Working memory and strategies in reasoning

Abstract

The study of thinking and problem solving is organised into fairly broad areas such as reasoning, planning, and decision making, and within each area into families of tasks such as syllogistic reasoning, decision making under uncertainty, and planning in move problems. Even within specific tasks it is clear that individuals vary in how they tackle them, and that individuals vary over time. Virtually every task of interest to the study of thinking and problem solving, I suggest, is open to a variety of approaches, or strategies, for solution. For example, given the task of finding the sum of the digits 1 to 100, an obvious approach is to begin adding the numbers in sequence. However, a better approach is possible if the person notices that the digits 1 to 100 form 50 pairs each of which add to 101 i.e., (1 + 100), (2 + 99), (3 + 98), (4 + 97) and so on. This second ‘grouping and multiplying’ procedure readily gives the correct answer (5050). It is reported that the famous mathematician Gauss used the second procedure when given the above task as a young child in school (Hall, 1970) enabling him to produce the correct answer very quickly while his classmates struggled with the ‘simple summing’ procedure. Intuitively, it would seem that the ‘grouping and multiplying’ method is better because it produces the answer with less load on memory. This chapter will be discussing work which relates to the link between strategies in reasoning tasks and memory load and, in particular, working memory load.