Simpson, M. L., Olejnik, S., Tam, A. Y.-W., & Supattathum, S. (1994). Elaborative verbal rehearsals and college students’ cognitive performance. Journal of Educational Psychology, 86(2), 267–278

In a small test of “high-risk” college freshman, the authors find that Elaborative verbal rehearsal improves performance on both immediate tests and (more substantially) on delayed tests, relative to a control group engaging in simple verbatim rehearsal. The authors had fairly strong success in training these students to produce effective elaborative rehearsals (3 hours total training).

Among other issues, they examine which aspects of the rehearsal seem most correlated with performance; they find that “completeness and organization” explains most of the variation. This suggests that Self-explanation alone may be sufficient, without much explicit elaboration.

To find a subset or subsets of components that would
predict almost as well as all six components, we computed
squared multiple correlations for all possible regressions.
The overall completeness and organization component was
the single best predictor, explaining 82.1% of the total
variance in overall exam performance. Among the two-
variable models, any one of four components (generaliza-
tions, creative responses, text examples, or personal exam-
ples) combined with the completeness and organization
component explained between 86% and 87% of the variance
in overall exam performance. Beyond these two-component
models, the three-, four-, and five-component models ex-
plained additional variance that we considered trivial

The definition of that component:

Oh. Well… alright. If a high organization/completeness score requires getting 5s on three other criteria, then it makes sense that it would have the most explanatory power.

I don’t like that their rubric is mostly numerical: three personal examples for full points, not just one or two! Supporting examples for all five key ideas for full points! Sounds boring and annoying. Three personal reactions!