The ability to represent and comprehend written language is an important survival skill in modern life; this is why research on reading and spelling occupies such a dominant place in cognitive psychology.
Over the past 50 years, much has been learned about the processes involved in reading and writing. Of particular interest is the question of how children learn to read and spell. So far, we know that children's ability to process the sounds in language plays a vital role in reading and spelling acquisition. In the early stages of literacy, children who are able to detect how two words sound alike are at an advantage over those with poorer phonological skills (Bradley & Bryant, 1983). This is because the ability to hear the sounds in words depends, in part, on the ability to segment a word into its constituent sounds. Segmentation skills are a vital aspect of learning to read and spell new words.
However, reading and spelling are not just about linking letters to sounds and vice versa. In languages like English and French, words are not spelled solely on the basis of sounds. Some words are spelled in a particular way because some letter strings constitute units of meaning, or morphemes. For instance, although kissed could be spelled kist on the
basis of its sounds, it is actually spelled with an -ed ending because this denotes the past tense status of the verb. Children's understanding of morphology, therefore, plays a vital role in reading and spelling acquisition. Despite its importance, children's development of morphological awareness in relation to reading and spelling has not received the same level of attention as the role of phonological awareness. Given that phonological awareness plays such a vital role in early literacy, it is not surprising that children who have difficulties learning to read and spell tend to have poor phonological skills. There is a consensus in the field that developmental dyslexia, in which children with no constitutional or general cognitive impainnent have extreme difficulty learning to read, is underpinned by a deficit in phonological processing. This finding has had far reaching implications for the way in which dyslexic children are taught; many
remedial reading programmes place phonological awareness training at their core.
But when dyslexic children's phonological awareness improves and they are able to
read and spell words on the basis of grapheme-phoneme correspondences, the majority do
not go on to become fully literate; their reading may reach an adequate standard, but their
spelling and writing usually remain poor. This has led some researchers to wonder whether
dyslexic children hav~ wider reaching linguistic deficits; specifically, do they have deficits
in their ability to process grammatical information?
Grammar is really a portmanteau term that encompasses a range of language processing devices: word order (syntax), punctuation, and the ability to denote the appropriate status of a word so that it agrees with other words in a sentence ( derivational and inflectional morphology).
This thesis focuses on how children with dyslexia process one aspect of grammar - regular past tense inflectional morphology - in relation to groups of normally developing children. Regular past tense inflectional morphology refers to the way in which the -ed suffix is added to verbs to place them into the past tense.
Studies reviewed in Chapter 3 show that data exist to suggest that dyslexic children are impaired in written and spoken domains of inflectional processing. However, many of these studies did not methodically compare dyslexic children to both spelling and reading age matched controls and chronological age matched controls, nor did they concurrently
assess children on a range of written and spoken measures of inflectional morphological processing.
The use both chronological age and spelling and reading age control groups can be extremely informative when attempting to conclude where the source of dyslexic children's difficulty may lie. For instance, if dyslexic children are poorer than those of the same age on spoken language tasks, but similar to those with the same reading and spelling levels,
the conclusion would be that the dyslexic children's spoken language performance is due to their poorer levels of literacy. However, poorer performance relative to younger children would indicate that the dyslexic children have spoken language deficits.
The studies in this thesis aim to address three main questions:
1.Do children with dyslexia exhibit difficulties with inflectional morphology in
reading, spelling or spoken language~ in comparison to normally developing
children?
To address this question, a pilot study was first carried out (Chapter 4, Pilot study)
comparing 12 dyslexic children (DR) aged 11 years, to chronological age (CA) and spelling
and reading age matched (SA-RA) controls. The DR group was poorer at spelling regular
past tense inflectional endings, but no worse at spelling the ending of one morpheme
words, compared to the SA-RA group. There were no differences between the DR and SARA
groups on spoken language measures.
In the first study (Chapter 4, Experiment 1), 28 dyslexic children aged 9 years were
tested. It was found that the DR group was poorer than CAs on their reading and spelling of regular past tense inflected verbs. Relative to the SA-RA group, they were no worse at reading regular past tense inflected verbs but were a great deal poorer at spelling the regular past tense -ed ending. This finding occurred on two different spelling tasks, which
suggests that the finding is robust, at least for the children used in this study.
2. What are the likely causes for dyslexic children's impairments in spelling of the past tense -ed ending?
The three groups of children from Experiment 1 were assessed on a range of written and spoken tasks. Their regular word, irregular word, and non-word reading and spelling performances were compared (Chapter 4, Experiment 2). There were no differences between the DR and SA-RA groups on any of the tasks. This suggested that their difficulties with spelling the regular past tense -ed ending were not related to deficiencies in orthographic knowledge or phoneme-grapheme/grapheme-phoneme conversion abilities relative to the SA-RA group.
The groups' spoken language morphological and phonological awareness was compared, and no differences emerged between the DR and SA-RA groups, other than superior performance by the DR group on one task of morphological awareness (Chapter 4, Experiment 3) .. The DR group was poorer than the CA group on phonological awareness
tasks, and on most of the morphological awareness tasks. This suggests that dyslexic children's difficulties on explicit spoken language tasks can perhaps be accounted for by their impoverished experience of print.
A regression analysis showed that, for all three groups, orthographic knowledge was the most important indicator in their correct spelling of the -ed ending. In addition, for the SA-RA group, morphological awareness also emerged as a significant predictor. It has been found that awareness of morphology in spoken language plays a role in spelling of the -ed ending (Nunes, Bryant, & Bindman, 1997a). Application of morphological rules to spelling was assessed through a task in which children had to spell It! and /d/ sound ending non-words in either a noun or verb context (Chapter 4, Experiments 4a and 4b ). While adults spelled non-word endings according to their syntactic class, the children's data revealed that none of the three school aged groups made this type of distinction. This led to the conclusion that perhaps the role of morphological
awareness in spelling ofthe-ed ending is not as important as has been suggested.
The overall conclusion was that the cause of the dyslexic children's poor use of the -ed ending was not evident on the basis of studies carried out in Chapter 4.
3. Is there a sub-group of dyslexic children who have particular difficulties with
morphological spelling?
A close look at the dyslexic group's scores in Experiments 1-4 showed that some of these children were particularly poor at using the -ed ending in the light of near ceiling performance on applying the correct phonetic ending to irregular verbs and one morpheme words.
The following year, a new set of dyslexic children were screened (Chapter S) and three children were selected for case study research on the basis of their poor use of the-ed ending in comparison to their good use of phonetic endings on irregular verbs and one morpheme words. A further child, who over-used the -ed ending, was also recruited. These
children were assessed on a range of morphological and phonological tasks. There was
some variation of scores within this group, but the main findings were that: ( a) all had
morphological and phonological awareness within the normal range for their age; (b) they
all had particular difficulties with irregular spelling and reading, in relation to their regular
and non-word reading and spelling.
The general conclusion from the studies described in Chapters 4 and 5 is that dyslexic children do, as a group, have difficulties with spelling past tense -ed endings.
However, this does not mean that they are deficient in inflectional morphological processing. Poor use of the -ed ending can occur for a number of reasons, and it is important to assess children's use of this ending in relation to their ability to spell other types of words. For those with phonological impairments, their unstable phonemegrapheme skills can account for their problems; these children are poor at spelling most types of words and so will apply incorrect graphemes to phonemes. In addition, these children are still in the alphabetic stage of spelling development, and so cannot move on to
learning spelling rules for complex phoneme-grapheme relationships.
For another group, particularly those with normal or remediated to normal phonological skills, their poor use of the -ed ending relates to weak orthographic skills.
This latter set of children, all of whom have good morphological awareness, would particularly benefit from teaching strategies that emphasise the links between grammar and spelling, as they do not have inherent difficulties with grammar.
Before describing the experiments in dep~ a review of the relevant literature is provided. Chapter 1 discusses the development of reading and spelling. In Chapter 2, the nature and causes of developmental dyslexia are outlined. In Chapter 3, the ways in which adults, normally developing children, and dyslexic children process morphologically complex words are discussed.