Discourse processing

Text comprehension does not entail only understanding isolated words and sentences. Rather, it is an attempt to comprehend the author's communicative intention expressed in the text. A number of models focus on text comprehension aiming to find principles governing this process. Below, several models and a selection of studies inspired by the models are presented.

3.1.1. The Kintsch model

The Kintsch model (1974) views comprehending a text as developing the representation of the text in memory in the form of a network of interrelated propositions. The key notions of the model – propositions are defined as “ideas that can be expressed in words, not the words themselves” (McNamara, Miller and Bransford 1991: 342), “units of meaning roughly corresponding to phrases or clauses” (Louwerse and Graesser 2006: 426). It is useful to note that during their studies Kintsch and Keenan (1973) observed very high correlations between reading times and the numbers of propositions, not the numbers of words in sentences.

The Kintsch model from 1974 evolved into the next conceptualisation of discourse processing – the Kintsch and van Dijk model (1978), which led to the rapid development of discourse comprehension research. According to this model, processing the text is done automatically in cycles due to limited memory capacity. Comprehension operates at two different semantic levels: the micro level, referring to the local discourse of the text, and the macro level, referring to the more global discourse of the text.