UC Merced

We tested a dual process model incorporating constructs that reflect both performing the target behaviour (behaviour directed habit) and habits that run counter to the target behaviour (opposing behaviour habit) in accounting for variance in two health behaviours: eating the recommended serves of fruits and vegetables a day and restricting sugar-sweetened beverage consumption. A prospective correlational design with two waves of data collection separated by one week was adopted. Participants (N = 606) comprising middle school students (n = 266) and university students (n = 340) completed an initial survey comprising self-report measures of past behaviour, intention, and habit to perform the target behaviour and habits that run counter to the target behaviour. One week later, participants (N = 414) completed a self-reported measure of behaviour. Results revealed that behaviour directed habits predicted fruit and vegetable consumption in both samples, while opposing behaviour habits predicted restriction of sugar-sweetened beverages in the middle-school sample only, with a moderating effect also observed. Current findings indicate that habits specifying avoidance of the target behaviour did not predict future behaviour. However, the moderating effect observed provides preliminary evidence that strong habits to perform a behaviour may override habit to avoid the behaviour.