Skip to Content
Study Completed: 2015
College of Humanities & Social Sciences
Hierarchical Bayesian Modelling of Criterion Variance in Probabilistic Categorisation as an Analogue to Signal Detection
Signal detection theory concerns the measurement of a person''sperceptual discriminability and response bias. Inconsistencies in an observer''schoice behaviour are often overlooked and can lead to spurious conclusions regarding their actual perceptual abilities. Contextual factors, such as the presence or absence of corrective feedback, can also affect observer performance and are routinely disregarded. Mr Taylor''sresearch addressed these fundamental oversights by mathematically modelling the cognitive processes underlying observer choice behaviour. Critically, he demonstrated that choice behaviour was reliably sub-optimal and that corrective feedback does not guarantee improvements in performance; in fact, behaviour can deteriorate with feedback in some instances. Mr Taylor developed a mathematical model that could predict the interplay between corrective feedback and response inconsistency. His findings call into question the validity of using signal detection to interpret choice data unless significant modifications are made to the extant theory.
Page authorised by Web Content Manager
Last updated on Tuesday 04 April 2017