Why do people overestimate the effectiveness of blocked learning?

Abstract Interleaved learning has been shown to be better for delayed memory performance than blocked learning. Nevertheless, people judge blocked learning to be more effective. In the present work, we tested different explanations for this metacognitive bias. Across three experiments, participants studied sequences of object-color associations for a final color-reproduction test. In Experiment 1, colors of a sequence were selected from one color category (blocked-study condition) or distinct color categories (interleaved-study condition). Participants gave higher judgements of learning (JOLs) for objects studied in the blocked condition, although performance was better for objects in the interleaved condition. In Experiment 2, participants’ immediate memory performance after each sequence was additionally measured. JOLs were again higher for objects in the blocked condition, and they mimicked participants’ immediate memory performance suggesting a link between participants’ evaluations of the learning strategies and their immediate memory effects. In Experiment 3, the objects of one sequence were either grouped by category (blocked-study condition) or derived from distinct categories (interleaved-study condition). Neither JOLs, nor immediate performance was affected by this manipulation of blocked learning, speaking against the possibility that people prefer blocked learning because of habit only. We conclude that people overestimate the effectiveness of blocked learning due to the immediate memory boost caused by blocked learning and not due to their previously acquired habit to study in blocks. This study provides insights into how people evaluate the effectiveness of learning strategies and why these evaluations are not always accurate.