Behavioural Science is a rapidly expanding field and everyday new research is being developed in academia, tested and implemented by practitioners in financial organisation, development agencies, government ‘nudge’ units and more. This interview is part of a series interviewing prominent people in the field. And in today's interview the answers are provided by Susan Michie. Susan is Professor of Health Psychology and Director of the Centre for Behaviour Change at University College London. She is co-Director of NIHR’s Behavioural Science Policy Research Unit and leads UCL’s membership of NIHR’s School of Public Health Research. She serves as an expert advisor on the UK’s Scientific Pandemic Influenza Group on Behavioural Science (Covid-19) and is a consultant advisor to the World Health Organisation on Covid-19 and behaviour. In addition (!!!) she is also expert advisor to Public Health England and the UK Department of Health and Social Care, is Chair of the UK Food Standard Agency’s Social Sciences Advisory Committee and chaired the Academy of Social Science’s ‘Health of People’ project. Her research focuses on behaviour change in relation to health and the environment: how to understand it theoretically and apply theory to intervention development, evaluation and implementation. She is also leads the Human Behaviour-Change Project.
Who or what got you into behavioural science? Well, that's a complex question. It wasn’t one person or one event, rather I would say it was a cumulation of events and people over the years. My father and stepmother were both interested in behavior and I was always interested in people in general. What made them tick? Why things went wrong for them, especially what went wrong within their brains? I remember reading a novel about somebody with schizophrenia, and seeing a television program about somebody with Down’s syndrome and being very interested in that. I did Experimental Psychology at Oxford University and found that I was very interested in behavior. We did classic experiments on rats, training them in various ways. And I thought that was very interesting; how you could use very simple stimuli to change behaviors of complex animals. And then I went on to do clinical psychology and within clinical psychology one can apply many of the very basic learning principles that we learned at university to help people solve their problems and achieve their goals, which I thought was a very attractive proposition: apply science to solve people’s real-world problems.
My mother, Anne McLaren, was a huge influence in my life and in my thinking. She was a scientist who was also very active in the arenas of ethical and social implications of science. I was brought up in a household where things were questioned, discussed, analysed, nothing was taken for granted or as it appeared without thinking about it. They were both Marxists and the philosophy of dialectical materialism which I read about at University has been a great framework within which to think about all questions. When I asked my mother why we weren’t taught dialectical materialism at school and University, I remember her saying “All great scientists are dialectical materialists, but most don’t realise they are.” I was always interested in behavior. I worked as a clinical psychologist for many years. And behavior is relevant to pretty well everything. I did research for many years looking at psychological aspects of genetic testing. There wasn't much room to study behavior, and that was quite frustrating. But I had one day a week working in an occupational health unit of a large Teaching Hospital, where I worked as an organizational consultant and developed and evaluated training programs for staff, and so that got me very interested in people's behavior within organizations and organizational systems and procedures as ways of supporting or preventing behavior change. It was fascinating and satisfying to work at an organizational level, to help teams, departments, and individuals change a whole range of different ways of being and behaving. And then I was offered a new post at University College London, which is where I still am. Because it was a new post I could develop it as I wanted to. So I started from scratch, having made my research name in psychology as applied to genetics and also to occupational health. I thought, “well, actually what I really want to do is study behavior change.” When I began studying it, I realized that it was not only complicated, but also that it was quite difficult to study it for various reasons. So much of my work has been problem solving in terms of how to develop methods to enable us to understand behavior, and also to help communicate the principles of behavior and behavior change to other people. I've always been very committed to translating academic expertise to individuals, organizations and populations who could potentially benefit from it.
What is the accomplishment you are proudest of as a behavioural scientist? And what do you still want to achieve? I think what I am proudest of is developing frameworks and tools that make the understanding and application of behavior and behavior change accessible to wide audiences, thus providing a bit of a bridge between academic expertise and changing policy and practice to address the world's problems. What I would really still like to achieve is the vision of the Human Behaviour-Change Project; a collaboration between behavioral science, computer science and system architecture to develop a knowledge system, using a behavior change intervention ontology that we're developing to organise evidence, and artificial intelligence and machine learning to automatically extract relevant information from the world literature to answer the kind of questions that policymakers and practitioners have. These are along the lines of what interventions are likely to work best? And for how long? And with what processes of change or mechanisms of action? The answers draw on all the evidence that there is out there, because the rate at which intervention evaluations are being published is estimated to be more than 100 a day. And it's impossible for human beings to be able to keep up with it, to be able to synthesize evidence across all the different domains that are relevant. And to be able to detect patterns in a way that machine learning and reasoning algorithms can do. In this project, we’ve been working with just one behavior, smoking cessation, and now moving on to another: physical activity. This will have taken us five years. But hopefully it will lead to an end-to-end system, including a user interface so that users can pose their questions, and include constraints they may have, for example, what modes of delivery the intervention might be or what type of population it might be targeting. Or what kind of processes of change they might want to limit it to: this will be of especial interest to researchers studying theory. Because the relevant information will be identified, extracted, and then organized into an ontology, which is an organizing framework for knowledge, evidence can be synthesized across very diverse studies using very diverse types of concepts, language and formats for reporting studies. One of the reasons I'm very keen on this is that I was a member of a national guideline committee for eight years, called eth Ppublic Health Interventions Advisory Committee, and I saw just how slow and not fit for purpose the current systematic reviewing and evidence synthesis methods are. I also became very aware that the research literature that's drawn is dominated by the US, whereas the world's greatest problems in terms of health and sustainability are in lower and middle income countries. What the knowledge system that we are developing will be able to do is answer questions where evidence is lacking by making informed inferences from huge amounts of other relevant information. I hope that it will not only provide much more and much better up-to-date evidence in a format that people can immediately use for their own situations, but I also hope that it will help to reduce the massive inequalities in terms of the usable evidence for all populations in all parts of the world.
If you weren’t a behavioural scientist, what would you be doing? One of the things I'd love to do is be an architect; save buildings that are falling down, fix buildings that are unloved and need rescuing. Whilst using new designs and new materials, especially ecological materials, to build really exciting new buildings. Not just restoring what was there but blending the old with the new.
How do you apply behavioural science in your personal life? Well, I have three children who are in their 30s and three grandchildren. And I find having an understanding of behavior and how to change it really helpful with them. I think it's very difficult to know where you as a person stops and where behavioral science begins. One of the things I hope I do is validate people in terms of reinforcing good things that they do and say. Expressing appreciation is something that I've learned from the many decades of research into rewards and seeing the huge power that rewarding behaviors of others has. It’s something we don't do nearly enough. And it costs nothing. But for some reason, we don't do it hugely. It gives so much pleasure to people when you do praise and acknowledge good things. It also makes for good relationships. I think I now naturally do this with friends and family, and I hope I do it with colleagues, but I am sure there is room for improvement. Does that come from being interested in people and wanting the best for the world and wanting the best for people around me? Or does it come from an understanding that behaving in that way, helps other people and oneself to achieve the goals in life one wants to achieve? I don't know.
With all your experience, what skills would you say are needed to be a behavioural scientist? Are there any recommendations you would make? To me, an analytic mind, openness, focus, persistence and collegiality are incredibly important. I think an analytic mind is important for science in general- to be able to think laterally, be able to pull things apart in precise ways. The other side of that is dialectics; seeing things or systems, seeing connections, and how things are put together, how links are made across boundaries, I think these are really important. But also having an open mind. To be prepared to be surprised, to be prepared to be proved wrong, and to embrace that and enjoy that. At the same time, it is important to be able to focus and to avoid a scattergun pot-pourri of interests and go up lots of different paths at the expense of pursuing one. An important quality is to explore lots of paths but also to have the ability to tunnel in and focus on one particular question. And to be persistent, not to let go of something that you think is important, keep going…. just because grants get rejected, just because experiments don't work, and just because other people don't see the point of what you're doing, if you believe in it, then keep going. Lastly, it’s important to be collegiate. I've been blessed over my life with having wonderful collaborators to work with. More should be done to teach the skills of collaboration because not everybody finds it easy. I think being trained in collegiality, and how to collaborate, especially in difficult circumstances, would be really helpful. Part of being a good scientist is to make things fun – for yourself and for others you work with. All the projects that I lead, or that I'm part of, I try to inject a sense of humour, a sense of fun. I try to have an interest in the people that I'm collaborating with, so that there's a social and enjoyable dimension to what is also often quite serious work.
How do you think behavioural science will develop (in the next 10 years)? I think there's a real danger in behavioural science being perceived as common sense. Everybody behaves, and everybody sees other people behaving, so that everyone develops their own lay ideas and models about behavior and behavior change, in other words our own common sense. This can lead people to think that we don't need more than common sense. I know I can't perform open heart surgery. I know I can't build a bridge. But everybody thinks that they know about behavior and they know how to change behavior, because, they themselves behave and they see other people behave. Now, if that were true, there wouldn't be a need for behavioral science. But it's not true. So it’s important to bring methodological rigor and all the tools of thinking that science brings, to think about behavior, and how to change it. And not to think about behavioral science as any ‘softer’ or in less need than other sciences of being totally precise about how one defines entities, relationships between them, and tests those relationships to build up theoretical understanding in addition to empirical evidence. I think a lot of what I've tried to do in my career is to try to bring that kind of rigor into thinking about behavior. What I hope is that there are people who think along similar lines as to what needs to happen to behavioral science: to make it a more rigorous science, starting with defining terms and ensuring that the highest standards of methodology, including studies and their analysis plans being pre-registered, so there's no post-hoc fishing trips going on. Another need for advancing behavioural science is to develop large collaborations so that we can get properly powered studies that can look at not only single effects, but the more interesting complex system type of questions where you can look at mechanisms of action, along with modifying variables. I hope is that by bringing more rigorous scientific thinking and developing more and larger collaboration between behavioral scientists across fields and across countries, and between behavioral scientists and other scientists. Harnessing the power of machine learning and artificial intelligence has the potential to make much more progress and be much more efficient and productive in terms of the ratio of input to output, not just in terms of scientific and theoretical understanding, but also in terms of application. I hope we can embed ourselves in much broader collaborations that will show the need for behavioral science in pretty well every area of life. The experience of being on behaviour scientific advisory committees has shown me that behavioral science can make a difference to the big issues of the day, but we need to go much further than that. We're not going to get out of this pandemic just by waiting for a vaccination or for drugs. Myself and two colleagues (Professors Jeremy Grimshaw and Paul Glasziou) have just initiated a global collaboration called BESSI, which stands for Behavioral, Environmental, Social, and Systems Interventions, bringing together people from all over the world, from many different disciplines and backgrounds, to collaborate in developing those types of interventions. That is an example of behavioral science being at the heart of a new initiative, where we are working with for example, people who know about urban planning and indoor use of space,. in relation to human behavior. To sum up I would say: more and bigger collaborations within behavioral science and between behavioral science and the whole range of other disciplines that are relevant to behavior and to do bigger and better studies. In terms of when I say better, I mean, scientifically better studies. Lastly, we need to work alongside those who are on the ground in terms of practitioners and policymakers and relevant populations and communities so that interventions are very much developed with the people for whom change is desired. Rather than scientists doing their work, and then taking the results of their work and trying to implement it, we should, from the beginning, work in partnership with a wide range of people.
Which other behavioural scientists would you love to read an interview by? The first is Marie Johnston, Professor of Health Psychology in Aberdeen. She employed me when I was a clinical psychologist after having had my first baby at 28 years old. She's a collaborator on all my main methodological science projects, so we've had a lifelong collaboration. She has a very impressive mind in terms of being both analytic but also open and good at lateral thinking. And she's also very conscious of the relationship between science and its application. The second person is Robert West, who's a Professor of Health Psychology at UCL and my husband. I knew him as a colleague for many years before we became close. He's got a very bright, analytic brain, and he's somebody who I could never tire of talking with about pretty well any subject on earth because he's always got really interesting thoughts and good insights. I hope he thinks the same about me!
Thank you so much for taking the time to write down these amazing answers Susan.
As I said before, this interview is part of a larger series which can also be found here on the blog. Make sure you don't miss any of those, nor any of the upcoming interviews!
Keep your eye on Money on the Mind!