Building a cumulative science of memory development

Abstract

Cumulative science hinges on consolidating empirical evidence. However, both narrative reviews and meta-analyses often restrict integration by using construct labels in searches, which vary greatly across traditions and eras in psychology. A complementary approach is the mapping review, more common in other disciplines, which focuses on operational definitions and experimental design. Here, we pioneered using this approach in experimental psychology, applying it to memory development. We classified 506 publications in the last 5 decades using 16 design features and identified methodologically convergent work that originated from different theoretical frameworks. New insights emerge from this approach. First, characterizing the relative data density along each dimension of experimental design identifies where future research is most necessary. Second, we linked relevant evidence previously separated by nonoverlapping construct labels. Third, we illustrated the potential application of this technique as a precursor to subsequent research syntheses with an analysis of the development of pattern separation (also called mnemonic discrimination). To facilitate the process of literature integration and identification of methodological overlap, we have created a freely available interactive web application using the current database.