Cognitive abilities as determinants of success in acquiring logic skill

Abstract

In three studies, the authors examine the relationship between four cognitive factors— processing speed, working memory capacity, declarative knowledge and procedural knowledge—and performance on short-term logic learning tasks. One task taught skill in tracing signals through logic gates, and the other was a novel task (“color equations”) involving logical rules isomorphic to but superficially distinct from the rules of logic gates. Performance on the learning tasks was partitioned into separate declarative and procedural learning phases. In Study 1 (logic gates), only working memory capacity contributed uniquely to the prediction of declarative ( r = .73) and procedural ( r = .73) learning success. In Study 2 (color equations), working memory capacity ( r = .70) and procedural knowledge ( r = .35) predicted learning success (a confounded declarative/procedural learning factor). In Study 3, we employed a transfer paradigm, where, prior to administration of the logic gates task, half the subjects were administered color equations. Although there was a transfer effect for the initial declarative portion of the learning task, there was no procedural learning transfer: transfer subjects were no better able to solve problems than were untrained controls. For both untrained and transfer subjects, working memory capacity was the only significant determinant of success in both the declarative ( r = .76) and procedural ( r = .99) phases. We argue that the ability to acquire cognitive skill is little more than working memory capacity and a highly task-specific component.