Adolescence is a critical period for establishing weight-related behaviors that track into adulthood. Rapid physical, psychosocial, and neurological development during this time often favors reward responses, which are not always counteracted by a less developed inhibitory control system. Increased autonomy over eating choices may lead to poor eating habits and weight gain, increasing the risk of adult obesity. Understanding contributing factors is crucial for identifying intervention targets. One such factor is the relative reinforcing value (RRV) of food, an index of motivation to obtain food. Higher RRV is associated with greater BMI and weight gain. This relationship is moderated by delay discounting (DD), the preference for smaller, immediate rewards over larger, delayed ones. High RRV and high DD, known as reinforcement pathology, are associated with higher BMI, weight gain, and resistance to weight loss. Adolescence is an understudied period regarding reinforcement pathology. While RRV is considered a trait, it can be modified by state (hunger), alternatives (e.g., television), or experience (food intake variety). Repeated intake of highly liked snack foods can decrease RRV in non-obese adults but increase it in obese adults. This increase is termed sensitization, a phenomenon described in substance use literature, responsible for neurobiological changes in substance abuse. Sensitization of the RRV of HED foods is associated with greater BMI and weight gain in adults and is cross-sectionally linked to higher zBMI in adolescents. This study focused on the potential protective effects of sensitization to repeated LED food intake. While sensitization is more likely with HED foods due to their dopamine-releasing properties, some individuals might sensitize to LED foods, offering potential protection against weight gain. Prior studies showed low rates of LED food sensitization and no relationship with weight. This study aimed to examine the impact of sensitization to repeated HED and LED food intake on zBMI change over two years in a large adolescent cohort. The hypothesis was that HED food sensitization is positively associated, LED food sensitization is negatively associated, and an interaction exists between HED and LED sensitization, impacting zBMI change. Furthermore, the study hypothesized that the relationship between HED sensitization and zBMI gain is moderated by DD.

The literature review extensively explores the established link between food reinforcement, delay discounting, and obesity across the lifespan. It highlights the importance of adolescence as a critical period for establishing long-term weight-related behaviors and the role of relative reinforcing value (RRV) of food as a behavioral phenotype associated with obesity risk. Studies demonstrating the cross-sectional and prospective associations between RRV, especially for high-energy-density (HED) foods, and BMI/weight gain are reviewed. The moderating role of delay discounting (DD) in these relationships is discussed, emphasizing the concept of reinforcement pathology. The literature also examines the influence of repeated food intake on RRV, introducing the concept of sensitization and its parallels in substance use research. Previous studies on the sensitization to HED and low-energy-density (LED) foods in both adults and adolescents are summarized, laying the groundwork for the current investigation's hypotheses.

This 2-year prospective observational cohort study involved 201 adolescents (aged 12–14 at baseline) with a baseline zBMI between −1.5 and 2.0. Participants completed five laboratory visits over 5–6 weeks. The RRV of HED and LED foods and DD were assessed at baseline (visits 1, 2, and 4). RRV was reassessed after participants consumed daily portions of HED and LED foods for 2 weeks (visits 3 and 5; order counterbalanced). Increases in RRV after daily intake were categorized as 'sensitization', while decreases were 'satiation'. Follow-up visits were conducted at 6, 15, and 24 months to measure height, weight, and zBMI. The RRV task involved choosing between responding for food or a preferred sedentary activity. Sensitization was calculated by subtracting baseline RRV from post-daily intake RRV. DD was assessed using an adjusting amount task. Demographic data, pubertal development, subjective food responses (hunger, thirst, liking, wanting), and dietary restraint (DEBQ) were also collected. Multilevel modeling was used to analyze zBMI changes over time, with month of visit, sex, DEBQ score, pubertal development, and baseline zBMI as fixed effects. Participants were categorized as sensitizers (>1) or satiators (≤1) based on their change in RRV. Models examined the effects of HED and LED sensitization, their interaction, and the moderating role of DD on the relationship between HED sensitization and zBMI change.

The study's primary finding is that sensitization to HED food is significantly associated with greater zBMI change over time (β = 0.0074, p = 0.033). This effect was observed despite controlling for baseline zBMI, pubertal development, and other factors. Specifically, adolescents who exhibited sensitization to HED foods experienced a greater increase in zBMI over the two-year study period compared to those who did not show this sensitization. Visual representations of zBMI trajectories for sensitizers and satiators further support this finding. Conversely, no significant association was found between sensitization to LED food and zBMI change. There was no interaction effect between HED and LED sensitization on zBMI change, suggesting that HED sensitization is the primary driver of weight gain regardless of the response to LED foods. Delay discounting did not moderate the relationship between HED sensitization and zBMI change; there were no significant main effects or interactions involving DD in the models.

The findings support incentive sensitization theory, showing that sensitization to repeated HED food intake is a prospective predictor of increased zBMI gain in adolescents without obesity. This replicates and extends previous findings from adult populations. The lack of association between LED sensitization and zBMI change, along with the absence of an interaction effect between HED and LED sensitization, indicates that HED sensitization is the main risk factor. The lack of moderation by DD may reflect the dominance of motivational systems over inhibitory control during adolescence. The study's results highlight sensitization to HED foods as a potentially important target for weight gain prevention interventions, suggesting that focusing on reducing this sensitization might be more effective than addressing LED sensitization or DD.

This study confirms that sensitization to the reinforcing value of high-energy-density foods is a significant behavioral risk factor for increased weight gain in adolescents. The lack of a protective effect from LED sensitization and the absence of a moderating effect from delay discounting highlight HED sensitization as a key target for future weight gain prevention interventions. Future research should explore strategies to reduce this sensitization, potentially through behavioral interventions that focus on altering responses to rewarding foods.

The study's limitations include the sample's predominantly white and upper-middle-class composition, which limits the generalizability of findings. The age range at study entry resulted in differing pubertal development stages between boys and girls. The imbalanced sample size between sensitizers and satiators reduced statistical power. The COVID-19 pandemic affected follow-up timing, though the study maintained a relatively high retention rate.