The experimental study of habitual behavior in humans: A review of the validity of current protocols
Martínez-López, P., Garre-Frutos, F., Vadillo, M. A., & Luque, D.
PsyArXivPreprint
Abstract
Although habits are widely recognized as playing a key role across psychological science, current laboratory methods may limit our understanding of these behaviors. Within the dual-system framework, habits are defined as context-triggered, outcome-insensitive responses that emerge gradually with repeated instrumental training. Many human experimental paradigms have been developed in an attempt to test this characterization of habits. However, in most cases, they have not provided clear evidence of habit formation, suggesting a gap between habit theory and existing methodologies. In the present study, we systematically review evidence on the construct validity of fourteen paradigms in which the amount of training was manipulated. We also reanalyze publicly available datasets from studies using these paradigms to assess the psychometric reliability of their main measures, thereby extending the available validity evidence. Paradigms that most closely mirror classical approaches derived from habit theory have consistently failed to induce habits, whereas more recent approaches have revealed training effects. However, these novel paradigms often rely on broader operationalizations that substantially diverge from traditional definitions of habit. Among the paradigms reviewed, the forced-response task stands out as capable of inducing habitual behavior while providing a narrowly operationalized measure with acceptable reliability. Finally, we argue that strengthening the credibility of the field will require greater attention to the psychometric properties of habit measures, including precise operationalization, reliability, and systematic validity testing.