Current Issues in Developmental Disorders

Chloë Marshall

• Publisher: Psychology Press
• Year: 2012
• eBook ISBN: 9781136230677
• Topic: Psychology
• Subtopic: Developmental Psychology
• Index: Psychology

Cognitive development in children is a highly complex process which, while remarkably resilient, can be disrupted in a variety of ways. This volume focuses on two types of neurodevelopmental disorder: syndromic conditions such as fragile X syndrome, Down syndrome, Williams syndrome and velocardiofacial syndrome; and non-syndromic conditions including dyslexia, specific language impairment, autism spectrum disorder and attention deficit hyperactivity disorder.

This book provides a state-of-the-art review of current research and covers key topics across the full range of developmental disorders. Topics covered include:

• diagnosis and comorbidity
• genetics
• longitudinal studies
• computational models
• distinguishing disorder from disadvantage
• language and culture
• the modern beginnings of research into developmental disorders

The book also looks at how the study of developmental disorders has contributed to our understanding of typical development, and themes emerge that are common across chapters, including intervention and education, and the neurobiological bases of developmental disorders. The result is a fascinating and thought-provoking volume that will be indispensable to advanced students, researchers and practitioners in the fields of developmental psychology, neuropsychology, speech and language therapy, and developmental disorders.

Part I

Current issues in the study of developmental disorders

1	Comorbidity and diagnosis of developmental disorders
	David M. Williams and Sophie E. Lind

Introduction

This chapter explores two main themes in two separate sections. The first section explores some of the challenges involved in the diagnosis of complex developmental disorders such as specific language impairment (SLI), developmental dyslexia, attention deficit hyperactivity disorder (ADHD), and autism spectrum disorder (ASD). The second section considers the issue of comorbidity between developmental disorders, and discusses the various models that have been proposed to explain potential overlap.

Part I: Issues in the diagnosis of developmental disorders

Many developmental disorders that affect cognition and behaviour, such as Down syndrome, Fragile X syndrome, and Turner syndrome, have a known genetic basis. Some of these disorders also involve specific physical abnormalities, such as short stature or particular facial features that are characteristic of a particular disorder. The presence of testable genetic markers and outward physical signs mean that such disorders are relatively straightforward to diagnose. Williams syndrome (WS) is a case in point.

Phenotypically, WS is characterised by particular physical abnormalities, including facial dysmorphology (“elfin-like” appearance) and heart disease (most commonly, supravalvular aortic stenosis – narrowing of the aorta). Individuals with WS typically show “hyper-social” personalities but tend to lack skills in social judgement. On the cognitive level, WS is characterised by mental retardation (full-scale IQs usually in the range of 50–60), alongside a somewhat uneven cognitive profile, with relative strengths in expressive language and face processing, and particular weaknesses in visuo-spatial abilities (Karmiloff-Smith, 2008; Martens, Wilson, & Reutens, 2008; see also Filippi & Karmiloff-Smith, this volume).

WS is caused by a deletion of approximately 26 genes on the long arm of one copy of chromosome 7q11 (Peoples et al., 2000). Most significantly, 96 percent of individuals with classic WS show a deletion of one ELN allele (Lowery et al., 1995). The ELN gene codes for elastin, a structural protein found in connective tissue in multiple organs. Hemizygous ELN deletion is thought to result in abnormal elastin production and to ultimately cause the supravalvular aortic stenosis that affects individuals with WS. However, given that ELN is expressed only negligibly in the human brain, its deletion is unlikely to account for the cognitive characteristics of WS (Frangiskakis et al., 1996). Even though ELN deletion cannot completely account for the full WS phenotype, it nevertheless provides a useful genetic marker for the disorder.

ELN deletion can be detected using a chromosomal screening technique called fluorescent in situ hybridization (FISH), which utilises fluorescent probes to detect particular DNA sequences. As is true of virtually all developmental disorders, there is considerable variation in the expression of the WS phenotype. Some cases of WS, in which all the classic clinical signs are clearly apparent, are relatively easy to diagnose on the basis of the clinical phenotype. However, subtler, more difficult-to-diagnose cases in which, for example, facial abnormalities are not obvious or cardiac problems are mild, are not uncommon. FISH screening is particularly useful in such instances and provides an invaluable tool for confirming a diagnosis of WS.

The example of WS clearly illustrates the importance of genetic screening in the diagnosis of particular developmental disorders. However, such techniques can only be utilised when disorders have an established genetic basis. Indeed, there are numerous heritable developmental disorders for which genetic basis is yet to be established. For such disorders, diagnoses must be made purely on the basis of phenotypic characteristics. Although the phenotypic characteristics of some developmental disorders may include outwardly observable physical signs, many disorders involve no such diagnostic clues. Thus, diagnoses must be made on the basis of neurobiological, cognitive, or behavioural markers of the disorder. Dyslexia, ADHD, SLI and ASD are each an example of such disorders. These disorders can be more challenging to diagnose, given that they have no characteristic physical manifestations and no known set of necessary and sufficient genes to allow objective genetic confirmation of a diagnosis.

SLI

SLI is diagnosed among individuals who, despite no frank sensory or neurological dysfunction, and no significant ASD features, achieve scores on standardised tests of language significantly below that expected on the basis of their age and non-verbal abilities. For example, the International Classification of Diseases (ICD-10; World Health Organisation, 1993) specifies that language ability must fall more than 2 SDs below that expected for the individual’s chronological age and at least 1 SD below their non-verbal ability. The SLI consortium (2004) specified that either receptive or expressive language skills should be at least 1.5 SDs below that expected for chronological age and that non-verbal IQ should be at least 80. In a large epidemiological study of SLI, Tomblin et al. (1997) specified that for a diagnosis of SLI, performance on at least two measures of (receptive or expressive) language should be at least 1.25 SDs below the mean (i.e., a standard score ≤ 80), with non-verbal IQ in the normal range (i.e., ≥ 85).

In reality, SLI is substantially heterogeneous, and has several empirically-derived subtypes defined according to profiles of ability across comprehension and expression, and according to the degree to which phonology, grammar, semantics and pragmatics are affected (see Leonard, 2000). However, SLI is a useful umbrella term and a substantial proportion (around 50 percent) of pre-school and school-aged children with SLI present with a common profile of language difficulties, which is characterised by problems in language production and comprehension. Moreover, deficits in phonology and syntax are more severe than are deficits in higher order, lexical, or pragmatic language skills.

Although there are differences between studies and between diagnostic manuals in terms of how severely language must be impaired in order to receive a diagnosis of SLI, agreement is almost universal that language ability must be discrepant from non-verbal ability. However, this can create problems for the detection and diagnosis of SLI, and some have questioned the validity of the criteria. Firstly, although language-impaired children with normal non-verbal IQ (NVIQ) tend to have better outcomes than language-impaired children with depressed NVIQ (e.g., Bishop & Edmundson, 1987; Stothard, Snowling, Bishop, Chipchase, & Kaplan, 1998), this does not show that the language impairment in the former case is qualitatively different from the language impairment in the latter case. Rather, high NVIQ may allow some children to compensate for their language problems, a route not open to those with low NVIQ.

Second, nonverbal ability appears to decline over time among people with SLI, with several studies reporting a drop in NVIQ of 10 points or more across development (Botting, 2005; Mawhood, Howlin, & Rutter, 2000; Tomblin, Freese, & Records, 1992). Therefore, receiving a diagnosis of SLI depends, in part, on the age at which an individual is assessed. A child may be referred at 5 years of age and have a NVIQ of 85, thus meeting criteria for SLI. If the same child had been referred at 8 years of age, their NVIQ could have dropped to 75 and thus they would not meet criteria for the diagnosis.

Related to both of these issues, language impairment in SLI is highly heritable, but the discrepancy between verbal and non-verbal ability appears not to be (Bishop, North, & Donlan, 1995). In twin studies, the heritability of a given trait (i.e., the proportion of variation in a trait that is accounted for by genes) is established by exploring the relative similarity of identical (monozygotic; MZ) and non-identical (dizygotic; DZ) twins on that trait. The basic logic here is that, for a given trait, the greater the degree of similarity between MZ twins (who share 100 percent of genes) relative to the degree of similarity between DZ twins (who share only 50 percent of genes, on average), the greater the contribution of genes to variation in that trait. The size of the difference between MZ and DZ twins is used to calculate the univariate heritability of the trait in question (DeFries & Fulker, 1985, 1988). To illustrate, imagine that one member of an MZ twin pair (the “proband”) and one member of a DZ twin pair each has SLI and each scores 2 SDs below the typical mean on a test of language. Now imagine that the co-twin of the MZ proband also scores 2 SDs below the mean on the same language test, whereas the co-twin of the DZ proband only scores 1 SD below the mean. This would give a heritability estimate for language ability of one (i.e., 100 percent of the variance in language ability is due to genetic variation). Now, in the twin study of SLI by Bishop et al., the heritability of language scores was very high (indeed, depending on which measure of language ability was used, it was close to one), whereas the heritability of the discrepancy between language scores and scores on a test of nonverbal ability was close to zero. This suggests that reference to NVIQ may not be essential when diagnosing SLI, given that language impairment has the same aetiology in SLI as it does in “non-specific” language impairment.

Given the difficulties associated with defining SLI (and other complex developmental disorders; see below), Bishop (e.g., 2006, p. 1153) has argued that we could “cut loose from conventional clinical criteria for diagnosing disorders and to focus instead on measures of underlying cognitive mechanisms. Psychology can inform genetics by clarifying what the key dimensions are for heritable phenotypes”. For the purposes of conducting genetic studies of SLI (and other disorders) and for remediating the core language impairment in the disorder, perhaps defining SLI according to its cognitive endophenotype would be more productive. Generally-accepted criteria for an endophenotype (or “cognitive marker”) are that it is associated with the disorder in question, is present at all stages of the disorder (even if superficial behavioural difficulties have resolved), is heritable, and is present in non-affected family members at levels greater than would be expected by chance (e.g., Gottesman & Gould, 2003). The most promising candidate for a cognitive marker of SLI is diminished nonsense word repetition (NWR). In a NWR test, the participant listens to non-words spoken by the tester, and repeats each immediately after hearing it. It is well established among typically developing (TD) individuals that NWR skills are strongly associated with structural language ability (Baddeley, Gathercole, & Papagno, 1998; Service, 1992), independent of NVIQ (e.g., Conti-Ramsden, Botting, & Faragher, 2001).

Critically, poor NWR distinguishes children with SLI from TD children in over 80 percent of cases (Conti-Ramsden et al., 2001) and even characterises “resolved cases” of SLI who receive an early clinical diagnosis, but who perform in the normal range on broad standardised language measures later in life (Bishop et al., 1995; Conti-Ramsden et al., 2001). Moreover, diminished NWR runs in the families of individuals with SLI (including among non-affected relatives; e.g., Lind-gren, Folstein, Tomblin, & Tager-Flusberg, 2009) and is highly heritable (e.g., Barry, Yasin, & Bishop, 2007). Importantly, the inclusion in molecular genetic studies of NWR as a marker of SLI has become routine, and has led to significant advances in our understanding of the aetiology of the disorder (e.g., SLI Consortium, 2002, 2004).

Reading disorder

Reading disorder is diagnosed when an individual’s “reading achievement, as measured by individually administered standardized tests of reading accuracy or comprehension, is substantially below that expected given the person’s chronological age, measured intelligence, and age-appropriate education” (American Psychiatric Association, 2000). A major difficulty with this definition is that no distinction is drawn between distinct aspects of reading, namely reading accuracy and reading comprehension. An individual with “reading disorder” may be able to decode only a small fraction of printed words that they are exposed to, but may understand all of the words they can decode. In contrast, an individual might be able to decode the majority of printed material they encounter, but understand very little of the meaning of the material. The former kind of reading disorder is termed developmental dyslexia, whereas the latter is termed reading comprehension impairment. We’ll focus our discussion on dyslexia, which is defined by the National Institute of Neurological Disorders and Stroke (http://www.ninds.nih.gov/disorders/dyslexia/dyslexia.htm) as,

a brain-based type of learning disability that specifically impairs a person’s ability to read. These individuals typically read at levels significantly lower than expected despite having normal intelligence. Although the disorder varies from person to person, common characteristics among people with dyslexia are difficulty with spelling, phonological processing (the manipulation of sounds) and/or rapid visual-verbal responding.

Using a criterion of scoring more than 2 SDs below the mean on a measure of reading accuracy, plus normal IQ, Rutter et al. (2004) found that between 3 percent and 6 percent of children in the UK could be classified as having dyslexia. However, as with the diagnosis of SLI, relying on discrepancy scores (in this case between IQ and reading ability) may obscure the underlying problem in dyslexia. Indeed, there is little evidence that greater gains in reading accuracy are made by poor readers with high IQ than poor readers with low IQ (e.g., Hatcher & Hulme, 1999). Nonetheless, as with SLI, it is possible to diagnose dyslexia on the basis of objective performance on standardised measures. Diagnosis of the two disorders that we consider next is rather more complicated.

ADHD

ADHD is defined as a “persistent pattern of inattention and/or hyperactivityimpulsivity that is more frequent and severe than is typically observed in individuals at a comparable level of development” (American Psychiatric Association, 2000). Specifically, a diagnosis requires that an individual shows six or more signs of inattention (e.g., not listening when spoken to directly; being forgetful in daily activities), hyperactivity/impulsiv...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Contents
• List of figures
• List of contributors
• Acknowledgements
• Introduction
• Part I Current issues in the study of developmental disorders
• Part II Reflections on the study of developmental disorders
• Glossary and abbreviations
• Index