Understanding food cravings and consumption: From group averages towards a personalized approach

Overweight and obesity are major global issues that contribute to significant healthcare expenditures on a societal level. Despite the efforts against it, the rise of overweight and obesity prevalence is not stopping, indicating a need for innovative approaches. This proposed project aims to address this issue by going beyond the current state-of-the-art using a multimethod approach, together with advanced analysis, to address the large heterogeneity among individuals experiencing food cravings and problems with controlling their intake. The project aims to address three interrelated problems related to overweight and obesity. First, research and public policies have primarily focused on the environment, such as marketing and pricing of food, and largely ignore the psychology of eating. To address this, the project's first objective is to gather data on personality, neurocognitive and emotional processes that contribute to increased food cue reactivity, craving and overconsumption. Second, while many studies aim to understand the neurophysiological correlates of (un)successful food cue reactivity regulation, little is known about the ways in which neural correlates of food cue reactivity regulation correspond to food-related outcomes in everyday life. Thus, the second objective is to conduct a study combining EEG, self-reports, and real-time experience sampling, following participants for several days in their own environment. Third, the ‘one-size-fits-all’ approach to understanding food cravings and consumption is not effective. Instead, this project will follow a more personalized approach by studying a large sample of young adults with varying weight (third objective). With the use of unsupervised machine learning methods, the project will aim to identify personalized profiles (clusters) of participants based on neurocognitive markers, self-reports, and experience sampling data. The project will use the so-called ‘fingerprinting approach’, which addresses the heterogeneity among individuals belonging to the same group (e.g., overweight, normal-weight participants). More specifically, individuals will be clustered and re-assigned to more homogenous data-driven clusters. This will allow us to create a summary measure that identifies which critical data are necessary to reliably identify which profile of food-cue reactivity and craving behaviour the individual belongs to. Later in the project, the usefulness of the identified clusters will be examined with the use of a longitudinal approach, where participants will be followed-up over a period of six months. These aims will be achieved with several different studies, all conducted with the same participants. This will allow us to profile each individual using several different methods and gain a better understanding of the underlying processes that contribute to craving and consumption for that individual. We expect to uncover several different clusters of individuals driven by different processes related to food consumption and craving. This is an important contribution to this area of research, since obesity research tends to be based on averages and neglects individual aspects that contribute to weight gain and food-related issues. It is crucial to recognize the marked heterogeneity of individuals that seem to belong to the same group based on observed characteristics (e.g., weight) but are in reality driven by completely different processes. The only way of improving the treatment of weight-related issues in the long run and, more importantly, contributing to possibly preventing the beginning of these issues in the first place, is to acknowledge this marked variability between individuals. Before personalized interventions can be developed, a deep understanding of the underlying factors that drive craving and consumption is needed. The proposed project undertakes an important initial step in this direction.