Studies of learning to learn X. Nonspecific transfer effects in free-recall learning

Abstract

Learning to learn in multiple-trial free recall was investigated. The experiment comprised a training and a test phase. Three methods were used in the training phase: random order of presentation and free recall (RF); constant order of presentation and free recall (CF); constant order of presentation and serial recall (CS). The method in the test phase was always RF. Consistency of output order was manipulated by instructions in both phases. Acquisition of the training lists was faster for CF and CS than for RF but practice gains were comparable for the three methods. Method of prior practice per se did not influence test-list performance; however, there was an interaction of conditions of training and instructional treatments.