Studies in spaced and massed learning; paired-associate and serial learning.

Abstract

An experiment on paired-associate learning was carried out to test Hovland's prediction that spaced practice should be more advantageous than massed if the order of presenting the pairs of nonsense syllables were fixed. In the results, spaced and massed practice showed little difference in learning efficiency. Hovland's prediction was not confirmed. In a previous experiment of this series of studies it had been found that spaced practice was more efficient than massed in serial learning when syllables of “low-meaningfulness” value were used; but that there was little difference in efficiency between the two forms of distribution when “high-meaningfulness” syllables were used. The results of the two experiments naturally suggested that, as regards the relative efficiency of spaced and massed practice, paired-associate learning of low-meaningfulness syllables differed from serial learning of similar syllables, but matched serial learning of high-meaningfulness syllables. It seems justifiable to draw the conclusion that the relative efficiency of spaced and massed practice is different in paired-associate and serial learning, not because remote associations can be formed in the one case and not in the other, as Hovland (1939a, 1939b) assumed; but because the organization of the material and the learning processes are different in the two forms of learning.