Zwaan's Model of Literary Comprehension

Summary of Zwaan, 1996. "Toward a model of literary comprehension"

Nonliterary texts (or texts meant to be read as content-oriented discourse) aim to be CONSIDERATE TEXTS that minimize processing load on the reader through, for example ...

• simple words
• simple sentences
• stating main point at the outset

Literary texts contain devices that side-track routine comprehension processes through ...

• FOREGROUNDING [Mukarovsky, 1964]
• DEFAMILIARIZATION [Shklovsky, 1916]
• INCONSIDERATE at the level of SURFACE STRUCTURE:
  • The reader places SELECTIVE ATTENTION on surface structure rather than the content, creating the AESTHETIC EFFECT; this shift of attention occurs at the expense of other processes, such as the construction of a text model and situation model. [Kintsch, 1992]
  • Examples:
    • Enjambment (disrupts syntactic parsing)
    • Story (Surface order of events) vs Plot (chronological order) (Flashbacks and flash-forward can disrupt comprehension but create suspense. [Ohtsuka and Brewer, 1992])
• INCONSIDERATE at the level of the TEXT MODEL
  • Considerate texts make the main point clear from the outset - they are coherent and unambiguous
  • Literary texts may never state the main point - they are often deliberately incoherent and ambiguous (Seemingly irrelevant information has to be kept active in memory for a long period, or found by re-reading.)
• INCONSIDERATE at the level of the SITUATION MODEL
  • Settings and events that are unfamiliar or partly contradict reader's prior knowledge of the world make it hard for the reader to construct a situation model.
  • Perspective from which events are described - the narrator may perceive reality in ways different from the reader's, rendering the reader's knowledge base useless to some extent.
  • Texts with OPEN ENDINGS (i.e. without clear resolution) require the reader to simultaneously entertain multiple possible situation models, although empirical evidence suggests that readers actually end up refraining from constructing any situation model at all [Mani and Johnson-Laird, 1982].

Readers can activate the LITERARY CONTROL SYSTEM even in the absence of literary devices. [Zwaan, 1991, 1993, 1994] When told that texts are literary, they ...

• Read slower;
• Process information that contradict prior knowledge faster when compared to reading texts as nonliterary;
• Construct stronger surface structure representations
• Remember causally unimportant information better (perhaps because they expect it to become important later) in their textbase, and
• Construct weaker situation models

In literary comprehension, readers ...

• Keep seemingly irrelevant information active longer in their working memory and construct a partly incoherent, loosely organised text model (and have less trouble processing information that contradicts the existing text model)
• Use information about the genre of the text to activate prior knowledge relevant to literary comprehension
• Delay the construction of a situation model as long as possible, until there is sufficient information

Readers with different working memory spans may use different strategies:

• Low span readers use a TOP-DOWN STRATEGY, forming interpretations early and forcing the rest of the text to fall into that pattern
• High span readers use a BOTTOM-UP STRATEGY, holding off on interpretation and allowing the text to guide them

References:

Mani and Johnson-Laird, 1982. The mental representation of spatial descriptions.

Mukarovsky, 1964. Standard language and poetic language.

Kintsch, 1992. Cognitive models of text comprehension: Literary texts.

Shklovsky, 1916. Art as technique.

Ohtsuka and Brewer, 1992. Discourse organization in the comprehension of narrative order in texts.

Zwaan, 1991. Some parameters of literary and news comprehension: Effects of discourse-type perspective on reading rate and surface-structure representation.

Zwaan, 1993. Aspects of literary comprehension.

Zwaan, 1994. Effect of genre expectations on text comprehension.