The effect of constant vs. varied encoding and massed vs. distributed presentations on recall of paired associates.

The variables of coding (constant vs. varied) and spacing (massed vs. distributed) were factorially combined in an independent-groups design. Sixteen nominal pairs were presented twice, with the two presentations being either massed or distributed. A given nominal pair (e.g., AR-LE) was accompanied either by the same word-pair code on both presentations (e.g.,arm-leg) or by a different code on each presentation (e.g.,arm-leg; arrive-leave). On a subsequent recall test, the nominal stimulus only was shown. Better recall of the nominal response was found after constant coding and after distributed presentations, and there was no interaction between the spacing and coding variables. The results (a) support none of the current proposed explanations for the distributed practice effect, and (b) suggest that the critical factor in determining the probability of response recall is the strength of the strongest potential route to the response, not the number of different routes. This research was supported by a grant from the Graduate School of the University of Wisconsin.