Affect and Decision Making
: Disentangling Underlying Processes

Student thesis: Doctoral ThesisDoctor of Philosophy (PhD)

Abstract

Judgement bias, which examines decision-making under ambiguity, is a promising measure of affect in non-human animals. However, there are significant gaps in our understanding of the cognitive processes underlying judgement bias and the adaptive significance of affect-induced shifts in decision-making, which we aimed to resolve.
To this end, we investigated the relationship between reward and punisher experience, judgement bias, and affective state in rats and humans. Firstly, we conducted a systematic review and meta-analysis of pharmacological manipulations of judgement bias which demonstrated that pharmacological manipulations designed to alter affect overall influenced judgement bias in the predicted direction, hence supporting the validity of judgement bias as a measure of affect (Chapter 2). Then, we combined behavioural experiments with computational modelling, which permitted a deeper insight into the decision-making processes underlying behaviour. Specifically, we examined the influence of fluctuating primary (Chapter 3) and secondary (Chapter 4) rewards and punishers on judgement bias and self-reported affect in humans. Additionally, we manipulated rats’ reward and punisher experience either prior to (Chapter 5) or within (Chapter 6) a judgement bias test session. Data from the rat and human judgement bias studies using a within-test manipulation (Chapters 3, 4, and 6) were
analysed using a novel computational model in which we defined decision-making as a partially observable Markov decision process. We found that reward and punisher experience, particularly the predictability (in humans) and absolute favourability (in rats) of these experiences, altered judgement bias primarily by influencing reward or punisher sensitivity. Humans that reported more positively valenced affect exhibited greater predictability-dependent modulation of judgement bias. Finally, we assessed the extent to which the explore/exploit trade-off could provide a novel measure of affect and found that inducing a putatively negative affective state increased the rats’ tendency to explore, rather than exploit a food source (Chapter 7).
Date of Award12 May 2020
Original languageEnglish
Awarding Institution
  • The University of Bristol
SupervisorIain D Gilchrist (Supervisor) & Michael T Mendl (Supervisor)

Cite this

Affect and Decision Making: Disentangling Underlying Processes
Neville, V. M. (Author). 12 May 2020

Student thesis: Doctoral ThesisDoctor of Philosophy (PhD)