John Wiley & Sons, Ltd.

eLS subject area: Neuroscience

How to cite: Stice, Eric; and Burger, Kyle S (June 2012) Neurobiology of Overeating. In: eLS. John Wiley & Sons, Ltd: Chichester.

DOI: 10.1002/9780470015902.a0024012

Advanced article

Neurobiology of Overeating

Eric Stice, Oregon Research Institute, Eugene, Oregon, USA

Kyle S Burger, Oregon Research Institute, Eugene, Oregon, USA

The fact that some people are able to maintain a healthy weight, despite the omnipresence of high-fat and high-sugar foods, whereas others experience unhealthy weight gain, has prompted theories regarding individual difference factors that may increase risk for overeating. Some have proposed that individuals who experience less reward from food intake overeat to compensate for this reward deficit. Others have suggested that it is individuals who experience greater reward from food intake who are at risk for overeating. There is data from neuroimaging studies that investigate individual differences in response to food intake and food cues, as well as animal studies that support the reward deficit model, yet there is similar data to support the reward surfeit model. However, neither of these models fit with all of the data. We have therefore proposed a dynamic vulnerability model that appears to better account for the findings in the literature. This working model points to initial vulnerability factors as well as neural adaptations to overeating which may perpetuate overeating in a feed forward manner.

Key Concepts:

• Advances in neuroimaging techniques allows for the direct assessment of how food reward can impact eating behaviour.
• Multiple, conflicting theories of food reward and obesity have been presented.
• Emerging animal and prospective human data suggest that reduced reward when consuming foods and hyper-responsivity to food cues may evolve from overeating and perpetuate further overeating.
• Initial hyper-responsivity of brain regions that encode the reward value of food cues, coupled with greater responsivity of gustatory/oral somatosensory regions may be initial vulnerability factors of overeating.
• Reduced inhibitory activation in response to food stimuli, and genotypes associated with compromised dopamine function may increase risk for overeating/weight gain.
• Collectively, data point toward a dynamic vulnerability of overeating and obesity.

Nearly 70% of American adults are overweight or obese, defined as body mass index (BMI) >25 (Flegal et al., 2005). Excess adiposity is major risk factor for atherosclerotic cerebrovascular disease, coronary heart disease, cancer, hyperlipidemia, hypertension, and diabetes mellitus, resulting in as many as 300000 deaths in the USA each year (Luppino et al., 2010). Despite considerable efforts, most prevention programs do not significantly reduce risk for future weight gain and virtually all nonsurgical treatments result in only transient weight loss (Jeffery et al., 2000). The limited success of prevention and treatment interventions may be due to an incomplete understanding of the processes that increase risk for obesity. Theorists have suggested that reward from palatable food intake overrides the homoeostatic processes that historically balanced energy intake with energy expenditure, which has contributed to marked increase in obesity in western culture (e.g. Lutter and Nestler, 2009).

Brain imaging techniques, particularly functional magnetic resonance imaging (fMRI) and positron emission tomography (PET), allow the study of the neural basis of food reward in vivo, thereby providing valuable insight to eating behaviour, weight regulation, and the development of obesity. Frequently these studies focus on brain regions associated with dopamine (DA) functioning, such as the striatum (caudate and putamen), or brain regions that are within traditional dopaminergic pathways (e.g. mesolimbic or mesocortical pathways). As a result, neuroimaging studies have identified regions that respond to food receipt and encode the relative perceived pleasantness of foods; for example, consumption of palatable food results in increased activation of the striatum, dopaminergic midbrain, insula, subcallosal cingulate and prefrontal cortex, and these responses decrease as a function of satiety and declines in food pleasantness (Kringelbach et al., 2003). Further, the magnitude of DA release in the striatum correlates with ratings of meal pleasantness during consumption (Small et al., 2003). Data from these neuroimaging studies frequently point to individual differences in reward responsivity to intake of palatable food, anticipated receipt of palatable food, and food images, frequently dovetailing with traditional addiction literature. Herein we summarise theories regarding neurological risk factors that may cause overeating, review the available data, and propose a new working model that seems to better fit with extant findings.

Reward Deficit and Surfeit Theories of Obesity

Some theorists postulate that obese individuals show hypo-responsivity of reward circuitry, which leads them to repeatedly overeat to compensate for this deficiency (e.g. Wang et al., 2001). Consistent with the reward deficit and surfiet model, obese versus lean individuals have lower basal DA levels and dopamine D2 receptor availability (Volkow et al., 2008), implying that they show reduced DA receptor binding in reward circuitry. Further, obese versus lean adolescents show less activation in the dorsal striatum in response to consumption of chocolate milkshake versus tasteless solution in two studies (Stice et al., 2008a, b), though this finding did not replicate in another sample (Ng et al., 2011). Another study suggested that the reduced striatal response to palatable food receipt might be limited to individuals at genetic risk for reduced DA-signalling capacity by virtue of having a TaqIA A1 allele and not those with the A2/A2 genotype (Felsted et al., 2010).

Other theorists posit that individuals who experience greater reward from food intake are at risk for overeating (e.g. Dawe and Loxton, 2004). This is akin to the reinforcement sensitivity model of substance abuse, which posits that certain people show greater reactivity of brain reward systems to reinforcing drugs (Dawe and Loxton, 2004). Consistent with the reward surfeit theory, in response to pictures of palatable and energy-dense foods, obese versus lean humans show greater activation in the striatum, insula, orbitofrontal cortex (OFC), and amygdala (Stice et al., 2010b; Stoeckel et al., 2008; Rothemund et al., 2007), which are brain regions within dopaminergic reward pathways (Beaver et al., 2006). Further, in response to a cue predicting receipt of palatable food, obese versus lean humans show greater activation in the primary taste cortex (anterior insula, frontal operculum), which encodes tastes such as sweetness, and in oral somatosensory regions (postcentral gyrus, rolandic operculum; Ng et al., 2011; Stice et al., 2008b), which encode properties such as viscosity of the fat content and possibly energy density of the food (de Araujo and Rolls, 2004). Critically, young women who showed greater activation in response to cues for appetising versus unappetising food images in a region of OFC that encodes the reward value of stimuli, showed elevated future weight gain (Yokum et al., 2011).

Although there are findings that appear to provide support for both the reward deficit and the reward surfeit theories of obesity, these two theories seem fundamentally incompatible. One potential explanation for this mixed pattern of findings is that one set of abnormalities represents the initial vulnerability factors and the other set of abnormalities are a consequence of overeating or obesity (Stice et al., 2011). We now consider animal experiments and brain imaging studies, particularly prospective ones, that collectively suggest that a history of overeating may contribute to the development of some of the abnormalities characterised in the above reviewed cross-sectional studies, thereby suggesting a more dynamic model of reward responsivity and overeating.

Food Reward Neuroplasticity

Development of hypo-reward responsivity

Animal studies indicate that repeated intake of energy dense foods leads to down-regulation of D2 receptors and decreased D2 receptor density, sensitivity, and reduced reward sensitivity (Johnson and Kenny, 2010). These findings prompted us to test whether overeating contributes to reduced striatal responsivity to palatable food receipt; results from a prospective, repeated-measures fMRI study revealed that weight gain over a 6-month period was associated with reduced striatal response to palatable food receipt relative to both baseline and relative to women who showed a stable weight over the 6-month period (Figure 1; Stice et al., 2010a). These data provide the first evidence from humans that overeating may reduce reward-region responsivity to food receipt. Further, there is evidence that intake of energy dense food in particular contributes to this reduction in reward circuitry functioning and capacity. Specifically, rats fed a diet of only energy dense (high-fat, high-sugar) food relative to isocaloric intake of low energy dense food resulted in down regulation of striatal D1 and D2 receptors (Alsio et al.,...