Glossary Updates12 new terms added to the glossaries · October 2, 2026, 22:44 CEST
AI TechDocKnowledge

Cognitive psychology in technical communication

Reading comprehension

Also known as: Text comprehension · Comprehension of technical text · Mental representation of text

Reading comprehension is the construction of a mental representation of what a text describes. Research distinguishes several levels of this representation, from the wording of the text to a model of the situation it refers to, and shows that readers continually fill gaps with inferences from prior knowledge. For technical communication the theory explains why concept information, consistent terminology and coordinated words and pictures help readers form an accurate understanding of a product.

By knowledge.aitechdoc.world · Last reviewed

Definition

Comprehension is the process by which a reader builds a coherent representation of a text and of the state of affairs it describes. Inferences connect sentences and supply information the text leaves implicit. A situation model is the reader’s representation of the described situation, integrated with prior knowledge. A mental model in the cognitive science sense is a person’s internal representation of how a system works, which is used to explain and predict its behavior.

Historical development

Kenneth Craik (1943) proposed that people reason with small-scale internal models of the world. Philip Johnson-Laird (1983) developed mental model theory for reasoning and discourse. Teun van Dijk and Walter Kintsch (1983) distinguished the surface form of a text, the textbase of its propositions and the situation model, and Kintsch (1988) described comprehension in the construction-integration model, in which text and knowledge first activate many associations that are then integrated into a coherent representation. Allan Paivio (1971) proposed dual coding theory, which holds that verbal and imagery-based information are represented in separate but connected systems. Donald Norman (1986, 1988) applied mental model concepts to design: the designer’s model, the user’s model and the system image — the product, its interface and its documentation — through which the user forms a model; the gulfs of execution and evaluation describe the distance between a user’s intentions and the actions and feedback a system offers. David Ausubel (1960) studied advance organizers, introductory material that relates new content to what readers already know.

Main models

  1. Levels of representation. Surface code, textbase and situation model; readers can remember the wording without understanding the situation, and the reverse.
  2. Construction-integration model. Comprehension as the interaction of text-based activation and knowledge-based integration.
  3. Inference. Bridging and elaborative inferences fill gaps; readers with less domain knowledge make fewer correct inferences.
  4. Dual coding theory. Words and pictures provide two connected routes to representation.
  5. Mental models and the system image. Users form models of a product from what it and its documentation show them.

Function and purpose

The models explain why readers with the same text arrive at different understandings, why gaps that are obvious to an expert writer are not obvious to a novice reader, and why a product can be operated correctly in routine conditions but misunderstood when something unusual happens.

Applications in technical communication

Documentation practice treats the documentation as part of Norman’s system image. Concept information is provided before or alongside tasks where readers need a model of how a system works — for example of operating modes or interlocks — so that they can interpret behavior the procedure does not describe; the context card Mental models and concept information follows this line, and DITA concept topics are one implementation. Consistent terminology supports the integration of information across topics, because a changed term invites the inference that a different thing is meant. Plain language and comprehensibility guidelines, such as ISO 24495-1:2023, address the surface and textbase levels; advance organizers and overview pages support the situation model. Figures and text are coordinated so that each supports the other, in line with dual coding. Comprehensibility is evaluated with readers, for example with think-aloud protocols, rather than inferred from the text alone.

Limitations and evidence

Readability formulas, such as Rudolf Flesch’s (1948), estimate difficulty from word and sentence length; they do not measure comprehension, ignore domain knowledge and should not be used as targets for rewriting. Situation models and mental models are theoretical constructs inferred from behavior, and the term “mental model” is used with different meanings in cognitive science, usability and decision-making (see the separate entry Mental model (thinking tool)). Claims that readers process text with one brain hemisphere and pictures with the other simplify research on lateralization beyond its findings and are not a basis for design. Plain language improves access but does not on its own make information correct or complete.

Relation to other areas

Comprehension depends on attention and working memory and in turn feeds judgment and action: a reader acts on the model formed from the text. Cognitive load theory addresses the cost of building that model, and the analysis of mistakes in Judgment and human error traces many of them to incorrect models.

Conclusion

Comprehension research describes reading as model building. Documentation supports it with concept information, consistent terms, coordinated words and pictures and testing with real readers.

Further reading

Cite this article

knowledge.aitechdoc.world. “Reading comprehension.” Encyclopedia, AI TechDoc Blog. Last reviewed October 5, 2026. https://knowledge.aitechdoc.world/encyclopedia/cognitive-psychology-in-technical-communication/reading-comprehension