Original title: The Experimental Study of Habitual Behavior in Humans: A Review of the Validity of Current Protocols
Pablo Martínez-López, Francisco Garre-Frutos, Miguel A. Vadillo, David Luque · September 2026
Scientists define habits as automatic actions triggered by a situation, which keep happening even when the reward stops mattering. Researchers usually try to create habits in labs by having people repeat actions over and over. However, most lab tests fail to show that habits actually form. To investigate this issue, the researchers reviewed fourteen experimental methods that varied how much training people received. They also reanalyzed public data to test how reliable the measurements were. They discovered that traditional lab tests based on original habit theory consistently fail to form habits. Newer tests do show training effects, but they change what counts as a habit. Only one method, the forced-response task, successfully created habits while sticking to a strict definition and showing high reliability.
You are trying to learn to tie your shoes without looking. At first, you tie them differently every time—sometimes too tight, sometimes loose. Then you practice until your fingers move the same way each time, even when you are distracted by a conversation. The problem is that in most practice setups, you accidentally change the type of knot each day, so you never build a stable routine. In this paper, most lab methods for studying habits failed to produce that stable, automatic pattern, just like practicing different knots never teaches you one reliable procedure.
Habits are central to everything from breaking bad routines to building good ones. But if the experiments meant to study them don't reliably produce habitual behavior, the science behind habit-change advice—for smokers, dieters, students—rests on shaky ground. This review shows that some popular methods don't work as assumed, pointing the field toward more trustworthy measures.
The review does not introduce new experimental data; its conclusions are based on reanalysis of existing public datasets and the subset of paradigms included. The forced-response task shows reliability but the measure is narrow—it may not capture all aspects of habit, and it has not been tested against real-world habit outcomes. The abstract does not specify the number of participants or studies analyzed for each paradigm.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
The Experimental Study of Habitual Behavior in Humans: A Review of the Validity of Current Protocols — Pablo Martínez-López, Francisco Garre-Frutos, Miguel A. Vadillo et al. — PsyArXiv — doi:10.31234/osf.io/r62pa_v2
Abstract licence: CC BY
Explanation generated by Synthesis from the abstract — it is not the original paper. This is a preprint, so it hasn't necessarily been peer reviewed.