ABSTRACT
The assumption that cognition implies purpose was one of the main reasons for the development of stimulus–response learning theory, because behaviourism was certainly intent on being deterministic, and in consequence held that there were no cognitive processes. Now that the pointlessness of trying to deny the reality or behavioural import of cognitive processes has become plain, and their role in learning is again a subject of study, it becomes necessary to dissociate cognition from purpose, and to show what view of the nature of cognition is required if there is to be viable empirical study of it. Cognition and purpose do not form an inseparable pair; one can recognise the causal role of cognitive processes in the determination of behaviour without at all committing oneself to the view that cognition-guided behaviour must be purposive.
