Orthographic Processing

Orthographic Processing

First Author: Nicole Conrad -- Saint Mary's University
Keywords: Orthographic Knowledge, Word reading
Abstract / Summary: 

Orthographic processing has been defined as “the ability to form, store, and access orthographic representations” (Stanovich & West, 1989, p. 404). This definition encompasses orthographic learning, orthographic knowledge, and orthographic memory, yet in past research, these constructs, and how they are measured, have not always been differentiated. Our understanding of the conceptual complexities surrounding orthographic processing has blossomed over the last 25 years. The papers within this symposium reflect the diversity of contemporary thought on how we define, measure, and conceptualize the various aspects of orthographic processing. We present research on (a) factors that contribute to establishing orthographic representations in memory (Tsantali & Georgiou; Conrad, Kennedy, & Hanusiak), (b) measuring the quality of those representations (Masterson & Apel), (c) orthographic learning as a predictor of reading skill (Deacon, Tims, Marinus, & Castles) and (b) how orthographic knowledge plays a role in the reading skills of special populations (Wolter).

Symposium Papers: 

What is the mechanism underlying the RAN-orthographic processing relationship?

First Author/Chair:Anastasia Tsantali -- University of Alberta
Additional authors/chairs: 
George Georgiou

Purpose: According to Bowers and Wolf (1993), rapid automatized naming (RAN) is related to reading because of its contribution to orthographic processing. However, the nature of the RAN-orthographic processing relationship remains unclear. Thus, the purpose of this study was two-fold: (a) to examine the relationship of RAN with different measures of orthographic processing (lexical and sub-lexical; accuracy and response time) and (b) to examine what processing skills may account for the relationship between RAN and orthographic processing.
Method: One hundred fifteen university students (85 females; mean age = 21.42 years, SD = 2.59) were tested on measures of RAN (digits and objects), orthographic processing (orthographic choice and word likeness), discrete naming (digits and objects), phonological recoding (phonological choice), speed of processing (visual matching), and visual attention span.
Results: The results indicated that RAN correlates only with lexical orthographic processing response time and that only when controlling for phonological recoding speed the contribution of RAN to orthographic processing drops to non-significant levels.
Conclusion: These findings suggest that RAN contributes to how quickly letter sequences are mapped in order to form the orthographic representations which are important for whole word recognition.

Orthographic representations established through spelling: What can we learn from errors?

First Author/Chair:Nicole Conrad -- Saint Mary's University
Additional authors/chairs: 
Kathleen Kennedy; Laura Hanusiak

Purpose: Children establish word-specific representations in memory through experiences with print (Perfetti, 2007). Compared to our understanding of how this learning occurs during reading, we know relatively less about spelling as a self-teaching mechanism. The purpose of this study was (a) to determine whether spelling practice establishes memory representations that can be used to support reading, and (b) to explore what can be learned about these representations through the errors that children make while spelling.
Method: Children in Grade 2 first learned one of two alternate spellings accompanying a drawing of a non-object (slurst/slirst). Children then practiced spelling each word four times within a story context. For each story, children could (a) consistently use the spelling provided during the first phase (correct response), (b) consistently use the alternate spelling (an “alternate” error), (c) consistently use another self-generated spelling (a consistent error), or (d) produce inconsistent spellings within a story (an inconsistent error). One day later, orthographic learning was assessed using three tasks.
Results: Chi-square analysis revealed that response types during the practice phase were not evenly distributed. Children most frequently made correct responses or consistent errors; less frequently produced were inconsistent errors, and almost never were alternate errors produced. Orthographic learning was evident for correct responses, and less so for consistent errors. Factors that differentiate which type of response is made are also explored.
Conclusion: Children establish strong representations when spelling that may or may not be “correct”, suggesting that if these representations established during spelling are to support fluent reading, corrective feedback during spelling may be required.

Facilitating Data Analysis and Inter-Lab Collaboration with the CSSS

First Author/Chair:Julie Masterson -- Missouri State University
Additional authors/chairs: 
Kenn Apel

Purpose: This presentation will focus on how to use the Computerized Spelling Sensitivity System and procedures for accessing the system, which is available free of charge to researchers or practitioners.
Method/Results: The Spelling Sensitivity System was developed by Masterson and Apel (2007; 2010; 2013) to reflect the level of linguistic knowledge demonstrated by individual spellings. Spellings are coded for the extent to which they represent adequate phonemic awareness, orthographic pattern awareness, morphological awareness, and storage of mental graphemic representations (MGRs). The Computerized Spelling Sensitivity System was created by Masterson and Hrbec (2011) to increase both efficiency and reliability of the system. The SSS has been used to characterize response to intervention (Masterson & Apel, 2013), spellings associated with speech disorders (Masterson & Preston, 2012), and spellings associated with cultural differences (Williams & Masterson, 2010 and developmental changes (Masterson & Apel, 2010). In CSSS, an individual’s spellings are parsed or segmented into elements, aligned with the elements in the target spelling, and scored on a 4-point scale representing linguistic accuracy. Spelling elements are defined as (a) the letter or letters associated with each phoneme in a base word or stem, (b) spelling modifications associated with changes to a base word or stem when adding an affix (i.e., junctures), and (c) the letter or letters used to spell an affix. Each element in the student’s spellings is scored from 0 to 3 based on level of linguistic knowledge represented.
Conclusion: Discussion will address possibilities of sample sharing among researchers.

Testing the Self-Teaching Hypothesis: Does orthographic learning predict gains in word reading?

First Author/Chair:Helene Deacon -- Dalhousie University
Additional authors/chairs: 
Talisa Tims; Eva Marinus; Anne Castles

Purpose: Orthographic processing plays a prominent role in several theories of reading. More recently, however, different aspects of orthographic processing have been delineated. These are crystallised orthographic knowledge, or children’s established knowledge of orthographic patterns, and orthographic learning skill, or children’s skill in acquiring new orthographic patterns. Based on the Self-Teaching Hypothesis, we predict that it is the latter skill that determines children’s progress in word reading.

Method: We tested this hypothesis in a longitudinal study with a sample of 90 second and third grade students tracked over a one and a half year. At both Time 1 and Time 2, participants completed measures of crystallised orthographic knowledge, orthographic learning, word reading accuracy, and word reading fluency. At Time 1, they also completed control measures of phonological awareness and nonverbal reasoning.

Results: We used cross-lag hierarchical regression analyses to predict gains in word reading skills; as such, all analyses included an earlier measure of word reading skill entered prior to the predictor variable. We also included controls for age, nonverbal reasoning, and phonological awareness in all analyses. Our preliminary analyses show that performance on measures of crystallised orthographic knowledge did not predict gains in either word reading fluency or accuracy. In contrast, measures of orthographic learning predicted gains in measures of both word reading fluency and accuracy.

Conclusions: Our findings that orthographic learning predicted gains in word reading accuracy and fluency lend support for the importance of self-teaching in children’s reading development.

Orthographic Knowledge and Morphological Awareness and The Relation to Literacy Success in Children With and Without Language Impairment

First Author/Chair:Julie Wolter -- Utah State University

Purpose: The purpose was to determine whether orthographic knowledge acquisition, specifically the rapid development of mental orthographic representations (MORs), and morphological awareness was significantly different between kindergarten and 1st grade children with and without language impairments (LI), and whether these linguistic skills were significantly related to reading and spelling success. Method: Participants: 66 children, with (n = 33) and without (n =33) LI tested in the kindergarten or first-grade. Procedures: The children completed an orthographic knowledge acquisition task (Wolter, & Apel, 2010) and a morphological awareness generation task (e.g., farm, “My uncle is a ___; farmer”). Children’s phonological awareness, reading, and spelling were assessed. Results: Children with typical language skills performed significantly higher that children with LI on tasks of orthographic knowledge acquisition, F(1,64) = 10.19.1 , p < .01, and morphological awareness, F(1,64) = 10.79.1 , p < .01. For children with typical language, in addition to phonological awareness, orthographic knowledge was correlated significantly with spelling (r =.52, p <.01), and morphological awareness was correlated significantly with word-level reading (r =.22, p <.05). For children with LI only orthographic knowledge was related to spelling in (r =.67, p <.01). Conclusions: Results suggest that the ability to quickly develop MORs of novel written words is related to their ability to spelling words. Additionally, although early measures of morphological awareness were related to reading success in children with typical language abilities, early tasks may be too difficult for children with LI.