Figure 2.

Decision making in a diffusion model. In diffusion models, a particle drifts and diffuses between two absorbing boundaries. A decision is triggered as soon as the particle hits either boundary. The drift rate is proportional to the reliability of the mean of the momentary evidence, while the diffusion represents its fluctuation. A correct decision corresponds to hitting the boundary towards which the particle drifts – in this example the upper bound. The figure shows various example trajectories leading to different decision times, one of which causing the decision to be incorrect (i.e. hitting the lower boundary).