Hovland, C. I. (1938). Experimental studies in rote-learning theory. I. Reminiscence following learning by massed and by distributed practice. Journal of Experimental Psychology, 22(3), 201-224.
http://dx.doi.org/10.1037/h0062123
In this experiment 32 subjects learned 16 lists of 12 nonsense syllables by massed and by distributed practice with and without a two-minute rest period between learning and relearning. Learning was by the serial anticipation method. During the rest pause between learning and testing and between successive trials during distributed practice, the subjects named colors. Reminiscence of the type obtained by Ward in his study of retention over short intervals of time (see XII: 3441) was verified and was found to be much more pronounced when the learning had been by massed practice than when it had been by distributed practice. Recall showed the effect less than did the number of trials required to go from seven syllables correct to mastery. Theories based upon (1) rehearsal, (2) fatigue, (3) motivation, (4) perseveration, and (5) removal of inhibition are discussed and criticized. To the author the last-named theory most adequately explains the results obtained. (PsycINFO Database Record (c) 2016 APA, all rights reserved)