Investigates the time course of predictive inferences by using naming and word reading times. Explains the setup and results of two different experiments. Suggests that predictive inferences occur online, but require time to be drawn and are initially encoded to some degree, but completed later. (PA)

Proposes a "partial explanatory" model of document processing. Describes research underlying the model, presents a grammar of documents, defines and illustrates the variables underlying the model using a set of tasks relating to a bus schedule, and demonstrates the advantages of explanatory over exploratory models of document processing. (SR)

A study was conducted to determine the boundary conditions under which schemata influenced the encoding and retrieval of prose. Results showed that during immediate recall, either an encoding or retrieval perspective influenced recall; however, when recall was delayed, only the encoding perspective influenced recall. (FL)

Examines encoding differences between bridging and predictive inferences in four experiments. Suggests that bridging inferences are encoded into the textbase, whereas predictive inferences are not--predictive inferences are held in working memory, perhaps as part of the reader's situation model. (PA)

Describes two experiments showing that experienced readers use text signals to guide their attention to relevant information in a text. Indicates that subjects (1) took longer to read a summary sentence if it was signaled in advance, especially poor readers, and (2) recalled important information better if the information was signaled. (JD)

Presents evidence that indicates recognition memory consists of three separate representations: a surface representation, a propositional textbase, and a situation model. (KEH)

Investigates the role of paragraph markings in text: how informative paragraph cues are and how they affect interpretation. Reports one study asking subjects to reinstate deleted paragraph markings. Reports a second study examining the effect of paragraph boundaries on reading time, ratings, and judgment of important ideas in the text. (JAD)

Finds that perceptual identification of target words explicitly presented was significantly higher than that of new targets and that targets inferred to maintain coherence and targets predicted by the text were identified as well as explicit targets and significantly better than targets schematically appropriate to the text. (SR)

Demonstrates that reactivation of unsatisfied goal information depended on the type of original goal context and whether it was fully reinstated and that reinstatement of goal information from the distant target episode was successful only when the context was fully reinstated. Supports models of comprehension that assume reactivation of relevant…

Tests two views of how relevant global information becomes readily available to readers. Demonstrates that backgrounded information became readily available if it shared features in common with the current contents of working memory. Shows that this occurred independently of whether the information was relevant or thematically related;…

Presents results of a study indicating that, over many propositions appearing in passages that vary widely in content and vocabulary difficulty, early and later propositions are better recalled, and that the rated importance of a proposition predicts probability of recall independent of serial position. (HTH)

Tests how well two models of text comprehension (one referential and one causal) predicted the importance ratings and the recall of procedural texts. Finds better predictive power of the causal model for importance ratings; neither model performed well with recall; and neither model was as predictive of procedural text as they were in previous…

Investigates whether individual differences in working-memory (WM) capacity are related to the ways readers use inferences to facilitate text comprehension. Finds several differences between low-memory-span and high-memory-span readers. Notes that readers with adequate WM capacity can keep interpretations open ended and await more information from…

Hypothesizes a resonance process in which concepts and propositions in the discourse resonate in response to related elements in the current sentence, initiating a process that makes available a subset of the information in the representation. Reviews research; presents explicit assumptions about the discourse representation; and describes results…

States that previous research shows that readers who are given multiple perspectives on a domain excel in solving problems using material from that domain. Reports on three experiments that provide online evidence for a reinstatement-and-integration theory hypothesized previously as a possible processing account for problem solving with domain…