The Effect Of Behavioral Automaticity On Behavior Is Moderated By Cognitive Self-control Abilities

Abstract

PURPOSE: Dual-process theories assume that physical activity (PA) behavior is regulated by two different processes: implicit (i.e., automatic, unconscious) and explicit (i.e., effortful, conscious) processes. In this study we examined the interaction of implicit (i.e., behavioral automaticity) and explicit (i.e., cognitive self-control abilities) processes on PA behavior. We expected significant interaction effects between behavioral automaticity and cognitive self-control abilities (i.e., inhibition): highly automatized behaviors will prevail when inhibition abilities are poor, while high inhibition abilities might help to inhibit unwanted automatic behavioral tendencies (e.g., highly automatized sedentary behaviors [SB]). METHODS: A prospective study with two points of measurement (N = 114 undergraduate and graduate students) was conducted. At t1 age, sex, past PA behavior (control variables) and automaticity of a) PA and b) SB were assessed with standardized questionnaires. Inhibition was assessed with a computerized Stop-Signal and a Go/No-Go task. At t2 (4 weeks later), PA behavior was measured as dependent variable with a standardized questionnaire. Hierarchical multiple linear regression analyses with interactions Automaticity x Inhibition on PA behavior and subsequent moderation analyses were calculated for automaticity of a) PA and b) SB respectively. RESULTS: The expected interaction effects Automaticity x Inhibition on PA behavior were significant for automaticity of a) PA (b = 55.23, p < .01) as well as b) SB (b = -27.40, p < .05). Moderation analyses revealed that PA automaticity was a significant positive predictor of PA behavior when inhibition abilities were poor (b = 105.75, SE = 23.13, t = 4.57, p < .001), but not when they were high (b = -11.10, SE = 24.71, t = -0.45, p = .65). Furthermore, automaticity of SB was a significant negative predictor of PA behavior when inhibition abilities were poor (b = -42.83, SE = 20.87, t = -2.05, p < .05), but not when they were high (b = 13.67, SE = 17.61, t = 0.77, p = .44). CONCLUSIONS: In line with theoretical assumptions, automatic behaviors prevailed when the ability to inhibit prepotent responses was poor. However, higher inhibition abilities erase the significant associations between automaticity and behavior for both, PA and SB automaticity