How is reward sensitivity related to bodyweight in children?

Abstract

Previous research assumes that there are two seemingly opposing hypotheses for the relation between reward sensitivity (RS) and bodyweight: hyper-responsiveness model and Reward Deficiency Syndrome (RDS), leading to the proposition of a feed forward process of weight gain. High RS may contribute to overeating and weight-gain among normal weight individuals. Over time the excessive food-intake may evolve in a down-regulation of dopamine (RDS), resulting in overeating as a form of self-medication and the progression to obesity. This process was evidenced in adults showing a curvi-linear relationship between self-reported RS and BMI. The aim of the current study was to investigate the association between self-reported RS and BMI in children (10–15 years). The results confirm the non-linear relationship between RS and bodyweight and support the suggestion of the same feed forward process in children. These findings imply that it is crucial to reduce the intake of high palatable foods in high RS children to prevent the decrease in RS and reduce the risk for future weight gain.

Introduction

The recent boom of childhood obesity challenges worldwide public health (Orsi, Hale, & Lynch, 2011). Besides the recognized genetic predisposition, the impact of the modern food–environment is well-established. A greater food palatability, a wide variety of foods, the high and easy availability (in the home and workplace), the stimulation by advertising, the food saliency, a larger portion size, and a higher energy density of food, all contribute to an increased reward value of foods, which overrides existing satiety signals and fosters overeating (Rolls, 2011). Consequently, in an increasing part of western society a positive energy balance is likely, which could lead to weight gain. Although, the fact that not everyone in the same high rewarding food–environment becomes overweight points also at the role of interacting individual factors. It is reasonable to propose that individual differences in reward sensitivity (RS) or the tendency to engage in motivated approach behavior in the presence of rewarding stimuli may be one of the factors that contribute to a vulnerability to overeat and become obese (Small, 2009).

Obese individuals find palatable foods more rewarding than non-obese (McGloin et al., 2002, Rissanen et al., 2002), but it remains unclear why this is so (Lowe, Van Steenburgh, Ochner, & Coletta, 2009). According to Gray’s Reinforcement Sensitivity Theory (RST, Gray, 1994), RS reflects functional outcomes of the behavioral activation system (BAS), which is organized primarily by the neurotransmitter dopamine (DA) (Di Chiara, 1995, Pickering and Gray, 1999). Additionally, it has been postulated that dopamine deregulation contributes to the development of obesity and binge eating (Davis et al., 2009, Mathes et al., 2010). The Dual Vulnerability Theory of dopamine deregulation presents two opposing hypotheses (Davis and Fox, 2008, Stice et al., 2009). The first hypothesis, the hyper-responsiveness model, states that hypersensitivity to reward due to increased dopaminergic functioning, may motivate individuals to seek rewarding stimuli simply because the reinforcement value of the reward is so great (Davis et al., 2004, Dawe and Loxton, 2004). Alternatively, Reward Deficiency Syndrome (RDS), states that individuals with relative insensitivity to reward because of low dopaminergic functioning, seek more rewarding substances to increase endogenous dopamine levels and improve mood (Volkow, Wang, Fowler, & Telang, 2008).

The hyper-responsiveness model was supported by experimental research in healthy volunteers. In a large group of adults who habitually consume a high-fat diet, Blundell et al. (2005) identified individual differences which make some people susceptible to weight gain and others resistant. The results clearly indicate that hedonic attraction to palatable foods and eating could significantly differentiate between individuals who gained weight and those who remained lean. This finding was further supported in children by Guerrieri, Nederkoorn, and Jansen (2008). They identified high and low RS-children based on their performance on a behavioral task and measured their caloric intake via a Bogus Taste Test. Interestingly, when varied food was offered, the high RS children ingested significantly more calories than their low RS counterparts. When monotonous food was offered, RS did not really affect caloric intake.

Similarly imaging research in adults found that RS as measured with the BIS/BAS self-report scale (Carver & White, 1994), significantly predicted activation to appetizing foods (relative to bland foods) in brain areas implicated in food reward (Beaver et al., 2006, Schienle et al., 2009). Additionally, fMRI data indicate that obese children and adolescents versus their lean counterparts showed greater activation in brain reward areas in response to visual food stimuli (Batterink et al., 2010, Bruce et al., 2010) and in response to food consumption (Stice, Spoor, Bohan, & Small, 2008). Especially relevant is the fact that it was previously shown in mice that activation of these brain areas produces overeating and increases the preference for foods high in fat and sugar (Kelley, 2004).

The assumed initial vulnerability may be a generalized hyper-responsiveness to various reward types as opposed to a specific deficit within the eating domain (Stice, Yokum, Burger, Epstein, & Small, 2011). Stice et al. (2011) found that adolescents at high-risk versus low-risk for future obesity by virtue of parental obesity not only showed greater activation in reward regions in response to palatable food, but also to monetary reward. Similarly, compared to lean individuals, obese children continue to play a rewarded computer game longer (Nederkoorn et al., 2006, Verbeken et al., 2009) and report higher generalized RS (Davis et al., 2004, Mobbs et al., 2010). Furthermore, substantial longitudinal research indicates that children with higher generalized RS (measured with a self-regulation task) were more likely to be classified as overweight or obese several years later (Francis and Susman, 2009, Graziano et al., 2009, Seeyave et al., 2009). However the paradigm used in these longitudinal studies provides a mixed measure of RS and inhibitory control, and therefore it is impossible to support the unique predictive value of RS.

The second hypothesis, here labeled as Reward Deficiency Syndrome (RDS) rests on the premise that palatable food can be used in the same manner as addictive drugs, and that risk for its overuse should therefore be greater among those at the anhedonic end of the RS continuum. In other words decreased dopamine activity reduces the sensitivity to natural rewards, and this deficit might temporally be compensated for by overeating (Davis et al., 2004). In obese children, imaging data indeed show a lower activation of a part of the dopaminergic reward system in response to food cues (Davids et al., 2010) and to food receipt compared to normal-weight children (Stice et al., 2008). These findings suggest that food may be experienced as less rewarding by obese children.To our knowledge, there is until now no evidence for this model based on behavioral measures or self-report data in children. Few studies in adults and adolescents found evidence for the relation between reduced self-reported RS and uncontrolled eating, emotional eating, binge eating, and obesity (Davis and Fox, 2008, Davis et al., 2004, Keränen et al., 2010, Pagoto et al., 2006).

These seemingly opposing data might reflect a dynamic vulnerability (DV) model for obesity that may evolve and change over time in response to overeating (Stice et al., 2011). The DV-model states that it is possible that heightened generalized RS is an initial risk factor for excessive food intake among normal weight individuals resulting in a positive energy balance and weight gain. However, the excessive food-intake can overload the DA system in such a way that it reduces the DA activity. An adaptive decrease in dopamine D2 receptor (D2R) is suggested to contribute to the reduced responsiveness of the striatum to palatable food (for the biochemical mechanisms see review Kenny, 2011). Hence, excessive overeating is assumed in the long run to lead to an insensitive reward system, which enhances further overeating to reach an acceptable level of hedonic satisfaction (Davis et al., 2004, Lowe et al., 2009, Stice et al., 2011).

This DV-model was already evidenced in adults showing a curvi-linear relationship between BMI and RS, based on self-report (Davis & Fox, 2008), but has never been examined in children. Such knowledge seems however pivotal in unraveling differential mechanisms leading to overeating but also in tailoring early intervention.

The current study aimed to investigate the DV model in children by analyzing the association between self-reported RS and bodyweight in children. From the age of 10 years, it seems relevant to assess RS via self-report. Based on the findings of Davis and Fox (2008), we expect a positive association between self-reported RS and bodyweight, which will change to a negative association among children with obesity.