ABSTRACT
In recent years, we have been studying how students learn and reason from history texts, with the emphasis on the plural-multiple texts. We began with a simple assumption about the form of the typical history textbooks read by students, namely that they were essentially narratives. This implied that students’ understanding of history texts would typically include a representation of temporal-causal event sequences. Work by Tom Trabasso and his colleagues on the causal analysis of narratives provided a fine starting point for our goal of describing this temporal-causal component in history stories. Building on early story-understanding research (Mandler & Johnson, 1977; Omanson, 1982; Rumelhart, 1975; Stein & Glenn, 1979; Thorndyke, 1977), Trabasso and colleagues proposed that causal connections play a critical role in establishing meaning and coherence in a story. By providing a detailed and explicit model of causal analysis and spelling out tests for necessary and sufficient conditions, Trabasso and colleagues were able to predict what readers recall (Trabasso & van den Broek, 1985),
include in a summary (Trabasso & van den Broek, 1985; van den Broek & Trabasso, 1986), and judge as important (Trabasso & Sperry, 1985; Trabasso & van den Broek, 1985). Although this type of causal analysis clearly captured an essential part of narrative understanding, an analysis of the use of multiple texts required more than this.
