Academic journal article Education

The Effect of Gender on the Construction of Backward Inferences

Academic journal article Education

The Effect of Gender on the Construction of Backward Inferences

Article excerpt

Successful reading comprehension requires that the reader not only identifies and understands individual text words, and processes the individual sentences of the text by detecting their syntactic structures, but also identifies the relations between the various parts of the text, as well as between the text and his/her world knowledge (van den Broek, 1994). This is crucial for comprehension because for most texts writers imply more than what they write explicitly. Therefore in order to construct the coherence of a text, a good reader actively deals with inferential meaning.

There are many classifications and definitions of inferences in the literature (Graesser, Singer & Trabasso, 1994; McKoon & Ratcliff, 1992; Richeit, Schnotz & Strohner, 1985; Singer, 1994; Singer, 1990; van den Broek, 1994). Despite different classification systems, the common ground is that three types of inferences are generated by the reader while trying to construct a situation model of what a text is about (Graesser et al., 1994). These are backward inferences (explanations), concurrent inferences (associations) and forward inferences (predictions). To comprehend a message, the reader may produce all three types of inferences. However, among them backward inferences play a crucial role in text comprehension since they identify unstated text connections by building relations between a current text proposition and a prior text proposition, and since discourse coherence is disrupted if they are not drawn (Singer, 1994; Singer & Remillard, 2004; Wiley & Myers, 2003).

Research on inferences points to the fact that backward inferences dominate associations and predictions during deep understanding. For example, Zwaan & Brown's findings (1996) indicate that skilled comprehenders generate significantly more backward inferences while they tend to generate fewer associations than less skilled comprehenders because this type of information does not contribute to a coherent mental representation. Laing & Kamhi (2002) point to the fact that average readers generate significantly more backward inferences than below average readers. Similarly, Magliano & Millis (2003) report that less skilled readers adopt a local processing strategy and they are less sensitive to the potential causal relationships that could be inferred between story sentences. To conclude, research has shown that backward inferences (explanations) are more necessary than other kinds of inferences to construct a coherent text base, and therefore understanding is explanation-based (Graesser, Singer & Trabasso, 1994; Laing & Kamhi, 2002; Singer & Ferreira, 1983; Singer, Halldorson, Lear & Andrusiak, 1992; Suh & Trabasso, 1993; Trabasso & Magliano, 1996; Zwaan & Brown, 1996).

Despite the richness of the literature related to inferences, the topic still attracts the attention of the researchers since inference generation is multi-dimensional in nature. Inferring is not only data driven but also knowledge driven and the differences in the cultural context, situational context, medium context, verbal context and personal context can affect the activated relevant world knowledge (Richeit, Schnotz & Strohner, 1985). Van den Broek (1994) describes the generation of inferences as a complicated cognitive activity consisting of sub-processes. For him and many others (Braten & Stromso, 2003; Narvaez, van den Broek & Ruiz, 1999; van den Broek, Lorch, Linderholm & Gustafson, 2001) the execution of these processes depends on many variables such as the reader's knowledge, attention capacity, high working memory capacity (Virtue, Haberman, Clancy, Parrish & Beeman, 2006), purposes in reading, the information provided by the text and the demands imposed by the task. Richeit, Schnotz & Strohner (1985) suggest personal context as one of the salient variables which affect the inferences the readers generate during comprehension. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.