Comparison of trial-and-error and graduated stimulus change procedures across tasks

Graduated stimulus change and trial-and-error methods were used to train a two- choice discrimination task to two groups of profoundly retarded individuals. The results were compared across four size- and four shape-discrimination tasks. Graduated stimulus change was more efficient than trial-and-error on the initial problem, continued to be more efficient across tasks, and was more efficient on two reversal problems. However, experience with similar problems was shown to be an important factor.