Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Revisiting the Construction-Integration Model of Text Compre
Transcript of Revisiting the Construction-Integration Model of Text Compre
A mental representation is some change in the way the mind views the world after reading the text (overly controversial subject).
Perceptual (images of how the word looked and sounded in a particular situation), verbal or linguistic (words, sentenses & discourses) and semantic (ideas expressed by the words) representation.
Idea units, combining more than one word in a schematic form, allows to represent the meaning of the sentence independent of the syntactic structure.
Why is it important? It allows us to distinguish what the most important ideas are in the text, and it predicts how the text is comprehended and remembered.
Mental representation of words, sentences, paragraphs is surface-level memory (important for poems!); semantic/propositional level (the ideas!).
The Process of Comprehension:
Construction and Integration
Typically, pretty straightforward process of looking at provided cues (units).
When ambiguity occurs, the CI model generates several plausible meanings in parallel and only later, when a rich context is available, sorts out which construction is the right one. This sorting out is done by means of an integration or constraint satisfaction process that suppresses those constructions that do not fit in well with context and strengthens those that do. (p. 815)
What can the CI theory do for teachers and/or researchers?
Final Task: Visual imagery is a powerful tool for comprehension. In pairs, think of a way to visually represent the construction-integration theory.
Understanding and comprehension are everyday terms, useful, but imprecise. We know what we mean when we say we understand a text, but understanding is difficult to define precisely.
Comprehension as a paradigm for cognition (from perception, constraint satisfaction process, to analytic thought, planning).
Atomic propositions: The boy is little -> [LITTLE, BOY]
Complex propositions: The little boy chopped wood -> [CHOP, BOY, WOOD], [LITTLE, BOY]
Propositional analysis is a valuable research (not teaching!) tool, because it helps to count ideas in the text and it helps to see where the break in comprehension might be.
Micro and Macro Structures
To distinguish between big discourse-level structure and sentence structure, Kintsch distinguishes between
Microstructure of a text is the network of propositions that represents the meaning of the text.
Macrostructure is the global organization of these ideas into higher order units.
Together, these two form the
Further, it is useful to distinguish the
Situation model represents the information provided by the text, independent of the particular manner in which it was expressed in the text, and integrated with background information from the reader's prior knowledge. What sort of situation model readers construct depends very much of their goals in reading the text as well as amount of relevant prior knowledge they have. (p. 811)
Situation models are not necessarily verbal.
It is important to form a good situation model and to integrate it with reader's prior knowledge. Failure to integrate leads to
(only retrieved via specific episodic memory).
How does the CI theory relate to the schema theory?
The hiker saw the grizzly bear. He was afraid.
van Dijk's Marostructure
The macrostructure of a text consists of those propostions that are globally relevant, that form its gist in everyday language. (p. 818)
vanDijk developed three rules that describe the formation of macrostructure (selection, generalization, construction), but these rules are post hoc/after the fact. These are not rules that allow us to generate macropropositions from a text.
--> Latent Semantic Analysis (LSA)
LSA allows one to measure the similarity of the content of sentences. The sentence in a paragraph that is most similar to all the other sentences in that paragraph is a good candidate for a macroproposition because it is the most central one. (p. 820)
Summary Street: allows to compare vectors of the text and the student summary, as well as to provide instantaneous content feedback. Summary Street has been shown to be effective in helping students write better summaries. (p. 827)
University of Colorado, Bolder: http://lsa.colorado.edu
Latent Semantic Indexing page of UT: http://www.cs.utk.edu/~lsi/
Word Identification & Inferences
P. 822-834: Share a quote you liked or by which you were puzzled.
Situation models are a form of “inference” by definition. Situation models vary greatly in their character. In the simplest case, their construction is automatic. On the other hand, situation models can be much more complex and result from extended, resource-demanding, controlled processes… The process may be shared by a social group or even by a whole culture and extend over prolonged period of time. (p. 833-834)