The aim of this study was to explore the effects of the following memory encoding variables on human learning: depth (implying progression through levels of encoding); spread (elaboration of information at a given level) and; congruence (integration of the form of encoding and the material to be learned). Encoding refers to the way in which material is learned. Four experiments were performed to evaluate these effects. Undergraduate students of educational psychology were trained and tested in one or several strategies for learning paired associates, word recognition, or free recall of words. The strategies employed differed in the depth, spread, or congruence of the encoding required. Those strategies assumed to emphasize depth or congruence of encoding were found to facilitate recall. Self-reported strategies were most likely to facilitate recall when they involved congruent encoding or were few in number. Results were interpreted as providing support for the view that depth and congruence of encoding facilitates retention, but there was no evidence for a beneficial effect of spread of encoding. Postexperimental questionnaire data indicated, however, that this strategy was not widely used by participants. In view of this finding, a case for spread of encoding is still possible. (Author/AEF)