1996 Volume 3 Issue 2 Pages 2_83-2_95
It is well known that “recursion” is a very difficult concept in computer programming. Pragmatic knowledge seems to be helpful to learn these kinds of unfamiliar concepts. That is referred to as the knowledge in which the functional relations among objects are represented like daily actions with purposes or intentions. In the present research we propose “subcontract instruction” as a pragmatic instruction to make a recursive LISP function. The instruction encourages learners to think that a LISP function as a worker divides a task and entrusts its portion to a subcontractor that has the same structure as the original LISP function. The effectiveness of the instruction was examined in an experiment, compared with “trace instruction” and “template instruction.” The results showed that the subjects given “subcontract instruction” could solve different types of problems flexibly and yielded high performance. It is argued that daily actions can be base domains on which people learn new concepts through analogical reasoning.