Keywords: decision-making, risk, uncertainty, cognition
Understanding how humans make decisions is a central theme in the behavioural sciences, with ramifications that stretch across the disciplines of medicine, engineering, economics, and public policy. Decisions tend to be driven by multiple goals and made in the face of uncertain information (e.g., whether you should sell or buy stocks given the current economic climate). The question of how people process uncertain information has a long and distinguished pedigree, dating back to Bernoulli and extending to more recent work in behavioural economics and psychology1. This large body of work has yielded a constellation of intriguing behavioural regularities obtained “in a vacuum” i.e., in oversimplified and stylised lab-based experimental tasks which oftentimes bear limited resemblance to real-life decision-making settings2. The aim of this project is twofold. First, to examine the way humans deal with uncertainty in naturalistic tasks that maintain fundamental features of real-life decision-making. And second, to delve deeper into the cognitive mechanisms via which humans represent and process probabilistic reward information in these naturalistic tasks.
Work in my lab is focuses on understanding the way humans sample information during multiattribute decision-making. I have a long-standing expertise and track record in mapping complex decision behaviours onto fundamental cognitive and neural processes3 . To achieve so, I utilise neurophysiological methods and devise bespoke experimental paradigms, which maintain key features of real-life decisions while also permitting the precise tracking of the information flow the cognitive system is processing at each moment in time4. A similar approach will be adopted in this project. Outcomes of this project will have both basic science and clinical relevance (e.g., how does decision-making under uncertainty goes awry in neuropsychiatric disease?).
The successful candidate will have demonstratable quantitative skills and good knowledge of probability theory. The student will be trained in eye-tracking and pupillometry, and in advanced computational modelling and statistical methods. Depending on the student’s interests, engagement with neuroscientific techniques (EEG, TMS) as well as secondary analysis of MEG and pharmacological datasets is possible. The student will enjoy close collaborative links with research groups situated in the School of Psychological Science, within the Bristol Neuroscience network, nationally and internationally.