Cognitive Transfer and Conservation in Chess Playing A DISSERTATION SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL OF THE UNIVERSITY OF MINNESOTA BY Michael Atherton IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

The relationship between memory representation and transfer of learning is a long standing question in cognitive psychology and education. It has been widely explored in many domains, but has rarely been investigated in relation to the game of chess. In 2001, Marmèche and Didierjean questioned whether concrete elements of chess problems are retained in memory when players build abstract schemas to represent semantic features of a checkmate strategy. They interpreted their results as showing that providing players with a more meaningful description of a solution helped them create a more abstract representation, and that surface features of the original chess positions were conserved. The study described in this paper used a variant of their design to attempt to replicate a subset of their findings using players with a wider range of abilities, and used chess problems that varied in complexity. The primary findings of Marmèche and Didierjean were not found to be significant in the current study; however, it was found that providing more meaningful descriptions of solutions did result in helping participants solve two complex chess problems more quickly: one problem that was similar to an explained example, and (U=163.0, p=.025, N=45) also one that was visually dissimilar, but structurally isomorphic (U=173.0, p=.028, N=45). Learning a principle helped less skillful players with the less difficult problems, but did not significantly improve their performance on the difficult problems. Conversely, learning a principle did not help the more skillful players with the less difficult problems, but did improve their performance on the difficult problems, suggesting the possibility of a zone of proximal learning.