Exploring the Valleys of the Applied Behavioural Science Landscape

This is part two of a series on the future of applied behavioural science. You can read part one here.

Revisiting the Core Challenge

Solutions to challenges solved in isolation inevitably stumble as they enter the messy, complex reality that they need to exist within

In part one of this series, I explored the technical and ethical limitations to innovation in the field of applied behavioural science. Each of these limitations was discussed in-depth, alongside opportunities for overcoming them. I then made the claim that although many of the identified limitations have clear solutions, these solutions don’t take into consideration the other constraining forces of the field, and therefore get stuck in the early stages of the innovation process.

The most obvious example of this is heterogeneity. One of the field’s biggest challenges is overcoming the individual-level differences in preferences and responsiveness to interventions. We’re realising that a one-size fits all approach isn’t the most effective one, and in some cases, it can actually do more harm than good. In overcoming this limitation, data-driven personalisation is the suggested way forward. This seems like a no-brainer. Although it may help to resolve the heterogeneity issues, it requires a large amount of individual-level data to be effective and strains the ethical components of the field (privacy, data security, third-party tracking). These ethical concerns aren’t just theoretical limitations; many people seem to have strong psychological sensitivities to how their data is collected, stored and used. And, as the data gets more personal, granular and complicated, more of this sensitivity arises. This is very much like the sort of balancing feedback loop that we see in a natural system of complex, dynamic relationships.

Unfortunately, it isn’t just the solutions to the heterogeneity challenge that seem constrained. Across almost all the identified limitations, there are examples of proposed solutions that make sense in isolation, but when you step back and look at them in relation to one another, optimism in their potential tends to wane.

A Way Forward

Sometimes you need to go into the valleys in search of paths to higher peaks

To unlock the vast remaining potential that exists on the applied behavioural science landscape, we need to find solutions that accommodate all of the identified limitations, as opposed to narrowly overcoming them individually in isolation, and therefore at the expense of other factors. This requires the exploration of counter-intuitive, and perhaps even seemingly obscure paths. Paths that diverge from the tried and tested ones. Paths that may seem to make no sense in the short run, in the hope that they’ll seem obvious over the longer run. Many of these paths will lead to dead-ends, or loop back to where they started. But without taking the risk to venture down these paths, progress in applied behavioural science will be limited to diminishing optimisations and horizontal adoption of what already exists. It could be so much more than this. Well, that’s my bet anyway.

Promising Paths

Leaders in the field have already started exploring new paths and hoping that, over the course of the decade, many others will do the same. I have tried to keep an eye out for examples of promising paths that I think will be worth getting involved in, or at least following attentively. These include the following:

• Rethinking the way Governments approach Decision Making (e.g. Michael Hallsworth & Jill Rutter’s Behavioural Government work)
• Building BI into the Fabric of Organisations (e.g. Melanie Kim and Dilip Soman’s BI Org work)
• Building Bridges Between Researchers & Practitioners (e.g. William Mailer work in bridging academics and practitioners at CBA)
• Using Collective Intelligence to solve urgent crises as they arise (e.g. work being done by Liam Delaney and Emma Watson in the fight against COVID)
• Rethinking the Role of Scientific Journals (e.g. Susan Michie and Marie Johnston’s HBCP work to create knowledge systems that leverage ML/NLP)
• Behaviourally-informed Coaching & Therapies (e.g. Rishi Mandal’s work on the digital coaching platform ‘Future’)
• Self-applied Behavioural Science (see below)

Self-applied Behavioural Science

One way to overcome the technical and ethical limitations is by giving individuals the tools to change themselves

The path that I have focused on over the past year is self-applied behavioural science. The goal here is to find ways to enable individuals to use the methods, insights and evidence-driven techniques of behavioural science, in service of improving their decisions, and acting on their intentions. Currently, there are a few important research papers, several published books and a handful of digital tools that aim at solving this challenge directly. My sense is that there is growing interest in this area and that the potential of impactful behaviour change is enormous. I also think an interesting opportunity may open for citizen ←→ scientist data partnerships that could overcome some of the technical and ethical limitations, that were discussed in the previous essay.

For those unfamiliar with this arm of behavioural science, here are some introductory resources:

Academic Papers Worth Reviewing:

1. Beyond Willpower: Strategies for Reducing Failures of Self-Control ~ Angela Duckworth, et al. (2019)
2. Nudging & Boosting: Steering or Empowering Good Decisions ~ Ralph Hertwig, et al. (2017)
3. Self-nudging & the Citizen Choice Architect ~ Samuli Reijula, et al. (2020)

Books Worth Reading:

1. Think Small ~ Owain Service & Rory Gallagher
2. How to Have a Good Day ~ Caroline Webb
3. Indistractable ~ Nir Eyal
4. Good Habits, Bad Habits ~ Wendy Wood
5. Tiny Habits ~ BJ Fogg
6. Behave ~ Robert Sapolsky
7. The Book of Why ~ Judea Pearl
8. Thinking in Systems ~ Donella Meadows
9. Spiritual Design ~ Steven Wendel

Digital Tools Worth Exploring:

1. Stickk
2. ReWi
3. Fabulous
4. RescueTime
5. BCTs
6. Streaks
7. Levels
8. Screen Time
9. Sleep Cycle

The Missing Piece

Moving from interesting ideas to everyday practices

Like many others in the field, I’ve devoured the literature on this topic as it has emerged. In doing so, I realised that, although the self-orientated frame of behavioural insights is interesting, putting this knowledge into practice is difficult. Like with many problems these days, there is an abundance of information, but a lack of implementation. The familiar old intention-action gap shows itself once again.

This is where I thought my background experience, interests and expertise may be useful. To solve the implementation issues associated with self-applied behavioural science for myself, I decided to try figure out the psychological, social and physical barriers, to explore the identity affects, and the glitchy mental models that I am holding. I then used these learnings to build out a framework that I piloted as part of two private programmes to test and iterate on. There is still a long way to go but I am feeling excited about the progress I’ve made and the feedback I received from the initial pilot groups.

Some of the big friction points to self-applied behavioural science I’ve identified so far:

1. Faulty mental models about the self, the mind and the brain-body loop.
2. Overestimating the importance of willpower and underestimating the importance of social and situational factors.
3. Mistakenly discounting the value of setting and holding structured mental representations (narrowly defined goals, implementation plans, etc.).
4. Overestimating the value of big life events, and underestimating trivial actions we take every day.
5. An overextension of agency and ownership regarding the sources of our goals, desires and needs.
6. An overreliance on memory for purposes of activity recall, and a general overconfidence inability to consciously recall intentions and motivations.
7. Forecasting failures when it comes to future energy levels, time affluence and enjoyment of particular experiences.
8. The necessary capacity to do the sort of reflection and design thinking required for building self-deployed interventions and self-experimentation.
9. A structure for seamlessly bringing the core ideas of self-applied behavioural science into one’s everyday life.
10. A framework for moving between the different modes required for self-applied behavioural science (exploratory, design, action and reflective modes, etc.).

I’m nowhere close to solving all of these friction points, but what I have seen is that a structured programme centred around active and social learning exercises can move the dial. I have also seen the value of community learning spaces, especially when complementary norms and feedback systems are present. Annie Duke’s decision pods concept is an inspiring example of what is possible here.

I’ve also got Kahneman’s famous quote glued to the back of my eyelids ~ “So this is my aim for watercooler conversations: improve the ability to identify and understand errors of judgment and choice, in others and eventually in ourselves, by providing a richer and more precise language to discuss them. In at least some cases, an accurate diagnosis may suggest an intervention to limit the damage that bad judgments and choices often cause”.

My Focus Going Forward

Online programmes, virtual communities, open-source toolkits and public conversations

It seems to me that investing my energy and attention in building and refining an online learning programme for self-applied behavioural science, complemented by a virtual community space, is a useful addition to the great work already done to build out this arm of the field.

As a result, at the beginning of June, I will be launching the public version of my previously mentioned pilot programme. I will also be deliberately growing and enriching an intimate virtual community space I have developed for ‘watercooler-style’ discussion on this topic and providing access to useful digital support tools there too. Finally, I am also starting an initiative to have public conversations on this topic, to continue learning and get others engaging with the key ideas.

Whether or not this path leads us out of the foggy valleys and to higher peaks is yet to be determined, but it seems like a worthwhile venture either way.

If you’re interested in participating in the June Public Pilot Programme you can access more info here:

https://citbesci.substack.com/p/welcome-to-circles-in-time-public