CORTEX-II: A randomized trial to test the effectiveness of multi-modal cognitive training on yearlong physical activity self-regulation among middle-aged adults

The purpose of this randomized controlled trial was to test the effectiveness of a multimodal cognitive training program for increasing physical activity engagement among middle-aged adults. Low-active adults (N=233; Mage=46.7) predominantly non-Hispanic (92.3%), White (81.1%), female (65.6%) with a college degree (74.7%) were recruited to participate in a 13-month study. Participants completed a battery of assessments (psychosocial, neuropsychological, and physical functioning) at baseline and subsequently were randomly assigned to receive 20 hours of cognitive training via Games (stationary computerized tasks emphasizing executive functioning and exerciser self-certainty, and exergaming with dual-task components; n=118) or attention-control health educational Videos (n=115). Training was split across five 2-hr lab sessions (overall adherence=98.3%), followed by five 2-hr home sessions. Among Games, 81.4% fully complied with our exergaming protocol (lab & home) and 88.4% completed the cognitive training protocol. Video engagement was assessed via accuracy to attentional checks (Ms=82.4% accuracy in lab, 82.0% accuracy at home). The primary outcome was physical activity engagement (PAE), a latent factor based on physical activity steps & moderate intensity minutes derived from yearlong Fitbit step recordings, visitations & classes, & Godin Leisure-Time Exercise Questionnaire). As hypothesized, Bayesian multiple-indicator, multiple-cause model analysis indicated a significant group effect (PAE score M±SD is 4.20±1.02 in Games and 3.77±0.92 in Videos; Cohen’s d=0.19 and one-sided p= 0.03) after adjusting for a set of background variables. Based on 58.4% respondents to our study evaluation survey, 87.3% enjoyed the intervention they were part of, 87.4% enjoyed the yearlong exercise program, and 97.1% would recommend it to friends/family. Only 4.1% of study completers were able to guess the study's true purpose. Results associated with hypothesized mechanisms of behavior change, including cognitive functioning and psychosocial outcomes, will be discussed. Our findings replicated our previous trial and show reliable effects for promoting exercise engagement within an unsupervised community-based exercise program for middle-aged adults.