ABSTRACT
The processes involved in analogical transfer are not unlike those involved in insight problem solving in that they involve a difficult problem that needs to be restructured or seen in a new light. To inform the study of insight, this chapter reviews research on analogical transfer (using Duncker’s “tumor” problem) and summarizes the conditions under which rates of spontaneous solution are improved. Many studies have shown that the transfer of relevant information from previously experienced instances is often limited by the tendency to focus on surface-level information rather than structural information. Although generally assumed to aid transfer, this review revealed that many manipulations designed to promote abstraction during encoding of source information have not been particularly robust. However, the use of dynamic visuospatial representations or comparisons among more than two analogs may be a more effective way in which to promote transfer. Further, research aimed at fostering re-representation at the time of solution attempts suggests a more promising approach, because it does not require additional processing during initial encoding of source information, but rather focuses on how solution processes can be employed in a manner that stimulates restructuring.
