The Role of Visuospatial Resources in Generating Predictive and Bridging Inferences
Titel:
The Role of Visuospatial Resources in Generating Predictive and Bridging Inferences
Auteur:
Fincher-Kiefer, Rebecca D'Agostino, Paul R.
Verschenen in:
Discourse processes
Paginering:
Jaargang 37 (2004) nr. 3 pagina's 205-224
Jaar:
2004-01-05
Inhoud:
It has been suggested that predictive and bridging inferences are generated at different levels of text representation: predictive inferences at a reader's situation model and bridging inferences at a reader's propositional textbase (Fincher-Kiefer, 1993, 1996; McDaniel, Schmalhofer, & Keefe, 2001; Schmalhofer, McDaniel, & Keefe, 2002). Recently, Fincher-Kiefer (2001) argued that the construction of the situation model may require visuospatial resources and demonstrated that this level of representation was impaired when readers were given a perceptual memory load but not a verbal memory load. This set of experiments test the hypothesis that if predictive inferences are generated at the level of a situation model, then this inference type but not necessarily bridging inferences should be disrupted by a concurrent visuospatial memory load. Experiments 1 and 2 employed a lexical decision task to measure inferential processing, and results supported this hypothesis. Experiments 3 and 4 employed a different measure, the contradiction effect, and results also provided support for the hypothesis. The selective interference effects found in these experiments suggest that predictive inferences require visuospatial resources but bridging inferences do not, and that these inferences may be constructed at different levels of text representation.