Predicting the Post-Covid World with Neuroscience: a UCL BIT Seminar
Of a few given snack products, which is going to be the most successful in the market? When will people start eating out again after a second lockdown? How will a furlough scheme affect job-seeking and unemployment? Leigh Caldwell, our seminar speaker on October 22nd, gets asked these kinds of questions on a daily basis. What is interesting, however, is that he uses neuroscience to inform his predictions.
Leigh Caldwell was initially trained as a mathematician, and worked in the 90s on the development of early internet technology. Today, he is a cognitive economist. He does research on how we simulate possible worlds in our heads and predict the good or bad things that may come from them. That’s the “cognitive” part. As for the “economist” in his title, he also studies how mental simulations influence economic decision-making and behaviour in the market. Leigh is author of The Psychology of Price, creator of the Cognitive Economics Society and founder of Irrational Agency, a behavioural research agency that aims to uncover some of the unconscious factors that drive the majority of consumers’ behaviour.
It is quite natural for business to ask questions about the coming days, months and years. But the future is hard to predict, and even more so in 2020. There are at least three approaches to forecasting in business. The traditional way is to simply draw inferences from the past. A more dynamic approach would be to extrapolate trends to forecast what change we are about to see next. Instead, Leigh tries to solve uncertainty in the corporate world with the tools of neuroscience research. Specifically, he builds models and uses them to make predictions given a set of starting points. Leigh says one of the most incredible abilities of human beings is the capacity to think about the consequences of their actions beyond the present moment. Over the last 10–20 years, the idea of the brain as a prediction machine has become an important consensus in cognitive science and neuroscience. However, we don’t just predict: we also update our models of the world based on the outcomes we observe, and use the increasingly accurate simulations to evaluate how we would feel if they came to realization (a process called the “how do I feel” heuristic). Leigh put all these ideas together and developed what he calls “System 3”, adding on to Daniel Kahneman’s System 1 and System 2. If System 1 refers to the more intuitive, unconscious, automatic mind and System 2 is the calculating, deliberative mind, then an additional System 3 is needed to cover mental stimulation and the generation of synthetic rewards as an outcome. System 3 is not automatic nor unconscious, but it’s also somewhat illogical, imaginative and creative.
Leigh’s research operates on System 3 and is designed to guide people in creating narratives for the future — for instance, around the question of what people want to do once the lockdown is over. He sets out to measure not only what people expect, plan and hope, but also how they will feel in different possible scenarios. He uses this kind of information to build “implication maps” like the one shown below.
Using these kinds of maps, Leigh and his team at the Irrational Agency tackle all sorts of questions in the business world. For example, they found out that perceived scarcity increments people’s desire for a product: so, snack companies could increase demand by limiting the available amount of packages of crisps available per person. In terms of communication, linking ideas of beauty and openness to a product will operate on people’s “how do I feel” heuristic, making them more likely to select it in a store. Using similar methods, they were able to predict that most people would be wary of pubs and restaurants in the first few days after reopening, with a majority returning after a couple of weeks, while some people plan not to go at all until the virus has been completely eliminated. By generating a range of projections on the progress of COVID-19, they also modelled the expected change in GDP in the next few months — unfortunately, not a rosy forecast.
After a great presentation where he showed us a blend of research and practice, Leigh was kind enough to let us ask him a few questions, some of which are shared below:
You said System 3 is not automatic, but it’s also not completely logical. So how can you predict what people will mentally simulate on?
Well, the idea behind implication maps is to predict exactly that: I would expect people to think most often about the things that are most frequently found in those maps and that are well connected to other concepts. But obviously, rewards take an important role here as well, so that the more rewarding the content, the more likely it is to be stimulated. In addition, I think the brain is always on the lookout for actions it could take (this is back to that predictive brain idea) so that simulations are partly triggered by environmental context. Finally, there are theories that support the idea that we simulate where there is uncertainty, so that we can try and extrapolate the most likely outcome and prepare ourselves for what’s to come.
How do you combine research and practice in your day-to-day life? What proportion of your time goes to either activity?
That very much depends on what specific projects are going on. For instance, there was a period this summer when I had the chance to do more theoretical work. But perhaps because of the overarching uncertainty of the last few months, our business has been in high demand recently, so I had to focus a lot on delivering projects for our clients. In general, I try to split it in a fair 50/50, and relying on my team has been really helpful in achieving this balance.
From the consumer’s perspective, how can we take into account System 3 to make smarter decisions?
Ah, it would probably take a whole other seminar to answer that question! But what I try to do is really to enjoy System 3. One of the classic effects has to do with misattribution. For instance, you’re in the store and you’re going to buy some food, and you imagine how hungry you’re going to be when you eat that food. But actually, you’re being influenced by how hungry you are in the moment; and if you’re really hungry, you’re going to buy too much stuff and it’s probably going to be unhealthy, but if you go to the store when you’re less hungry, or if you recognize that you’re hungry and you’re aware of it, you’re more likely to make a better decision because you can disconnect the current feeling from the anticipated future feeling. But in more general terms, you get a lot of value out of your own imagination and out of the pleasures and the experiences that you anticipate. A classic one is a holiday. Many of us may not have been able to travel much in the recent months, but you can still think about the holiday you’re going to take next year, where you’re going to go, who you’ll meet, what food you’ll eat and who you’ll be with.
We thank Leigh for being such an interesting and kind speaker, and the audience for truly engaging with this fascinating seminar.
Further resources:
- Leigh Caldwell’s TEDx talk explores the creation of a virtual world for understanding and solving real-life social problems.
- His own blog post about System 3.
- Clark, A. (2013). Whatever next? Predictive brains, situated agents, and the future of cognitive science. Behavioral and brain sciences, 36(3), 181–204. https://www.sciencedirect.com/science/article/pii/S0010027714002169
- Momennejad, I., Russek, E. M., Cheong, J. H., Botvinick, M. M., Daw, N. D., & Gershman, S. J. (2017). The successor representation in human reinforcement learning. Nature Human Behaviour, 1(9), 680–692. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6941356/
- Chater, N. (2015). Can cognitive science create a cognitive economics? Cognition, 135, 52–55. http://www.academia.edu/download/40419499/Cognitive_economics_revised.docx
Stay tuned for more exciting events from the UCL Behavioural Innovations Team!
Twitter: @UCLBIT
FB: UCL Behavioural Innovations Team
LinkedIn: UCL Behavioural Innovations Team
Instagram: @ucl.bit
