Sequence memories and their integration for planning: A spiking neural network model

We propose a biologically-inspired auto/hetero-associative spiking neural network combined with a working memory model, in which a state-driven forward sequence and a goal-driven backward sequence on the associative network are integrated in the working memory to make a plan. By discrete pulse-driven neural network simulations, we show that several characteristics of planning process such as goal-directed attention control at a branch point of a plan, incremental planning, and planning by combining episodes can be realized in our system.