding. To study such temporal (concurrent, prospective) relations, we analyzed intensive longitudinal data obtained through ecological momentary assessment during which 764 participants pooled across eight different primary studies answered questions 4-6 times throughout the day. To capture interindividual differences, trait-level eating style questionnaire scores were modeled as moderators for potential emotion/stress - craving relationships. Results obtained at the same timepoint (concurrent associations) showed that feeling bored, irritated, or stressed (primarily negatively toned affect) but also cheerful or enthusiastic (positive) co-occurred with food craving while relaxation co-occurred with less craving. These relationships were moderated by a range of eating style questionnaires. Feeling bored, stressed, or less relaxed were also associated with later craving which, in turn, related to feeling calmer, more relaxed, and less stressed at the next time point. Eating style questionnaires moderated only few of these prospective craving-emotion relationships. Results generally support the validity of the interindividual difference model of emotional eating and the stress eating concepts in regard to food craving in naturalistic environments. Yet, concurrent affect-craving relationships seem more robust and congruent with respective trait questionnaires than prospective relationships. We suggest novel terminology to reflect these findings. Further, questions regarding temporal ordering and causation remain open.