Stimulus-Stimulus Associations and Parallel Search
The S-R system developed so far is capable of assembling action sequences in order to achieve a final goal. However, there are other approaches to assembling action sequences which have quite different time/space characteristics. The best developed alternative to the S-R model is the stimulus-stimulus (S-S) model. In the S-S model the learned associations are between stimuli, rather than between stimulus and response (and stimulus and evaluation). In this chapter, two S-S models of “planning” by parallel search are developed. The first searches backward from the goal state looking for a sequence of states that connect the goal state with the current state. The second searches forward from the current state looking for a path to the goal state.
KeywordsGoal Node Shared Memory Travel Salesman Problem Goal State Conjunction Node
Unable to display preview. Download preview PDF.