Objective: Social cognition theories offer a pre-eminent account of behavior in health contexts and are based on the assumption that behavior is a function of reasoned decision making processes, represented by intention-mediated effects of theory constructs on behavior. However, these theories do not account for effects of constructs that represent the non-conscious processes that line up behavior, such as habits. Researchers have augmented these theories to test effects of the habit construct on behavior and broaden their predictive validity. Drawing from this research, we tested key hypotheses of habit effects on behavior in these theories using meta-analysis: the relative effects of habit and intention on behavior; the mediation of past behavior-future behavior effects by habit; and the mediation of habit-behavior effects by social cognition constructs and intention. We also tested the moderation of habit effects by habit measure type, behavior type (e.g., specific health behaviors, health behaviors vs. non-health behaviors), opportunity for habit formation, behavior complexity, behavior measure type, and measurement lag. Method: A database search identified studies (k=248) reporting relations among three habit measures (behavioral frequency x context stability, response frequency, self-report measures), intention, behavior, and social cognition constructs from the theory of planned behavior – a prototypical social cognition theory. Data were analyzed using multi-level meta-analytic structural equation modeling. Results: Habit measures and intention independently predicted behavior. Pastfuture behavior and habit-behavior effects were partially mediated by habit measures, and intention and social cognition constructs, respectively. Larger habit-behavior effects were observed when habit measures included behavioral frequency items, for complex behaviors (e.g., physical activity), for ‘rewarding’ behaviors such as alcohol consumption, and when habit was measured in close proximity to behavior. Conclusion: Findings indicate that past-future behavior effects can be partially attributed to habit, habit-behavior effects are partially subsumed by intention and social cognition constructs, and habit-behavior effects were larger for less complex and rewarding behaviors, and when measurement lag was shorter. Findings are expected to catalyze future research exploring habit effects in social cognition theories using experimental methods and non-self-report measures. The research also highlights the imperative of encompassing habit measures in predictive studies, as well as potentially tapping into strategies to target habit change in behavioral interventions aimed at promoting health behavior or breaking maladaptive habitual behaviors, particularly potentially rewarding and less complex behaviors.