How does prior knowledge affect learning? A review of 16 mechanisms and a framework for future research

Abstract

The knowledge-is-power hypothesis posits that individual differences in domain-specific prior knowledge are among the strongest positive determinants of learning. However, a large recent meta-analysis found that the correlations between prior knowledge and knowledge gains had a mean of zero and a large range. This Prior Knowledge Paradox demonstrates the need to understand better how prior knowledge affects learning. We give an integrative review of 16 learning processes mediating the effects of prior knowledge on learning outcomes in learners, for example, encoding, chunking, comprehension, interest, cognitive load, meta-cognition, transfer, and interference. Each process is moderated by further variables, such as content domain, learner characteristics, or instruction. We summarize these findings in the Multiple Moderated Mediations (Triple-M) framework for future research, which emphasizes that a full understanding of the effect of prior knowledge on learning requires the joint consideration of the multiple cognitive and motivational learning mechanisms and moderating influences on them.