What Is a Mechanism of Action in Behavioral Science?

A mechanism of action (MoA) is a process through which a component of an intervention is thought to influence a behavior. Connell and colleagues (2019) consider psychological, social, and physical processes. An intervention might build a skill, change what people expect from their peers, or give them easier access to a resource they need. The hypothesis concerns how that change affects what people do.

Behavior change techniques (BCTs) (in French) describe what an intervention delivers. A reminder in a work tool is a technique. Drawing someone’s attention to a task when they can act on it is the proposed mechanism. Completing the task is the behavior the team tracks.

How Do You Form a Hypothesis About a Mechanism at Work?

Take the fictional billing example. For standard contracts, the project manager is expected to send the client the monthly billing summary for approval by the first business day after the month-end close. When the data are spread across several tools, prefilling brings them together where the manager prepares the summary. The hypothesis is that direct access to the data reduces the work of reconciling them and makes it more likely the manager will send the summary on time. The time spent finding and correcting data gives the team a way to check what changed.

Suppose the portal also displays a cue when the summary is ready. For a manager who already had everything needed to act, the cue might bring the task back to mind in time to send the summary earlier. The APEASE criteria can help compare these options with a joint preparation session with accounting. That option tests whether having someone available at month-end helps resolve open questions that hold up the summary. To distinguish these explanations, the team looks at what changes between preparing and sending the summary.

What the Intervention Delivers Proposed Mechanism What to Observe
Prefilled summary Direct access to the data cuts the time spent finding and reconciling information before sending the summary. Time spent preparing the summary, number of corrections, and share of summaries sent on time.
Cue in the portal The cue draws the project manager’s attention to the action when the summary is ready. When the cue appears, when the manager opens the summary, and when they send it. Check that the summary was ready before the cue.
Joint session with accounting Support at month-end helps resolve open questions about what can be billed. Questions resolved during the session, time to resolve them, and share of summaries sent on time.

What Do We Know About the Links Between Techniques and Mechanisms?

Carey and colleagues (2019) examined 277 articles to identify how authors linked the techniques they used to the mechanisms they proposed. The review shows which explanations people offered in the published literature. The frequency of a link tells you how often authors proposed it.

Connell and colleagues (2019) asked behavior change experts which links they found plausible. The experts rated the links, discussed them, and rated them again. For many of the links they considered, the resulting judgments remained in the possible or uncertain range. That identifies questions future studies can investigate.

Johnston and colleagues (2021) compared the two sources. The literature review and initial expert consensus agreed on 37 links. A new expert panel reviewed the disagreements and uncertainties and identified 55 additional links. Together, the 92 links give intervention designers hypotheses to draw on. Whether any one link explains a change in a particular intervention is a question for evaluation.

How Do You Use the Theory and Techniques Tool?

The Theory and Techniques Tool (TaTT) presents that synthesis as a matrix. Start with a technique to see the mechanisms associated with it, or start with a proposed mechanism to see which techniques might act on it. Each entry shows the basis for that assessment. An inconclusive rating means the evidence and judgments gathered leave the question open.

Behavioral practice or rehearsal (BCT 8.1) is one example. In Johnston and colleagues’ analysis, the literature and the experts agreed on links with both skills and beliefs about capabilities. A team that asks people to rehearse a task can observe how well they perform it and ask how confident they are about doing it on their own. The COM-B model helps the team identify the barrier first. The TaTT then helps it document how a technique might address that barrier.

How Can a Pilot Test a Mechanism?

Track what was made available to people, the process the team expects to change, and the behavior itself. If the proposed mechanism involves reducing preparation time, gather observations while people are preparing the work, before they have to take the target action. Track an appropriate behavioral indicator (in French) as well. The sequence matters because the intermediate change needs to occur early enough to influence the action.

If the process changes and the behavior is still rare, look for barriers that remain. If people take the action more often and the process barely changes, investigate other explanations, such as a reminder in a meeting or a shift in priorities. When an intervention includes several components, observing how people use them or introducing them at different times can help assess their respective contributions. What the team can conclude depends on the quality of those comparisons and which other explanations it was able to examine.

Where Do Mechanisms of Action Fit in MEET©?

In MEET©, the team uses Model to describe the practice and understand what helps or hinders it. In Engineer, the team specifies what the intervention should change and how that change could affect the practice. It consults the TaTT when a technique needs a better-supported rationale or when several explanations are plausible. That hypothesis becomes part of the theory of change, which connects it to the other conditions for delivery and the business outcome.

In Experiment, the team checks the proposed mechanism against the pilot findings. In Transfer, it checks whether the conditions that supported the mechanism are in place elsewhere. A pilot that worked in a team with reliable data gives the company a reason to look at how other teams get those data. If another team has to reconstruct them, the person leading the rollout can adapt the intervention before extending it.

Frequently Asked Questions

What is the difference between a determinant of behavior and a mechanism of action?

A determinant is a condition that influences behavior in a given situation. A hypothesis about a mechanism says the intervention will change that condition and that the change will affect the behavior. A team might find that a piece of information is missing when people need to act. It can then test whether making the information directly available cuts search time and makes the action easier.

Can one technique work through several mechanisms?

Yes. A reminder can draw attention to a task and signal that the organization considers it a priority. The team uses its diagnosis to decide which explanations to pursue and chooses observations that help tell them apart.

What should you do if the TaTT shows an inconclusive link?

Go back to the observations that led you to choose the technique. They might justify a targeted pilot if the mechanism is plausible in this setting and the team can observe a sign of it before the behavior. If the uncertainty matters to the decision and the pilot would leave it unresolved, consider an option whose expected effect is easier to observe.

References

Carey, R. N., Connell, L. E., Johnston, M., Rothman, A. J., de Bruin, M., Kelly, M. P., and Michie, S. (2019). Behavior Change Techniques and Their Mechanisms of Action: A Synthesis of Links Described in Published Intervention Literature. Annals of Behavioral Medicine, 53(8), 693-707. 10.1093/abm/kay078

Connell, L. E., Carey, R. N., de Bruin, M., Rothman, A. J., Johnston, M., Kelly, M. P., and Michie, S. (2019). Links Between Behavior Change Techniques and Mechanisms of Action: An Expert Consensus Study. Annals of Behavioral Medicine, 53(8), 708-720. 10.1093/abm/kay082

Johnston, M., Carey, R. N., Connell Bohlen, L. E., Johnston, D. W., Rothman, A. J., de Bruin, M., Kelly, M. P., Groarke, H., and Michie, S. (2021). Development of an online tool for linking behavior change techniques and mechanisms of action based on triangulation of findings from literature synthesis and expert consensus. Translational Behavioral Medicine, 11(5), 1049-1065. 10.1093/tbm/ibaa050

About the Author

Maxime Rabéchault is the founder and president of UNREST, a change enablement consultancy. He applies behavioral science to organizational transformation and leadership development.