A local diffusion model (Staddon and Reid, 1990) can reproduce exponen
tial and Gaussian stimulus-generalization gradients. We show that a tw
o-dimensional diffusion model, together with simple reinforcement assu
mptions, can reproduce many of the empirical properties of goal-direct
ed spatial search, including area-restricted search, open-field foragi
ng, barrier and detour problems, maze learning and spatial ''insight.'
' The model provides a simple, associationistic ''reader'' for Tolman'
s cognitive map. (C) Elsevier Science Inc. 1997.