Learning by explaining to oneself and to others.

Abstract

One important source for the acquisition of knowledge, especially of factual knowledge, is the construction, transmission and comprehension of explanations. Two distinctive settings in which explanations are constructed are self-explanation, in which a single individual explains to himself and interactive explanation in which several individuals mutually and interactively explain to each other. While providing explanations to oneself might lead to the construction of additional knowledge by the explaining individual, providing explanations to each other involves forms of collaborative learning. To better understand the differences between and potential benefits of explaining to oneself and explaining to others, we compare empirical research on both kinds of explanation. The comparison is guided by three main goals. The first goal is to describe the learning that takes place during the construction of self-explanations. The second goal is to reveal how the construction of explanations as well as the learning changes if one moves from a selfexplanation setting to a more interactive explanation setting. With respect to interactive explanation settings, we particularly focus on learning by teaching. The third goal is to uncover differences in the learning between those individuals who construct explanations and those who receive them. Essentially, since learning by teaching has repeatedly been found to promote learning, this gave rise to the assumption that the beneficial effects of such arrangements are due to the successive construction of explanations. In particular, it has been hypothesized that explaining to others might be even more advantageous than explaining to oneself. However, so far empirical evidence does not strongly support this hypothesis.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 141 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.