What is it? Where is it headed?
Behaviour Science in India: What is it? Where is it headed?
Dilip Soman speaks with Divyani Diddi
“Behavioural Science” in India has grown and diversified out of various other applied fields in the last decade of its existence in India, in its current form. Of course, long before we began labelling there were researchers using Randomised Control Trials to ascertain the optimum scale and impact of welfare programmes, along with a highly intuitive and creative advertising and consumer marketing industry with a finger on the pulse of this giant, complex and confounding nation, among other initiatives that applied behavioural insights without much thought on theory or systems. But with a global awakening to the power of behaviourally informed programmes, policies, projects and products.
India too began compiling and streamlining its efforts to apply the knowledge of behavioural principles to achieve its various development and transformation objectives. As we enter an era of ubiquitous adoption of behavioural science across fields and disciplines, the question of what it is, what is still required and where it's headed arises. For this, we turn to Dilip Soman - a Behavioural Science practitioner turned academic, dedicated to exploring and explaining Behaviour Science in the Wild with some questions.
1. You have previously spoken about the different incentive structures for academics and practitioners in Behavioural Science and how it impacts each of these approaches manifesting in the way that the field has developed. How have these incentive structures evolved over time, and how do you see them reconciling as both sets of people increasingly collaborate with each other?
I'll start by repeating the claim that the incentive structures for practitioners and academics are completely different. There are the obvious things, things such as differences in time pressure for completing projects, and the freedom to choose topics to work on which academics have that practitioners don't. Beyond the more obvious ones, I think there's a deeper difference which is the fact that academics build a reputation by specializing in a theory or a particular phenomenon. For example, most of my early work is on mental accounting and it doesn't matter whether it's in the domain of saving or retirement or how people spend windfall gains, but it's all mental accounting, and so that's how academics make their reputation. They do it by specializing in a theory that applies to different kinds of domains, whereas the practitioner is more interested in the domain and the solution to a specific problem. They don't care about mental accounting or budgeting as much as they care about financial wellness. I think the problem is that practitioners are looking for, what I call, first-generation, first-degree solutions, "If I want to improve financial wellness, should I engage in financial literacy or should I help people budget?" Those are the kinds of big questions which academics usually don't have the tools to answer because of their narrow focus. I think the first set of challenges, the obvious ones, the time pressure, the selection of problems, I think those are the ones that can be mitigated with time. We've seen a lot of success stories of academics now in practice or practitioners working with academics, but I think the deeper problem, the fact that reputations are built in different ways, I think that's the one that's really going to be hard to reconcile over time. I don't think we'll ever work in a world where academics and practitioners are on an equal footing. I think each one of them has a different role to play, and I think we just have to find the best equilibrium as to what those roles are.
2. In my limited understanding, one of the differences between these approaches is the difference between looking back (a look at what has worked and why - an emphasis on documentation) and looking forward (predictive modelling). With reference to finding the answer to a problem, what do you think is the optimal balance between these two perspectives?
I'm not convinced that looking back versus looking forward is the right way to capture the distinction between academia and practice, even within academia and within the practice, you find people that look back versus look forward. I don't think that's the trick, but I think that the question of how we reconcile these approaches is an important one. Looking back, learning from history, and building models based on history, I think is great in a stable environment, but when the environment changes, when there are a lot of uncertainties, when we're taking learnings from one context and translating them to the other or whether it's just inherent uncertainty in the way the environment is going to shape up as, for example, we've seen with the COVID pandemic, then I think we need to focus more on looking forward in building predictive models than looking at history. There's a recent book by Robin Hogarth and Emre Soye which talks about the pitfalls of experience and I think they make the point that, again, if the world is stable, then experience which is a collection of insights that you get from looking back is good in a stable world, but when the world is unstable, when the situation is changing, then, in fact, it backfires. I think the trick is really figuring out how stable the environment is. If it's a stable environment, then I think it's perfectly fine to look back, figure out what works, and then apply that to the new situation, but if the world is unstable, then I think you can't blindly apply what worked in the past, it's important to test. I think it's really the testing that I think serves as the bridge, so I think history gives us great hypotheses, but I think it's really important for us to test on a small scale before we deploy those interventions or those ideas going forward.
3. Do you think there exists uniformity in the way that Behaviour Science is thought about or approached across its application spectrum? Especially with respect to Public Policy in the Indian context, do you think this form of thinking influences its application, for better or worse?
The question about uniformity is an interesting one because I have to confess that I'm not convinced that uniformity is really needed or, in fact, that it's a good thing. What I mean by that is different policy areas are different in terms of how much you can experiment, for example, or how much access you have to data about the end-user. Take the example of something like privacy policies or financial disclosure policies. These are not things that you can run randomized controlled trials with, whereas if you look at things like welfare programs, or thinking about how to encourage take-up of healthy behaviors, or diets, those are things where you can experiment and where you can get access to data. Actually, I don't think uniformity in how behavioral science is applied to policy areas is actually a good thing. I think it's important for us to understand that policy areas are different and that they should be treated differently in terms of how behavioral science informs them.
4. Your online course has been one of the most popular for learners in India - giving you a unique insight into how people feel about and experience this field. What have you seen change in successive cohorts - in terms of level of engagement, the kind of problem statements that they work on, etc.
The online course has been around now for, gosh, seven or eight years and it's changed the flavor. I think at the beginning, the course was more about exposure. The book Nudge had just been written. People were just at that point in time beginning to be aware of this animal called behavioral economics and so initially the course was more about presenting the idea, showing people the power of this new science. I think that's what we saw in terms of the work done by the cohorts. I think they were more interesting counterintuitive, flashy kinds of ideas but I think over the seven, eight years that the course has run, I think the field has become a lot more mature, a lot more stable. I'd say when the course started, the field was in the growth phase, now it's in the maturity phase, and in the maturity phase, we need to think more about questions like translation and scaling and when does something work, and when it doesn't work. I think the course too has evolved along those lines, has a lot more nuance now. It's not about just the fact. I think the idea that behavioral science can help is now established. I don't think we need to push that idea, but giving people a more nuanced understanding of when and how, and what's the process for making it work for you? I think that's what the course has developed to being all about, and I think over the last many years, you can see the evolution in terms of the kind of projects people work on, the essays they write, the self-reflection pieces. You can see that maturity that now it's no more about just cheerleading, but it's a lot more about application and thoughtfulness in terms of the more practical details.
5. What is a field in India that you think has gained the most from applying behavioural science insights into its work? What is a field that has potential and is currently not utilising all that the field has to offer?
In India, I think there's been a lot of great work. I have to confess that a lot of the amazing work that's being done in India over the past many years has been done by scholars physically located outside of India, but I think it's gratifying to see a lot more now being done by Indian scholars. In that domain, I think one of the biggest advances has been work on financial well-being. Things like how do we help people in poverty save more? How do we help people's smooth incomes across variable, seasonal consumption patterns? That sort of stuff. I think there's been a fair bit of work in that area, but what I'm really heartened by with the work being done in India is just the breadth of the application. We've seen examples of people in the sanitation area using behavioral science, in safety. That classic project about how do we safely help people cross railway lines? That's a cool one. Obviously, we've got a lot of work in the area of health, in the area of financial well-being. I think that breadth is really nice. I do wish we do a lot more work at the so-called bottom of the pyramid. Can we help the underprivileged sections of our communities, the poorer sections of our communities, the oppressed sections of our communities with behavioral science, be through education, be through creating more equitable social network. Raghuram Rajan has written a fantastic book called The Third Pillar, where he talks about the fact that for capitalism to succeed, you need a strong, active community, and I think behavioral science has a lot to offer to the community-building aspect. I wish we see a lot more of that happening in India. Of all the places in the world, I think India has a lot to offer the world of behavioral science on equity, fairness, inclusion, justice, all of those topics.
6. What are some of the things that courses, modules and lectures are not adequately preparing future practitioners (current enthusiasts) for, with respect to the application of behavioural science?
By its very nature, our course is really focused on a collection of content pieces, and I think most of our courses do a really good job of the content, but we know from behavioral science that the content, the theory, the principles are relevant to the extent that the context makes them relevant. The same idea, the same phenomena can operate differently in different contexts. I think that's the part of the process where the learner has to make the connection, the learner has to take the theory and say, "Ah, this works well in the context of a small village in Maharashtra," for example, or, "This will not work well in a large city in South India." Things like that. I think the course content is always standard. I think it should be standard, but good courses communicate the fact about this nuance that I spoke about, communicate the fact that we can actually come up with a slightly more nuanced view of the theory depending on the context in which that theory is being applied.
7. What are some of the non-negotiable skills and practices that future practitioners ought to keep in mind when beginning their journey as Behavioural Scientists?
What are the non-negotiable skills? Gosh, that's a tough question. A good level of curiosity, I think. I guess it's not a skill, but it's a prerequisite. Why do people do what they do? Is a question that drives most behavioral scientists. Why do they not do what we think they should do? Is another question that drives most behavioral scientists. Just having that mindset of trying to explore what motivates people, what prevents people, what are the frictions? I think that's an important skill, if you will, for applied behavioral scientists. In terms of specific tools, obviously, a good understanding of the science itself, the methods of the behavioral scientists, how to do experiments, how to run trials, how to do quasi-experiments. These are important skills because without those skills one will always remain a lithe theorist and not an evidence-based practitioner. I think in order to convert that interest, that promise of behavioral science into actual science, I think it's really important to understand the methods. Can you design an experiment? Can you analyze an experiment dataset or a randomized control trial? Can you think through quasi-experimental methods of taking a data set and looking at variations within? I think that is the other key one. Over time, I think big data is going to play a bigger role in behavioral science. If you look at the way our science is done, we have a very scientific way of testing hypotheses, but we don't have a very scientific way of generating hypotheses. I think that's where tools like machine learning can play a big role by looking at patterns of data and identifying what we call empirical regularities or generalizations. Those could be the starting points of hypotheses for the scientific process. I'd say machine learning, big data would be another set of skills that I think we should be developing as a community.
Key Takeaways
- Context is Key. It defines the stability of the environment within which we function, identifying the optimum balance of learning from the past, and developing predictive solutions through iterative testing.
- Customisation based on context is the only way to develop targeted and impactful solutions to specific challenges.
- Practice is a critical supplement to theoretical knowledge for any budding behavioural scientist. Beginning with a sense of curiosity, one must build a strong foundation of running experiments and trials to assess the validity of intuitive and learned knowledge. This requires frequent application of concepts and phenomena to real life examples, and allowing a healthy space to learn from being wrong.
- While the field has developed from establishing its credibility as a reliable source of insights to various domains and industries, it has gained momentum and spread its breadth of application to these fields.
- Incentive structures based on the foundation of the reputation of academics and practitioners is and may always be different. Given the recent success of individuals from both fields working with each other, a happy equilibrium for both to work together does not seem to be too far behind, with positive implications for the development and implementation of behaviourally informed interventions.
Dilip Soman is the Canada Research Chair in Behavioural Science and Economics, and serves as a Director of the Behavioural Economics in Action Research Centre at Rotman [BEAR]. His research is in the area of behavioural science and its applications to consumer wellbeing, marketing and policy. He is the author of "The Last Mile" [University of Toronto Press] and teaches a massive open online course [MOOC] "BE101X: Behavioural Economics in Action" on EdX. He can be found on Twitter here: @DilipSoman
*The views and opinions contained in the interview belong solely to the individual interviewee and do not necessarily reflect the views and opinions of any individual or institution mentioned in the piece.