Eating behaviour can be driven by non-homeostatic factors like stress. Both increased and decreased food intake in response to stress has been documented, but it has remained difficult to identify a trait that predicts who shows either pattern. Thus, we collected naturalistic data from Ecological Momentary Assessment in combination with the trait-level Salzburg Stress Eating Scale (SSES). In study 1, 97 individuals completed the SSES and 6 daily reports about stress, food craving and perceived food intake across 8 days, whereas in study 2, 83 diet-interested participants completed the same measures at 4 daily prompts across 14 days. Consistent across both studies, multilevel modelling revealed that participants with high SSES-scores showed relatively more positive intra-day stress-craving relationships than those with low SSESscores. On the day level, stress also predicted perceived food intake as a function of SSES-scores. Controlling for negative affect did not alter results. Results support an individual difference model of stress-eating where decrease vs increase of eating depends on SSES-scores. In affected individuals stress influences simultaneous food craving but might exhibit cumulative or delayed effects on food intake. Furthermore, the SSES provides a valid instrument for identifying at risk individuals and for tailoring interventions.

Background Many people aim to eat healthily. Yet, affluent food environments encourage consumption of energy dense and nutrient-poor foods, making it difficult to accomplish individual goals such as maintaining a healthy diet and weight. Moreover, goal-congruent eating might be influenced by affects, stress and intense food cravings and might also impinge on these in turn. Directionality and interrelations of these variables are currently unclear, which impedes targeted intervention. Psychological network models offer an exploratory approach that might be helpful to identify unique associations between numerous variables as well as their directionality when based on longitudinal time-series data. Methods Across 14 days, 84 diet-interested participants (age range: 18–38 years, 85.7% female, mostly recruited via universities) reported their momentary states as well as retrospective eating episodes four times a day. We used multilevel vector autoregressive network models based on ecological momentary assessment data of momentary affects, perceived stress and stress coping, hunger, food craving as well as goal-congruent eating behaviour. Results Neither of the momentary measures of stress (experience of stress or stress coping), momentary affects or craving uniquely predicted goal-congruent eating. Yet, temporal effects indicated that higher anticipated stress coping predicted subsequent goal-congruent eating. Thus, the more confident participants were in their coping with upcoming challenges, the more they ate in line with their goals. Conclusion Most eating behaviour interventions focus on hunger and craving alongside negative and positive affect, thereby overlooking additional important variables like stress coping. Furthermore, self-regulation of eating behaviours seems to be represented by how much someone perceives a particular eating episode as matching their individual eating goal. To conclude, stress coping might be a potential novel intervention target for eating related Just-In-Time Adaptive Interventions in the context of intensive longitudinal assessment.

Introduction: Coronavirus 2019 (COVID-19) quickly evolved into a global pandemic in early 2020, and most countries enforced social confinements to reduce transmission. This seems to dovetail with increasing, potentially problematic, screen use habits, such as gaming and “binge-watching.” Yet, the subjective experience of the common confinements may vary not only between individuals depending on age, sex, and living conditions (i.e., living alone) but also within individuals from day to day: confinements might interfere with habitual activity schedules more strongly on some days than on others. Such dynamic confinement experience has not been studied in relation to screen use yet but might guide targeted intervention.Method: In total, 102 participants (n = 83 female, n = 80 university students) completed 14 days of ecological momentary assessment during a COVID-19-related lockdown in Germany and Austria. Each evening, they indicated the extent to which they felt restricted by confinements in their social and work lives and whether they engaged in unusually high and intense levels of television watching, social media use, news consumption, internet surfing, and gaming. They also reported on how much they experienced their day to be structured.Results: Experienced work confinements were positively associated with social media usage. Further, work confinements were positively associated with gaming in males and with news consumption, especially in individuals living alone. Social confinements were positively associated with watching television especially in younger participants and with social media consumption in younger participants. Higher experienced day structure was related to less television watching, gaming, and internet surfing but more news consumption.Discussion: Screen use behaviors increased with higher confinements within person, dependent on sex, age, and living situation. Such knowledge allows tailoring on the person level (who should be addressed?) and the time level (when should interventions be scheduled?) as the negative consequences of excessive screen use behaviors on mental and physical health are well-documented. One potential low-threshold intervention might be day-structuring.

Background Prevention of binge eating through just-in-time mobile interventions requires the prediction of respective high-risk times, for example, through preceding affective states or associated contexts. However, these factors and states are highly idiographic; thus, prediction models based on averages across individuals often fail. Objective We developed an idiographic, within-individual binge-eating prediction approach based on ecological momentary assessment (EMA) data. Methods We first derived a novel EMA-item set that covers a broad set of potential idiographic binge-eating antecedents from literature and an eating disorder focus group (n=11). The final EMA-item set (6 prompts per day for 14 days) was assessed in female patients with bulimia nervosa or binge-eating disorder. We used a correlation-based machine learning approach (Best Items Scale that is Cross-validated, Unit-weighted, Informative, and Transparent) to select parsimonious, idiographic item subsets and predict binge-eating occurrence from EMA data (32 items assessing antecedent contextual and affective states and 12 time-derived predictors). Results On average 67.3 (SD 13.4; range 43-84) EMA observations were analyzed within participants (n=13). The derived item subsets predicted binge-eating episodes with high accuracy on average (mean area under the curve 0.80, SD 0.15; mean 95% CI 0.63-0.95; mean specificity 0.87, SD 0.08; mean sensitivity 0.79, SD 0.19; mean maximum reliability of rD 0.40, SD 0.13; and mean rCV 0.13, SD 0.31). Across patients, highly heterogeneous predictor sets of varying sizes (mean 7.31, SD 1.49; range 5-9 predictors) were chosen for the respective best prediction models. Conclusions Predicting binge-eating episodes from psychological and contextual states seems feasible and accurate, but the predictor sets are highly idiographic. This has practical implications for mobile health and just-in-time adaptive interventions. Furthermore, current theories around binge eating need to account for this high between-person variability and broaden the scope of potential antecedent factors. Ultimately, a radical shift from purely nomothetic models to idiographic prediction models and theories is required.

Food craving (FC) peaks are highly context-dependent and variable. Accurate prediction of FC might help preventing disadvantageous eating behavior. Here, we examine whether data from 2 weeks of ecological momentary assessment (EMA) questionnaires on stress and emotions (active EMA, aEMA) alongside temporal features and smartphone sensor data (passive EMA, pEMA) are able to predict FCs ~2.5 h into the future in N = 46 individuals. A logistic prediction approach with feature dimension reduction via Best Item Scale that is Cross-Validated, Weighted, Informative and Transparent (BISCWIT) was performed. While overall prediction accuracy was acceptable, passive sensing data alone was equally predictive to psychometric data. The frequency of which single predictors were considered for a model was rather balanced, indicating that aEMA and pEMA models were fully idiosyncratic.