No abstract

No abstract

No abstract

No abstract

In an earlier paper I concluded from an analysis of the data on massing and distribution of practice that there was experimental evidence for a dual division of learning into 'achievement of insight' on the one hand and 'fixation' on the other. I also predicted that if (and as long as) a novel configuration were unstable, massing of repetitions would be more economical than distribution, since if the configuration were once lost the preliminary process of problemsolving must be gone through again. On the other hand, the fixation process was said to be favored by distributed practice. Two experiments with puzzles were reported supporting this prediction, showing massed practice more economical in early, and distribution in later, trials, with distribution having still more advantage when the task was relearned after a delay of several weeks (i).Since that paper was published, however, I have found reason to believe that (a) the primary factor in the advantage of massing is not achievement of insight as such, but the instability of the configuration. In other words, those materials which show a sharp drop in the forgetting curve should profit by massing of repetitions, particularly in the early trials. Now, a rapid drop near the beginning of the forgetting curve is not peculiar to puzzle-solving, nor indeed would this be selected by most observers as the most typical example. Far more familiar is the rapid forgetting of nonsense materials. It seemed probable, therefore, that nonsense materials would also show advantage for massed practice, provided the time interval between trials in the distributed procedure were considerably greater than the two-minute period shown by Ward (8) to yield reminiscence rather than forgetting. 1 In addition, it appeared plausible that two secondary factors might also favor massed procedure, (b) The first of these is freedom from interfering impulses set up within the learning process. It