Not Just STEM: Refuting Preconceptions About History

• In Event: Reducing the Impact of Misconceptions

Abstract

Readers commonly hold preconceptions, defined here as inaccurate or incomplete conceptualizations, that can interfere with learning and performance. One method of improving understandings involves the use of refutation texts (Tippett, 2010), which explicitly point out, rebut, and explain preconceptions. Participants who read refutation texts, as compared to non-refutation materials, are more likely to integrate text content with prior knowledge during reading, and perform better on memory tests even two months later (Diakidoy, Kendeou, & Ioannides, 2003; Kendeou & van den Broek, 2007; Mason & Gava, 2007).
While refutation texts seem applicable across topic domains, to date, most studies of their utility have focused on science concepts. In two experiments we assessed the generalizability of refutation benefits to history learning, a domain in which preconceptions are routinely documented (VanSledright & Brophy, 1992), and reading is regarded as an important pedagogical practice (Massey, 2015). Consider though that while scientific preconceptions are often derived from direct experiences with the physical world (diSessa, 1993), historical preconceptions often arise as conveyed through indirect sources including movies, novels, and even textbooks (Wineburg, 2001). Simply reading additional information, including refutational content on the same topic, may be insufficient to motivate knowledge revision (Tetreault, 1986).
We pre-screened undergraduate participants for preconceptions about a major event in American history: Rosa Park’s refusal to give up her bus seat as a seminal moment in the Civil Rights Movement. Participants exhibiting misunderstandings of the event, and participants who held valid understandings, were randomly assigned to read a refutation or non-refutation text. In Experiment 1, participants spoke their thoughts aloud as they read the text. Participants with preconceptions indicated they gained new knowledge during reading more often than did participants without preconceptions across both text types. But participants in the refutation condition more often recognized discrepancies between their understandings and the text content and generated more text-based inferences than did participants in the non-refutation condition. After reading, participants in the refutation condition, particularly those with preconceptions, recalled content that introduced new information more often than did participants in the non-refutation condition. Participants with preconceptions also exhibited substantial improvement in pre- to post-reading performance when the screening questionnaire was administered a second time.
These results could be due to the careful readings encouraged by think-alouds, so Experiment 2 asked participants to read the texts silently. We also tested whether any learning gains would also maintain one week later. The methods were otherwise identical to Experiment 1. The same recall and learning benefits accrued immediately after reading and one week later.
While additional replication attempts are warranted, these results provide evidence for the efficacy of historical refutations. Emergent benefits, regardless of topic, are likely attributable to the text content encouraging and supporting knowledge revision, as specified in a variety of models (e.g., Albrecht & O’Brien, 1993; Kendeou & O’Brien, 2014; Vosniadou, 1994). The think-aloud and recall data reported here provide evidence for a core proposition underlying these models as critical to refutation benefits: Readers must notice and act upon conflicts between texts and knowledge.

Authors

• Amalia Donovan, Northwestern University

• David Rapp, Northwestern University