A particularly critical aspect of intelligence, we propose, is the Context-Schema-Goal triad: a context, schema and goal that are interlinked according to the general pattern
- enactment of the schema S in the context C is likely to result in the attainment of the goal state G.
Technically, this means:
Context C PredictiveAttraction Evaluation done S G
(here done is a built-in schema which basically indicates that S has been executed on some input), or
Context C PredictiveAttraction Execution S A * G
where A is some particular input.
In OCP, sometimes a context-schema-goal triad will be represented on the Atom level, sometimes on the map level. These triads form the essence of OCP's behavioral reinforcement learning subsystem, which allows the system to learn new behaviors through experience. These behaviors include cognitive behaviors — i.e. the learning of the cognitive schemata that have been discussed frequently in previous chapters.