Comparison of the results of the studies of cognitive development and normal aging suggests a large degree of commonality in both behavioral and ERP effects across a wide age range. Whether measured in young children, adolescents, young, middle-aged or elderly adults, the size of the ERP repetition effect did not differ among the various age groups. This was true whether memory was tested directly during continuous recognition or indirectly during variants of semantic categorization tasks. Similarly, in the studies of adult aging, the degree of RT facilitation during the semantic task did not differ with age and, in both the studies of cognitive development and aging, the degree of RT prolongation during the explicit tasks did not appear to differ as a function of age. Moreover, in the studies of adult aging, the effects of three versus two exposures of a word assessed in the PM session (TABLE 1), modulated RT similarly in all three age groups. These data argue for continuity of information processing across a very wide age range during both direct and indirect memory tasks, when retention is assessed during the recognition (for explicit testing) and repetition (for implicit testing) phases of the task. Since ERP and RT modulation do not appear to differ with age during the retrieval phases of these experiments, how can the performance differential seen in young children and older adults be explained? Some evidence comes from the ERP data recorded during the study phases of our explicit tasks. During continuous recognition, both young children and elderly adults did not show the typical subsequent "memory effect." In the case of the children, the subsequently unrecognized ERP was larger than the subsequently recognized ERP, whereas for the older adults, there was no difference between these two ERPs. Moreover, during these same tasks, the young children did show the "crossover" (new greater than old) pattern for slow wave activity, whereas the older adults did not. Since these ERP findings were obtained during the acquisition phase (i.e., to new items that had to be encoded for subsequent retrieval), the data argue for encoding difficulties as one means of explaining the performance differences seen at the two ends of the age spectrum. However, since the older adults displayed a different new/old pattern for slow wave activity, the two age groups may differ qualitatively in the strategies employed to encode items for subsequent retrieval.(ABSTRACT TRUNCATED AT 400 WORDS)