Association for Behavior Analysis International

Overweight and obesity are problems of considerable social significance. Both overweight and obesity are primarily the result of an energy imbalance, with overweight and obese individuals consuming more calories than they expend. One of the primary behavioral factors related to this imbalance is physical activity. Increasing energy expenditure through increased physical activity can serve to remediate the energy imbalance, so long as calorie consumption does not increase. Moreover, increased physical activity can produce health benefits even in the absence of weight loss. Despite the obvious relevance of behavior analysis to these issues, relatively little behavior-analytic research has focused on physical activity in the context of overweight and obesity. This symposium will feature several studies focused on understanding and increasing physical activity levels in children and adults from a behavior analytic perspective. Two of the papers report the results of efforts to increase physical activity, one with children and one with adults. Two papers will report the results of assessments to determine conditions that promote more or less physical activity in children.

Physical activity is important for children as many children are considered overweight or obese. In the current study, a multiple baseline design across participants was used to assess the effectiveness of goal setting, reinforcement contingencies, and pedometers that provide feedback to increase step count by 5 participants. During baseline each participant wore a sealed pedometer to assess the average steps the participants took per day. In intervention 1, each participant set a reasonable goal of steps to achieve per day in order to receive a specific reinforcer chosen by the parent and participant. Intervention 2 added daily phone calls from the researcher to check participant progress. Data collection for 4 of the 5 participants showed a mean increase in steps taken per day during both interventions. All 3 participants that participated in intervention 2 further increased their mean count. Two participants participated in the follow-up phase of the study; both participants maintained their goals from intervention 2 and completed their goal step count on 100% of days.

The prevalence of overweight and obese children is an increasing concern in the United States and abroad. Studies have shown that higher activity levels correspond with lower body mass index and overall better health. Given the current percentage of children enrolled in both public and private schools, it may be the case that a significant portion of children's daily activity can be fulfilled in school. The aim of the present study was to evaluate the efficacy of step counts as reliable indicators of differential activity levels displayed by children during recess. Fifteen elementary school-aged children wore Fitbit accelerometers during their regular recess times. Both structured and unstructured activities were evaluated. Step counts were recorded from Fitbits in 1-min intervals, and specific recess activities were observed and recorded. The results indicated that step count data produced by the Fitbit were successful at identifying differential patterns of activity levels. Overall, higher step counts were observed during jogging and soccer, and lower step counts were observed during tag and free play. The data also indicated a gender difference, with boys reaching higher activity levels than girls during free play. Implications for interventions utilizing step counts for feedback and goal-setting will be discussed.

Behavioral contracting has been shown to be effective in improving a variety of behaviors, including weight management and exercise (Aragona, Cassady, & Drabman, 1975; Mann, 1972; Neale, Singleton, Dupuis, & Hess, 1990). Although research has examined the efficacy of various behavioral contract components (e.g., response cost), no studies have directly compared the efficacy of the type of deposit (e.g., monetary versus personal goods). In the current study, 2 female employees of a pediatric hospital, ages 32 and 24, wore a multidirectional accelerometer, which measured energy expenditure, during the workday. Following baseline and feedback-only conditions, participants were exposed to 2 behavioral contracting conditions, a monetary deposit and a personal goods deposit. Results showed that the most effective deposit contract at increasing energy expenditure differed across 2 participants. Additionally, for both participants only 1 of the deposit types resulted in meeting the goal and earning back the deposit entirely. These results suggest that to maximize the effectiveness of behavioral contracting, the most effective contract type needs to be identified and utilized prior to entering into a contract.

Direct observation of children being physically active often is viewed as the "gold standard" measure of assessing physical activity. Although direct observation can provide information regarding contextual variables (i.e., context and composition) related to moderate and vigorous physical activity, environmental variables often are not manipulated and thus causal relationships cannot be determined. We developed an experimental analysis methodology to assess the effect of outdoor activity context and group composition on level of physical activity in preschool children. The purpose of the current study was to compare the results from descriptive analyses of physical activity to those of experimental analyses. Descriptive analyses were conducted during 30-min sessions on an outdoor playground at a local daycare. Experimental analysis sessions were conducted on an outdoor playground within a multielement experimental design, with each condition lasting 5 min. The Observational System for Recording Physical Activity in Children (McIver et al., 2009) was used to define the conditions and various levels of physical activity within both the descriptive and functional analyses. Initial data suggest that the degree of agreement between the descriptive analyses and experimental analyses varies by participant.