The mental models approach is outlined for the following topics: combinatory semantics, anaphora, and inference. The flexibility of the language-processing system is discussed. (18 references) (Author/LB)

This paper demonstrates that language users exploit overspecification both in production and in comprehension of text. Controlled discourse production experiments show that the use of overspecification depends on theme shifts; comprehension studies test the claim that the differential use of anaphoric references correlates with thematic structure…

Research has shown that the comprehension of definite referring expressions (e.g., "the triangle") tends to be faster for "given" (previously mentioned) referents, compared with new referents. This has been attributed to the presence of given information in the consciousness of discourse participants (e.g., Chafe, 1994) suggesting that given is…

Four experiments demonstrated that the comprehension of novel compound nouns involves the recognition of a general relationship between two categories. They reinforce earlier findings that the process of creating meanings may be equally as efficient as the process of selecting meanings. (33 references) (Author/LB)

Two series of experiments investigate two hypotheses that may explain a conflict regarding pronomial reference. The hypothesis accepted is that the use of the gender cue is under the reader's strategic control; the hypothesis rejected is that implicit causality and gender cue interact differently with the complexity of the inference needed to link…

Three experiments are reported on the interpretation of conceptual anaphors, defined as those that do not have an explicit linguistic antecedent but one constructed from text. Two experiments showed that conceptual anaphors are quite easily understood but are processed with difficulty; the third one showed mixed results. (three references)…

The mechanisms involved in the assignment of an antecedent to an anaphoric element are examined. Taken together, four experiments suggest that conceptual, although grammatically illegal, anaphors do not cause comprehensive difficulties. (Author/LB)

In a study of knowledge activation and sentence mapping, subjects read stories that described concrete actions, and then the content of the stories was manipulated (i.e. stories were written that implied different emotional states). It is suggested that the more emotionally evoking situations one encounters the more memory traces are stored and…

Presents a view of the relative roles of logical presupposition and shared knowledge in "it"-cleft constructions. Indicators of shared knowledge relate to a speaker's "assumptions" about the state of the hearer's knowledge, whereas presuppositions indicate a speaker's "requirements" for what should be included within…