So What Is Reading Anyway?

Proposed Definition

           Reading comprehension is an increasingly automatic process by which readers strive to construct a coherent mental representation of a text by way of extracting literal meaning from printed words and filling in gaps in understanding through inferences that are informed by socio-cultural factors and prior knowledge.

Comparison with Other Definitions                

As per PISA (2003), “Reading literacy is understanding, using and reflecting on written texts, in order to achieve one’s goals, to develop one’s knowledge and potential, and to participate in society.” The proposed definition takes into consideration reading goals, as does the PISA definition. However, while the PISA definition focuses on the outputs of reading literacy (developing knowledge and potential; participating in society), the proposed definition focuses more on the sub-processes in a reader’s mind and their interaction with the text and socio-cultural factors, which together determine the development of reading skills.

The Simple View of Reading (Gough & Tunmer, 1986) posits that reading comprehension is the product of sets of skills: decoding and linguistic comprehension. The proposed definition is closer to the SVR definition, in that both focus on decoding and comprehension skills as essential sub-processes for skilled reading comprehension. The proposed definition additionally takes into consideration how these skills are informed by working memory resources, socio-cultural factors, and prior knowledge.

 Reading Comprehension: A Breakdown of the Proposed Definition

           Reading comprehension is a process that becomes increasingly automatic with extended practice over a period of time. This automaticity is a result of increasingly efficient utilization of working memory resources for the various sub-processes required to read a text with comprehension (Perfetti et al., 2005)

           The aim of every reader must be a coherent mental representation of the text. This coherence requires simultaneous sub-processes within the reader’s mind—decoding and extracting the literal meaning of the words and sentences read; putting those meanings together to establish the literal meaning of the text read; and filling in any gaps in understanding or supplementing the literal understanding by making inferences. These three levels correspond with the model proposed by Kintsch & Kintsch (2005), comprising propositions, the textbase, and the situation model.

The following sections will delve into each sub-process in play during reading comprehension. As the definition states, it is the increasing automaticity of each sub-process that makes one a skilled reader. Automaticity is a result of efficient use of working memory. Furthermore, as stated, socio-cultural factors and prior knowledge inform these sub-processes. Thus, each of the following sections will include the role of working memory, socio-cultural factors, and prior knowledge in informing the sub-processes.

Decoding and Word Identification

Decoding and word identification form the initial sub-processes of reading. Decoding is defined as the ability to map letters to sound and use this analysis to identify printed words without context (Vellutino et al., 2007). Studies conducted by researchers such Sticht (1979) and Hoover & Gough (1990) show that decoding plays a crucial role in comprehension abilities of low-skilled readers, both adults and children.

Beginning readers devote their working memory resources to decoding and word identification (Perfetti et al., 2005). With repetition, word identification skills get strengthened and more automatic (Graesser et al., 1994) and take up fewer resources in working memory, as deliberate decoding skills are needed by skilled readers only to identify unfamiliar words (Chall, 1996).

As Perfetti et al. (2005) point out, “phonological knowledge prior to literacy could play a role in the development of reading comprehension” (p. 239). Phonological knowledge forms the basis of decoding, and it is logical to assume that a different native language and level of exposure to the language of instruction in everyday life play a role in decoding abilities of beginning readers.

 Vocabulary and Syntactic Knowledge

To convert decoded words into propositions, readers need vocabulary and syntactic knowledge.

As established by Perfetti et al. (2005), vocabulary knowledge plays a dual role—in word identification and in comprehension. Especially among readers with weak decoding skills, the dependence on the meaning of the word for identification may be higher. Furthermore, the role of vocabulary and strategies for understanding the meaning of words in comprehension is undeniable (Alvermann & Eakle, 2003), as it is estimated that for comprehension, the reader must know the meaning of at least 90% of the words (Nagy & Scott, 2000). Vellutino et al. (2007) posit that vocabulary knowledge is critical for comprehension of sentences. As readers read more, they may encounter unfamiliar words, the meaning of which they may infer from context to ensure comprehension. However, such inference has two requirements: one, the reader must know the meaning of most words in the text to make the inference; and two, the retrieval of the meaning must be automatic so as to not overwhelm working memory resources and become a bottleneck to comprehension (Perfetti et al., 2005).

The socio-cultural background determines, to a significant extent, the pool of vocabulary knowledge a reader brings (Hart & Risley, 1995), as vocabulary is accumulated through reading experiences and oral conversations in everyday life (Snow, 2002).

Syntactic knowledge is defined as “implicit knowledge of grammatical rules for ordering, coreferencing, and inflecting the words in sentences” (Vellutino et al., 2007, p. 8). Syntactic elements in written text include a variety, such as morphology, nominalizations, causal noun phrases, and anaphoric and cataphoric referents. Deep understanding of syntactic elements enables readers to parse sentences for meaning and establish propositional structures (Perfetti et al., 2005).

As the definition suggests, syntactic knowledge is implicit knowledge. However, as long as parsing requires readers to make deliberate efforts, working memory resources are devoted to it. The aim is for this knowledge to become implicit with reading experience for comprehension to proceed beyond the propositional stage.

Socio-cultural factors also play a role in syntactic processing, as there may be drastic differences in grammatical rules among different languages, specifically the native language of readers and the language of the text to be read, over and above differences in the syntax of oral and written language (Perfetti et al., 2005).

Gap-Filling

           Once a reader is able to extract meaning from the semantic and syntactic components of a text, the components come together to make the text locally and globally coherent (Graesser et al., 1994). However, to construct a coherent mental representation, or a situation model, the reader must fill in gaps in understanding by monitoring comprehension and making inferences that align with reading goals.

a) Comprehension Monitoring

           Every reader encounters points in texts that are not explicit in their meaning or difficult to comprehend at the propositional or textbase level. For this, Kintsch & Kintsch (2005) suggest “consciously monitoring one’s ongoing comprehension, identifying the source for a breakdown in comprehension, and attempting to resolve the problem rather than passively reading on” (p. 84).

Once the textbase is established, the goals of the reader (Graesser et al., 1994) as well the reader’s standard of coherence in relation with that text also determine at what points the reader pauses to consider what mental representation he/she needs to construct to fill the gaps and what background knowledge needs to be pulled forth for such a construction. For example, reading a short story to critically review it requires a different mental representation of the story than reading it for pleasure. Similarly, knowledge of the text structure also contributes to better monitoring (Perfetti et al., 2005). For example, a mystery story requires a reader to keep a lookout for clues, while a persuasive essay requires attention to claims and supporting evidence.

Studies have shown that skilled reading requires monitoring of comprehension (Hacker, 1997). For successful monitoring, a reader must both have relevant prior knowledge and know that such prior knowledge needs to be retrieved (Perfetti et al., 2005). Furthermore, comprehension monitoring requires the text to be held in working memory while also retrieving relevant knowledge from long-term memory. Therefore, other sub-processes such as decoding and semantic and syntactic processing must be automatic for comprehension monitoring to be successful.

b) Inferences

           As skilled readers monitor their comprehension and notice inconsistencies within the text or between the text and their knowledge, they make inferences to resolve the inconsistencies, especially those who have high “standards of coherence” (Perfetti et al., 2005, p. 247). Simply put, this means that the more the reader wants to see how all the details in a text, little or big, fit together, the more he/she will strive to make inferences. Oakhill et al. (2003) established that inference-making skills better predicted comprehension among young readers than working memory, vocabulary, and word recognition ability.

           Just as with comprehension monitoring, skilled readers know when to make inferences and what kind of prior knowledge to draw on to make those inferences (Perfetti et al., 2005). From this we can logically assume that readers with relevant prior knowledge or a wider range of reading experiences will be able to make better inferences.

Graesser et al. (1994) list out thirteen types of inferences that readers make to establish local coherence and global coherence, and construct situation models. Some of these inferences can be drawn from clues given in the text, and such inferences may not require retrieval of much background knowledge, placing a smaller demand on working memory. Other inferences, that require integration of the text and prior knowledge, place a higher demand on working memory. Consider the following sentences: Neel was the world’s youngest detective. He had a hat and sunglasses to hide his face. Here, the anaphoric reference connecting the pronoun he to Neel can be directly inferred from the text. However, the inference that detectives work best when they are not noticed or recognized needs integration of the text and prior knowledge about detectives. Therefore, for a reader to construct a rich situation model, all other sub-processes must be fairly automatic to allow for such “knowledge-based inferences” (Graesser et al., 1994, p. 374) to be made in the working memory.

In conclusion, a skilled reader automatically decodes and identifies words; knows the meaning of most words and makes automatic inferences from context to understand unfamiliar words; and automatically parses sentences. He/She devotes most working memory resources to monitoring comprehension and making knowledge-based inferences to construct a rich situation model.

References

Alvermann, D. E., & Eakle, A. J. (2003). Comprehension Instruction: Adolescents and their multiple literacies. In A. P. Sweet, & C. E. Snow (Eds.), Rethinking reading comprehension. New York, NY: Guilford Press, pp. 12-29

Chall, J. S. (1996). A proposal for reading stages. In J. S. Chall, Stages of reading development, (pp. 9- 39). Orlando, FL: Harcourt Brace & Company.

Gough, P. B., & Tunmer, W. E. (1986). Decoding, reading, and reading disability. Remedial and special education, 7(1), 6-10.

Graesser, A. C., Singer, M., & Trabasso, T. (1994). Constructing inferences during narrative text comprehension. Psychological review, 101(3), 371.

Hacker, D. J. (1997). Comprehension monitoring of written discourse across early-to-middle adolescence. Reading and Writing, 9(3), 207-240.

Hart, B., & Risley, T. R. (1995). Meaningful differences in the everyday experience of young American children. Paul H Brookes Publishing.

Hoover, W. A., & Gough, P. B. (1990). The simple view of reading. Reading and writing, 2(2), 127-160.

Kintsch, W. & Kintsch, E. (2005). Comprehension. In Paris, S. G., & S. A. Stahl (Eds.), Children’s reading comprehension and assessment, (pp. 71-92). Mahwah, New Jersey: Lawrence Erlbaum Associates.

Nagy, W. E., & Scott, J. A. (2000). Vocabulary processes. Handbook of reading research, 3(269-284).

Oakhill, J. V., Cain, K., & Bryant, P. E. (2003). The dissociation of word reading and text comprehension: Evidence from component skills. Language and cognitive processes, 18(4), 443-468.

Perfetti, C. A., Landi, N., & Oakhill, J. (2005). The Acquisition of Reading Comprehension Skill.

Programme for International Student Assessment. (2004). PISA Learning for Tomorrow's World: First Results from PISA 2003 (Vol. 659). Simon and Schuster.

Snow, C. (2002). Reading for understanding: Toward an R&D program in reading comprehension. Rand Corporation.

Sticht, T. (1979). Applications of the audread model to reading evaluation and instruction. Theory and practice of early reading, 1, 209-226.

Vellutino, F. R., Tunmer, W. E., Jaccard, J. J., & Chen, R. (2007). Components of reading ability: Multivariate evidence for a convergent skills model of reading development. Scientific studies of reading, 11(1), 3-32.

Appendix I: Visual Representation of the Reading Comprehension Process of Skilled Readers

           The following representation shows how much of working memory resources readers must ideally devote to different sub-processes at play in the reading comprehension process with increasing reading experiences.