How Long Does a Nudge Last?

 

Decision Sciences

How Long Does a Nudge Last?

Todd Rogers Faculty ResearcherTodd Rogers, Assistant Professor of Public Policy, Harvard Kennedy SchoolPaper Title
Changing Behavior Beyond the Here and NowCoauthors
Erin Frey, Harvard Business School

The behavioral sciences have made a big splash in public policy. Following the insights of psychologists into the science of human decision making, researchers have lifted the veil from assumptions of logical human behavior and have begun to delve into why we so often make poor decisions (or sometimes fail to decide at all). What they have found is that giving a few small nudges can help us head in the right direction.

Those nudges have increased employee participation in retirement saving plans, helped people remember to schedule important medical tests, and boosted voter participation in local elections. Often only a small prod puts us on the intended path. Participation in saving plans, for example, soared after researchers realized that a major barrier to joining employee retirement programs was procrastination. The fix: Make the programs opt-out so that inaction would mean employees automatically saved more with each pay cycle. And when researchers looked at how to get more patients to have important colonoscopies, they found that a simple sticky note containing the relevant appointment information substantially increased attendance.

But how long do these nudges last, and can they be designed to have the desired effect at the right time? In a recent paper, “Changing Behavior Beyond the Here and Now,” Todd Rogers, Assistant Professor of Public Policy at Harvard Kennedy School, and coauthor Erin Frey, of Harvard Business School, looked at how they work over time.

“Behavioral interventions vary widely in the degree to which there is a lag between the moment that they are administered and the moment that the target behavior is to be performed,” the authors write.

They identify two types of interventions: those that are most effective at changing behavior soon after they are administered, and those that induce lasting changes.

“Defaulting individuals into a choice, capturing and directing individuals’ attention, and momentarily changing what thoughts are cognitively accessible to individuals in a given moment are especially likely to affect behavior when they are associated with short intervention-behavior lags,” Rogers and Frey write. “In some cases, simply correcting inaccurate but important beliefs can change mental content in ways that help interventions bridge time.”

The authors also turned their attention to the marginal benefit of continued intervention—examining whether each additional behavioral prod has a diminishing effect. “Specifically, we are interested in how additional administration of an intervention could result in behavior change above and beyond the behavior change that might persist from whatever treatment had already been administered,” they write.

In addition, they looked at the persistence of effects after the intervention has ended: “That is, what are the psychological and structural pathways through which behaviors can enduringly change as a result of interventions?”

“Behavioral science is increasingly being used to develop interventions to influence important behaviors throughout society,” Rogers and Frey conclude. “From firms to governments, from community organizers to teachers, behavioral science offers insight and tools for changing behavior. With this manuscript we sought to show that there are knowable aspects of behavioral interventions that affect whether, how, and for how long they change behavior. Our hope is that scholars find these frameworks productive for advancing and organizing future research, and that they help those who develop behavioral interventions to make them more effective.”

- by Robert O'Neill

“Behavioral interventions vary widely in the degree to which there is a lag between the moment that they are administered and the moment that the target behavior is to be performed.”