New publication examines the analysis of decisions

The mathematical drift-diffusion model (DDM) has been very successful in explaining behavioral and neurophysiological data. The model assumes that sensory information accumulates over time until this accumulated information is so clearly in favor of one alternative that a decision can be made - for example, which of two light signals is brighter. However, it has limitations when it comes to capturing the variability of consecutively executed decisions or the effects of internal factors on decision making.

Together with Université Paris-Saclay , researchers in BTU's Computational Neuroscience department propose a new model, a nonlinear version of the drift-diffusion model (nl-DDM), that solves these problems. "In the publication, we show that the nonlinear model performs better and captures time effects better than the original drift-diffusion model at the same complexity," says Prof. Dr.-Ing. Glasauer. "Our model paves the way to a more accurate analysis of perceptual decisions and also takes into account influences from the time before the presentation of the decision task."

About the scientist

Prof. Dr.-Ing. Stefan Glasauer conducts research on the principles and mechanisms of sensorimotor skills, perception, spatial orientation and navigation in humans and animals.