Transfer effects in a problem solving context

In recent years hypothesis theory has been applied to transfer between problems and thus used to explain a variety of transfer phenomena. A version of the theory is presented which leads to the prediction that identical variations in identical problems can lead to opposite trends in speed of solution of a critical problem depending on the nature of that problem. Using numerical tasks it was demonstrated that identical variations in preliminary problems either can transform a normally simple problem into an insoluble one or alternatively an insoluble problem into a simple one with the only alteration being in the degree of similarity between the hypotheses required for the initial problems and the critical problem. Implications for Einstellung, the sequence effect and general problem solving are discussed.