The Chocolate Study

Obese vs lean humans show greater gustatory/oral somatosensory and reward region responsivity to palatable food images/cues and this predicts future weight gain (Yokum et al., 2011; Stice et al., 2008, 2010b; Stoeckel et al., 2008), in line with reward surfeit and incentive sensitization models of obesity (Berridge, 2009; Davis et al., 2004). Yet, obese vs lean humans have fewer dopamine (DA) receptors in striatal reward regions, show reduced striatal response to palatable food intake, and low striatal response predicts future weight gain in those at genetic risk for reduced DA signaling (Felsted et al., 2010; Stice et al., 2008; Wang et al., 2001; Volkow et al., 2008), in line with the reward deficit model of obesity (Wang et al., 2002b). One explanation for the mixed findings is that some of these findings reflect initial risk factors and others result from overeating. Firing of DA neurons in reward regions shifts from food intake to cues that predict food intake after conditioning (Kiyatkin et al., 1994; Schultz et al., 1993) and overeating leads to reduced D2 receptor density, D2 sensitivity, and reward sensitivity in rats (Alsio et al., 2010; Kelley et al., 2003; Johnson \& Kenny, 2010) and striatal response to food in humans (Stice et al., 2010a), implying that overeating leads to increased incentive sensitization and down-regulation of reward regions. Further, reduced inhibitory region response to food images/cues predicts future overeating and weight gain (Cornier et al., 2010). Data imply that youth at risk for obesity initially show greater responsivity of regions that encode the reward value of food cues, coupled with greater responsivity of gustatory/oral somatosensory regions that encode the sugar and fat content of foods, and with reduced inhibitory region responsivity, which lead to overeating/weight gain that produces blunted striatal DA signaling, increased responsivity of reward valuation regions to food cues, and reduced inhibitory activation in response to food stimuli, increasing risk for further overeating/weight gain. We propose to conduct a rigorous test of this dynamic-vulnerability model using a novel repeated measures fMRI design in which teens complete scans annually over 4 years. Aim 1: test whether elevated gustatory/oral somatosensory and reward region responsivity and reduced inhibitory region responsivity to palatable food images, cues, and intake of food varying in sugar/fat content, and behavioral inhibitory control deficits/immediate reward bias predict initial increases in % body fat in 130 lean teens. Aim 2: use growth curve models to test whether initial increases in % body fat and energy dense food intake predict future decreases in striatal response to palatable food receipt, increases in reward circuitry response to palatable food images/cues, decreased inhibitory region response to food images/cues, and increased behavioral inhibitory control deficits/immediate reward bias. Aim 3: test whether decreased striatal response to palatable food, increased reward region response to food images/cues, reduced inhibitory region response to food images/cues, behavioral inhibitory control deficits/immediate reward bias predict further escalation in % body fat.