ABSTRACT
This chapter has reviewed the growing literature on planning in the context of well-structured problem solving. A primary distinction between initial and concurrent (opportunistic) planning has been highlighted and possible determinants of these different planning behaviours have been discussed. The literature suggests three factors that may determine planning behaviour. First, we considered the complexity of the problem itself. It seems clear that when problems remain relatively simple much of the solution to the problem can be developed in working memory. It is possible that more complex problems necessitate some degree of concurrent planning and when there is no possibility of externalizing the contents of working memory, pre-planning a solution may turn out to be counterproductive given the rather volatile and error prone nature of storage in working memory. Second, some environments are likely to support opportunistic problem-solving behaviour, especially where goals and subgoals can be easily externalized, thus reducing the potential load on working memory. The ability to return to this externalized information to make changes also seems to be a prerequisite to opportunistic planning. Third, systematic inter-and intra-group differences may exist in approaches to planning. For example, it is fairly widely reported that novices and experts in a particular domain may exhibit different strategies. Moreover, even within a single problem-solving episode planning strategies may change. Less emphasis has been placed on the possibility that there may be individual differences in preferred planning strategy, but some evidence is now emerging to support such a view. Finally, the effectiveness of initial planning has typically been ignored (or has not been controlled) in many studies of problem solving, but this phase is clearly an important stage in the problem-solving process, especially of puzzles of moderate difficulty. It may be a major factor in determining the strategies used whilst solving a problem. Planning is clearly a complex activity and research should consider the potential interactions between the putative determinants of planning behaviour that have been identified in this chapter. It seems unlikely that we will be able to fully capture the complexities of human planning by focusing upon any single factor in isolation.
