Can Explicit Instruction Boost Statistical Learning? A Meta-Analytical Review

This meta-analysis investigated whether and to what extent explicit instruction improves statistical learning (SL). While some previous studies individually demonstrated significant effects of explicit instruction on SL, there is a lack of a systematic and quantitative synthesis of this line of work so as to provide specific and detailed guidance for future research. To bridge this gap, a comprehensive meta-analysis was carried out, providing a first careful examination of the effect of explicit instruction on SL. Our analysis revealed a significant improvement in SL performance (g = .49) as a result of explicit instruction. Furthermore, we studied various factors that potentially influence the magnitude of the effect size of explicit instruction on SL. Specifically, we showed that explicit instruction in audiovisual compared to single-modality SL produced a more substantial enhancement in SL performance. In addition, the effect of explicit instruction was more pronounced in the linguistic domain compared to the nonlinguistic domain. These findings not only confirm the positive influence of explicit instruction on SL but also shed light on the specific conditions under which explicit instruction is more beneficial. Our findings have important implications for future educational research on developing effective instructional practices that potentially support SL in classrooms.