A Type-2 signal detection analysis of gambling behaviour: cognitions, metacognitions, expertise and optimality

Abstract

Cognitive gambling research has focused mainly on the irrational beliefs and cognitive biases that differentiate problem (PGs) and non-problem (RGs) gamblers. Whilst this research has been informative by highlighting that greater irrational beliefs are associated with gambling severity, the research has failed to determine cause and effect. This thesis proposes that metacognition is an area that may play a central role in the development and/or maintenance of problem gambling. Type-2 Signal Detection Theory (SDT) was used to analyse the data to measure three cognitive and metacognitive components of gambling performance: accuracy, resolution (metacognitive monitoring) and gambling criterion (metacognitive control). Optimality of gambling decisions was also explored. Experiment 1 used a simplified blackjack task, which demonstrated resolution differences between non-gamblers(NGs) and RGs. Experiments 2 to 5 examined the transference of gambling expertise of RGs and NGs in a novel dice gambling task. Experiment 6 demonstrated that the type of task can account for some cognitive and metacognitive variation observed between PGs and RGs, but impaired gambling criterion setting is a pertinent component of PGs’ gambling performance that is not dependent on gambling task. Finally, Experiment 7 showed that feedback enables participants to effectively shift gambling criteria to a more optimal position - and may have considerable implications for the treatment of problem gambling. The results are discussed in relation to four specific research questions and underscore the relative contribution of using a SDT approach in the study of gambling behaviour.

More information

Identifiers

ORCID for Philip A. Higham: orcid.org/0000-0001-6087-7224

Catalogue record

Export record