Australasian Mathematical Psychology Conference 2019

I don’t know… how to model this.

Angus Reynolds
Andrew Heathcote
University of Tasmania
Peter Kvam
Ohio State University
Adam Osth
University of Melbourne

We present the Multiple Threshold Race (MTR)- a framework to extend simple evidence accumulation models such as the Linear Ballistic Accumulator to judgments on a continuum rather than simple definitive response. The simplest example is to include a “Don’t Know” response as well as two definitive responses. We ran two facial recognition experiments: a two-alternative forced choice and a single item new/old identification task. In both experiments subjects could respond `Don’t Know’ as well as Left/Right or New/Old.

This talk uses the results of these experiments to outline the conception of an MTR model, demonstrating how we can include additional response options by incorporating a `Balance of Evidence’ (Vickers, 1979, 2001) like idea into our current evidence accumulation models. Additionally we present fits to the experiments from the original RTCON and RTCON2 papers (Ratcliff & Starns, 2009, 2013). Finally we discuss different variants of the model and applications of the MTR beyond confidence experiments.

Ratcliff, R. & Starns, J. J. (2009). Modeling confidence and response time in recognition memory. Psychological Review, 116(1), 59--83.
Ratcliff, R. & Starns, J. J. (2013). Modeling confidence judgments, response times, and multiple choices in decision making: Recognition memory and motion discrimination. Psychological Review, 120(3), 697--719.
Vickers, D. (1979). Decision processes in visual perception. Academic Press.
Vickers, D. (2001). Where does the balance of evidence lie with respect to confidence? In E. Sommerfeld, R. Kompass, & T. Lachmann (Eds.), Proceedings of the 17th Annual Meeting of the International Society for Psychophysics (pp. 148--153).