Revealing Preferences: a neuroeconometric approach to understanding decision-making behaviour

Rewards represent probably the most important objects for our life. Rewards are external items that an individual identifies as optimal to satisfy internal needs. Optimal rewards are associated with so-called ‘high reward values’. Reward values are therefore a key determinant of decision-making behaviour, the type of behaviour underlying the evaluation and acquisition of rewards. A key interest in the field of neuroeconomics is how the neuronal representation of reward value underlies this motivational behavioural. Hitherto, research in this area has examined single rewards, mostly in binary choice scenarios. This project aimed to integrate empirical laboratory studies for complex rewards, comprising of multiple components (bundles). We tested if preferences for bundles followed basic notions of revealed preference theory; preferences are only revealed at the time of choice and cannot be predicted by objective properties of a reward. Rather, preferences are subjective and independent of the bundle composition. Results from choice experiments provided empirical evidence for this formal economic theory. We investigated how psychophysically assessed revealed preferences of multi component rewards are represented by graphic ‘indifference curves’, a mathematical illustration of preferences for various bundle offers. Bundles that were equally preferred by the participants lined up on the same two-dimensional indifference curve despite varying bundle composition, thus demonstrating a characteristic trade-off between bundle components. More preferred bundles were represented by higher indifference curves. The shape and curvature varied between participants, emphasising the subjectivity of preferences. Indifference curves therefore demonstrated revealed preferences as being independent to varying bundle compositions. On a neurobiological level the results of a functional brain imaging experiment showed that the OFC represented preferences as defined by empirically obtained indifference curves. Based on the findings we reasoned that the OFC computes an economic value; that is a subjective preference for a reward, independent of its strict physical properties. This allowed us to identify a neural correlate supporting key principles of revealed preference theory. Additionally, results indicated that the vmPFC, dlPFC and ACC use information about preference and process internal states needed to coordinate reward acquisition. This processing of internal states we have likened to a cost-benefit analyses, ascribing these regions to a ‘control’ level of decision-making behaviour that serves to ‘check’ the economic values computed in the OFC. In conclusion, the novel experiments conducted in this study contributed towards developing a mathematically sound, behaviourally relevant, and neurobiologically detailed theory of choice.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2019

Author dc:creator

• Volkmann, Konstantin

Advisor dc:contributor.advisor

• Schultz, Wolfram

• Revealed Preference Theory
• Neuroeconomics
• fMRI
• Decision-making behaviour

Licence

• https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/e60cb8dc-35ca-41d4-9d73-13ccec6317f2/download
• https://www.rioxx.net/licenses/all-rights-reserved/

Language dc:language

en

DOI dc:identifier.doi

https://doi.org/10.17863/CAM.39373

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/292222