Date of Award
Doctor of Philosophy (PhD)
Ian M. Begg
Analogical transfer is using information from one problem to help solve a subsequent problem to which it is equally as applicable. This transfer is called "spontaneous" if it occurs without any hint being given to the solver that the information from the earlier problem (the training analogue) will help solve the current problem (the target problem). The research literature shows that although subjects can use information from a training problem to help solve a target problem when explicitly instructed to use the information, they do not often do it spontaneously. In this thesis, thirteen experiments work together to demonstrate that problem-oriented processing of the training information leads to significant levels of spontaneous analogical transfer. Experiments 1, 2a, and 2b show that when subjects attempt to solve training problems, fail in their attempt, and then are provided with the appropriate solution and explanation, transfer of the training problem's principle to the target is likely, resulting in high solution rates on target problems. Whether or not the scenario of the training problem is the same as that of the target problem makes little difference so long as the principles are the same. In Experiments 3 through 8, and 11 and 12, it is shown that if subjects try to solve a training problem before hearing its solution, or try to explain a training problem's solution before hearing the correct explanation, spontaneous analogical transfer is much more likely than if subjects study the training problem for memory before hearing its solution or explanation. The advantage of problem-oriented training over memory-oriented training prevails despite the fact that the solution and explanation attempts nearly always fail, and the advantage is just as robust if the target problem is tested approximately 16 minutes after training rather than immediately thereafter. In Experiments 9 and 10, the extent of the benefit when the explanation does not follow the attempt to explain is examined. In Experiment 9, when the experimenter-provided explanation is replaced with verbatim repetition of the story, the benefit decreases a little. In Experiment 10, when the subjects are asked to recall the training story rather than listen to the experimenter-provided explanation, the benefit is eliminated. Problem-oriented training is superior to memory-oriented training when subjects are tested on problem transfer, but when the test is a memory test, memory-oriented training is superior to problem-oriented training. I conclude that problem solvers can spontaneously transfer information from analogous problems if they try to solve or explain the earlier problems and receive the appropriate solution or explanation feedback. The reason for the effect is that problem-oriented processes performed at study are the appropriate processes to use at test.
Needham, Douglas Robert, "Problem-oriented training promotes spontaneous analogical transfer" (1991). Open Access Dissertations and Theses. Paper 3629.