Comparing Apples to Apples: Measuring Mental Representations with Linked and Complete Markov Chain Monte Carlo
Year: 2019
Type: dissertation
Abstract: Previously, research exploring mental representations of categories has faced two challenges: 1) Preselection of stimuli by the experimenter limited full exploration of the mental representation. 2) No task was applicable across age ranges to assess a developmental trend. Here, we utilized a novel methodology in the categorization literature, Markov chain Monte Carlo (MCMC) with people. Stimuli were directly sampled from the participant’s mental representations through a series of two-alternative forced-choice questions. In our first experiment, we replicated Sanborn, Griffiths, and Shiffrin’s (2010) findings on fruit categories where participants completed full Markov chains. In our second experiment, chains were completed by linking multiple participants to reduce the number of questions asked to each person. Stimuli produced were reflective of our intuitive representation regardless of the number of participants in a chain and were similar between Experiment 1 and 2. Our results highlighted MCMC’s potential to explore the developmental trend in categorization.
Source: TSpace (University of Toronto)
Author: Kay Sarah Otsubo
Institution: University of Toronto
Language: English
Subfield: Experimental and Cognitive Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W3140423690
