The researchers looked at published failed behavioural interventions across all areas that impact society, from healthy eating and organ donation, to tax compliance. They showed that whilst any type of behavioural intervention, applied in any type of setting, could be liable to fail, certain types of intervention were more likely to fail.

Current behavioural change programmes focus largely on promoting successes. This new study suggests that improved understanding of why and how interventions fail could help develop successful behavioural interventions in future, and avoid wasting time and money on interventions that will likely fail.

For the project, the researchers analysed 65 articles, published between 2008 and 2019, which identified failed behavioural interventions, including nudges. They identified eight different types of failures in total, which include 'backfires' whereby the introduction of the nudge intervention made the behavioural problem worse rather than better.

The most common type of interventions that resulted in failures were those involving social norming or social comparisons, where individuals are provided with information about the behaviour of their peers in order to encourage a desired behaviour change. Interventions that involved the provision of information through letters or text messaging, accounted for almost a quarter of the failed studies.

Dr Magda Osman, Reader in Experimental Psychology at Queen Mary University of London, said: "Our analysis provides the first attempt to systematically examine behavioural interventions that fail. We have shown that failures are quite common and can occur with nudges applied in any type of setting. We found that there are different types of failures, from interventions that simply don't achieve any behavioural change, to those that achieve negative changes such as backfire effects."

In the article, the researchers also show the benefits of using computational causal modelling techniques to map out the different factors that can influence specific behavioural interventions and their likelihood of success. This allows researchers and decision-makers a way of mapping out in advance what might work, as well as what might undermine the intervention ahead of time.

Dr Osman, added: "We believe that causal analysis can advance existing behaviour change frameworks as they allow us to formally model behaviour change problems and the context in which these interventions are situated. By incorporating these approaches into the early design of behavioural interventions, we can begin to understand what factors are relevant to the success of the intervention and how the intervention could influence these factors, and even prepare precautionary measures to help avoid failure."

The use of psychological insights to motivate people to change their opinions, attitudes and behaviours goes back at least as far as the 1950's when it was referred to as behavioural engineering. Many public and private institutions now use behavioural change techniques to influence positive change, from improving dietary choices to helping people save more for their retirement. More recently, governments have sought advice from experts on behavioural interventions to ensure public compliance with their proposed strategies to manage the Covid-19 pandemic, for example on behaviours such as social distancing and wearing masks.

Dr Osman, said: "It's clear to see that there's currently a great appetite for the use of behavioural techniques in society, and we're seeing terms like nudge being widely used in both scientific and public settings. However, the behavioural change enterprise disproportionately focuses on promoting successes at the expense of examining the failures of behavioural change interventions. Understanding why behavioural changes fail, and being able to anticipate possible types of failures when designing interventions could help to save time and public funds invested in these techniques, and overall increase their success in achieving the desired behavioural change."

Story Source:

Materials provided by Queen Mary University of London. Note: Content may be edited for style and length.

Journal Reference:

1. Magda Osman, Scott McLachlan, Norman Fenton, Martin Neil, Ragnar Löfstedt, Björn Meder. Learning from Behavioural Changes That Fail. Trends in Cognitive Sciences, 2020; DOI: 10.1016/j.tics.2020.09.009