< Back to Article

Model-based spatial navigation in the hippocampus-ventral striatum circuit: A computational analysis

Fig 8

Shallow control mechanism.

It exploits the state-transition and state-value models and local maximization to predict the expected value of each action primitive aⁱ and select the most valuable one. For each action primitive, the mechanism first finds the latent state that most likely would be achieved by applying that action and then finds among those states the predictably most valuable one. The action bringing to that state is the selected one: a_t = argmax_a P_v(r = 1|g,argmax_s′P_M(s′|s_t,a)).

doi: https://doi.org/10.1371/journal.pcbi.1006316.g008