Computational mechanisms of irrational decisions

Hosted by the School of Psychological Science

The Tsetsos Irrationality Lab; we aim to understand how humans process information when making decisions, especially when these decisions deviate from the normative standards.

Abstract: Imagine that you are about to choose a large house with a long commute over a small flat with a short train ride. Will the appearance of an irrelevant alternative–a small flat with a longer train ride–make you revise your initial preference? A large body of research suggests that you will, more likely than not, switch to the small flat with the short train ride. Changes of heart like that, dubbed preference reversals, defy the basic tenet of rational choice theory, according to which reward-maximizing choices should stem from stable and “menu-invariant” preferences. What do these irrational behavioural patterns tell us about the cognitive and neural processes that underlie preference formation? And why have these processes persisted in the face of evolutionary pressure for reward-maximizing choices? In this talk, I will provide answers to these questions by firstly delineating the mechanisms that give rise to preference reversals, and by secondly showing that these mechanisms are in fact reward-maximising in light of the simple fact that neural information processing is noisy. 

Bio: Dr Konstantinos Tsetsos is a cognitive scientist trying to understand how people make decisions. Following his undergraduate degree in Computer Science he completed a PhD in Cognitive Science at UCL. After pursuing postdoctoral research in cognitive neuroscience at the University of Oxford, he held a Welcome Trust Fellowship at Birkbeck, followed by a Marie Curie Fellowship at the University of Hamburg. While in Hamburg, Konstantinos started his own “Irrationality” lab funded by an ERC Starting Grant. He is now a Senior Lecturer at the School of Psychological Science (University of Bristol).

Contact information

Enquiries to psych-school@bristol.ac.uk