A pilot study evaluating a one-session attention modification training to decrease overeating in obese children

Kerri N. Boutelle, Jennie M. Kuckertz, Jordan Carlson, Nader Amir

There are a number of neurocognitive and behavioral mechanisms that contribute to overeating and obesity, including an attentional bias to food cues. Attention modification programs, which implicitly train attention away from specific cues, have been used in anxiety and substance abuse, and could logically be applied to food cues. The purpose of this study was to evaluate the initial efficacy of a single session attention modification training for food cues (AMP) on overeating in overweight and obese children. Twenty-four obese children who eat in the absence of hunger participated in two visits and were assigned to an attention modification program (AMP) or attentional control program (ACC). The AMP program trained attention away 100% of the time from food words to neutral words. The ACC program trained attention 50% of the time to neutral and 50% of the time to food. Outcome measures included the eating in the absence of hunger free access session, and measures of craving, liking and salivation. Results revealed significant treatment effects for EAH percent and EAH kcal (group by time interactions p<.05). Children in the ACC condition showed a significant increase over time in the number of calories consumed in the free access session (within group t=3.09, p=.009) as well as the percent of daily caloric needs consumed in free access (within group t=3.37, p=.006), whereas children in the AMP group demonstrated slight decreases in these variables (within group t=-0.75 and -0.63, respectively). There was a trend suggesting a beneficial effect of AMP as compared to ACC for attentional bias (group by time interaction p=.073). Changes in craving, liking and saliva were not significantly different between groups (ps=.178-.527). This is the first study to demonstrate that an AMP program can influence eating in obese children. Larger studies are needed to replicate and extend these results.