ABSTRACT

Kintsch’s models have distinguished three levels of cognitive representation that are achieved as a result of comprehending text (Kintsch, 1988; Kintsch, Welsch, Schmalhofer, & Zimny, 1990; Schmalhofer & Glavanov, 1986; van Dijk & Kintsch, 1983). The surface code preserves the exact wording and syntax of clauses. In most cases, comprehenders retain the surface code of only the most recent clause being processed. The textbase contains the explicit text propositions in a stripped-down form that preserves meaning, but not the exact wording and syntax of text. The textbase also includes a small number of inferences that are needed to establish text coherence. Finally, there is the referential situation model of what the text is about. A situation model is a mental representation of the people, setting, actions, and events that are explicitly mentioned or inferentially suggested by the text (Albrecht & O'Brien, 1993; Bower, 1989; Glenberg, Meyer, & Lindem, 1987; Johnson-Laird, 1983; Morrow, Bower, & Greenspan, 1987; Singer, 1990). Situation models may vary in abstractness from bare-bone conceptual sketches to lifelike renditions of episodes in the real world. Most inferences generated during text comprehension are part of the constructed situation model.