How behavioural science could help us stop coronavirus

Covid-19 transmission is dictated by human behaviour. So how do we change how people behave?

March 20, 2020
The way we behave isn't just dictated by what we are told, but how we are told it. Photo: PA
The way we behave isn't just dictated by what we are told, but how we are told it. Photo: PA

A few days back, I walked into a Dublin hospital that is getting it right. Immediately beyond the revolving door, just a few yards into the main lobby, stood a large, red “Stop” sign, the same shape as at road junctions. Below it was a dispenser containing hand sanitiser. At the reception desk, before requesting my name or appointment letter, the receptionist politely asked, “Did you use the hand sanitiser?”

The role of behavioural science, and the David Cameron-created “Nudge Unit” in particular, have been in the spotlight since the coronavirus crisis began. The government’s behavioural scientists of choice advised that the public’s willingness to comply with containment measures would be time-limited.

Once the chief medical officer for England, Chris Whitty, had remarked that “if you move too early… people get fatigued,” it was clear that an argument based in psychology was leading the government to delay its move. I will come back to that. Something important needs to be established first.

Behavioural science can save lives

I began in that Dublin hospital because it demonstrates just how much insight—cheap, practical but effective and potentially life-saving insight—behavioural science can offer. The behavioural intervention in that hospital foyer ticked multiple boxes. The novelty of the sign grabbed my attention, but the location mattered too. I pretty much had to walk around the sanitiser station to avoid tripping over it. As well as this physical context, the social context mattered. Had I walked past the station without using it, people might have looked disapproving, perhaps even said something. Experimental trials show that placing highly-coloured sanitiser stations in the middle of public spaces substantially increases use.

More generally, much evidence shows that concern about social disapproval makes the more selfish, or self-absorbed, among us behave in a more public-spirited fashion. The receptionist’s question, asked in front of others in the queue, exploited this same behavioural effect. I’m pretty sure one guy behind me lied to the receptionist when he said “yes”—his hesitation just discernible. He then used another sanitiser placed in front of the elevator doors.

So this intervention is backed by behavioural science. Every public building, business or workplace could adopt it, or something similar. Assuming that the public health authorities know about it, they can spread the evidence and promote adoption, perhaps encouraging an additional dash of creativity or, whisper it, fun.

“Behavioural science” is a catch-all that covers multiple social sciences, including parts of anthropology and sociology. Yet the increased use of this term is primarily a rebranding exercise to get society’s decision-makers to take psychology more seriously. It follows the success of “behavioural economics” in getting economists to do likewise.

Behavioural science has a crucial role to play in the response to this deadly virus. Like most scientific disciplines, it involves a mixture of theory and empirics. Some of the work is mathematical and technical, but much of it is not.

At its most simple, behavioural science is the observation of regularities in how people behave: in context A, people mostly do X; in context B, they mostly do Y. The logic is simple: transmission of Covid-19 depends on human behaviour and behavioural science reveals how context changes behaviour, therefore good use of behavioural science can reduce transmission.

Hand hygiene is one area, but behavioural science can help with others. For instance, there is good evidence that periods of isolation are tough going and can cause psychological harm. The worry is that this may undermine the effectiveness of self-isolation, because some people won’t do it, or won’t do it properly. Yet there are ways to mitigate this, tested by behavioural scientists, which maintain social connectedness and alter perception of time. If you need to isolate, tell as many friends and family as possible and encourage them to keep contacting you. Establish a daily routine, complete with alarm clock and regular tasks. The evidence suggests that these tips will make the necessary days pass by more easily.

Increased handwashing, changing how we cough and sneeze, self-isolation, avoiding travel, social distancing—all are impositions on the individual that benefit our fellow citizens. My behaviour affects your chance of contracting the disease and your behaviour affects mine. These sorts of collective action problems have been studied by behavioural scientists for several decades. There is robust evidence about how people respond. Some of us behave altruistically and some selfishly, but the majority will override self-interest and contribute to the common good, provided others pitch-in too. Many of us are “conditional co-operators,” who want to support the common effort, but not if we are one of only a few mugs doing it while others get a free ride.

As well as measuring how people respond to the social dilemmas at the heart of collective action, behavioural science has revealed what makes a successful social outcome more likely. Three things stand out. First, co-operation is more likely if people communicate to each other why the specific behaviour is best for all. Second, people are more willing to make sacrifices for group benefit the more they identify with the group. Third, moderate punishment is needed for those who don’t co-operate. This could merely be social disapproval from the majority, perhaps something a bit stronger, but nothing over-the-top.

Why the government is getting it wrong

At this point, you might wonder what advice behavioural scientists have actually given the UK government regarding Covid-19. As a behavioural scientist whose job is to produce evidence for policy in Ireland, I certainly do. There should be a clear articulation of a strategy that is best for all, encouraging behaviours that everyone can get behind in a national collective effort, and that people can badger their fellow citizens to stick to. Instead, there are foggy statements of rationale, for behaviours that are disputed by national experts, that some people are encouraging their fellow citizens to stick two fingers up to. My mother is in her seventies and lives in Oxford, where I grew up. Her response to current official advice is unprintable.

UK policy is now playing catch up on Covid-19, having departed from the global norm. Something went wrong and, to be clear, I am not the only behavioural scientist who thinks so. An open letter to government from more than 200 behavioural scientists expressed concern about the quality of behavioural evidence being relied upon. The early signatories read like a Who’s Who of British behavioural science. The international network of applied behavioural scientists to which I belong has been sharing the relevant tweets and posts, aghast and uncomprehending.

Let us take a pace backwards, for there is an irony here. The UK has led the world in the application of behavioural science to policy. After the “Nudge Unit” (or, more formally, the Behavioural Insights Team) was set up in the Cabinet Office 10 years ago, many were quick to deride it, or dismiss it as a fad. These detractors were wrong. An initially small team made successful contributions to multiple policy challenges, leading to expansion and emulators around the world. Some signatories of the open letter have supported and worked with the Behavioural Insights Team.

So what went wrong? Only those on the inside will truly know. Yet from a distance, as someone who does similar work, I see lessons that perhaps we can all draw upon. We are, after all, trying to fight the same lethal virus.

The touchstone must be evidence

One thing that makes it difficult is the need for government to coordinate multiple expert scientific communities. Arguably, this is especially difficult in countries with right-wing governments because, while there are many exceptions, most top scientists are liberals—that’s just what life in university is like. However, it’s not just the interface between government and scientists that has to work efficiently to meet this crisis, it’s the interface between scientific disciplines.

The scientific army includes epidemiologists, microbiologists (containing the subdiscipline of virologists), general medics, public health specialists, behavioural scientists and economists—with apologies to battalions I have missed. To fight together, they must work rapidly under pressure, keeping egos and insecurities in check. These scientists need to know the boundaries of their expertise, not overstretch themselves, and build relationships of trust with leaders in other disciplines, senior civil servants and politicians.

This is really hard. Good scientists are often natural sceptics who presume that self-professed experts don’t know what they are talking about until they produce work that proves otherwise. Having lived through the financial crisis in Ireland, I confess to this worldview.

The touchstone, always, has to be evidence. This applies as much to the study of behaviour as to any longer-established and supposedly “harder” science. Research is classed as science not because of the domain of study, but because of how that study is undertaken. The scientific method requires that we verify propositions with evidence, approach them critically, and handle them with meticulous care. The method is empirical: not about conjecture or assertion of abstraction, but open-minded inquiry, in which anything can be falsified by data.

Our present situation is too serious for people to go beyond the evidence, to act instead on hunches or, worse, a desire to be influential. Humility about what we know and what we don’t is fundamental. When I hear behavioural scientists trying to explain herd immunity to the public, I start to worry, just as I do when I hear a medic talking about how to solve collective action problems.

Like the authors of the open letter, I cannot locate the evidence behind the claim about behavioural fatigue that delayed the UK’s policy response. If such evidence exists, it needs to be made public; the rest of the world should see it too. If such evidence does not exist, it will be important to understand how the claim arose. Moreover, it should be made clear to senior decision-makers that there is nothing about behavioural science, as a discipline, that lends itself to unsupported claims—this was a one-off.

As described above, evidence from behavioural science really can be used to support behaviour that could slow the spread of Covid-19. Many behavioural scientists are now striving to generate new evidence for interventions and communication materials specific to the current crisis. If we understand where this work can be best applied and put it to use, it may save lives.

To read the full paper from Pete Lunn and colleagues, "Using behavioural science to help fight the coronavirus," click here.