A Simple Diffusion Model for Psychometric Analyses

The emergence of computer-based assessments has made response times, in addition to response accuracies, available as a source of information about test takers’ latent abilities. The predominant approach to jointly account for response times and accuracies are statistical models. Substantive approaches such as the diffusion model, on the other hand, have been slow to gain traction due to their unwieldy functional form. In the present work we show how a single simplifying assumption yields a highly tractable diffusion model. This simple diffusion model is straightforward to analyse using Gibbs sampling and can be readily extended with a latent regression framework. We demonstrate the superior computational efficiency of our model compared to the standard diffusion model in a simulation study and showcase the theoretical merit of our model in an example application.