REVIEW
CURRENT
OPINION Reading disorders and dyslexia
Charles Hulme a and Margaret J. Snowling b
Purpose of review
We review current knowledge about the nature of reading development and disorders, distinguishing
between the processes involved in learning to decode print, and the processes involved in reading
comprehension.
Recent findings
Children with decoding difficulties/dyslexia experience deficits in phoneme awareness, letter-sound
knowledge and rapid automatized naming in the preschool years and beyond. These phonological/
language difficulties appear to be proximal causes of the problems in learning to decode print in dyslexia.
We review data from a prospective study of children at high risk of dyslexia to show that being at family
risk of dyslexia is a primary risk factor for poor reading and children with persistent language difficulties at
school entry are more likely to develop reading problems. Early oral language difficulties are strong
predictors of later difficulties in reading comprehension.
Summary
There are two distinct forms of reading disorder in children: dyslexia (a difficulty in learning to translate
print into speech) and reading comprehension impairment. Both forms of reading problem appear to be
predominantly caused by deficits in underlying oral language skills. Implications for screening and for the
delivery of robust interventions for language and reading are discussed.
Keywords
decoding, dyslexia, intervention, language impairment, reading, screening
INTRODUCTION DYSLEXIA
Learning to read is one of the key outcomes for early
education and children who have reading difficul- Definition and prevalence
ties often enter a downward spiral of low edu- Developmental dyslexia is the most widely used term
cational attainment and poor employment for children who experience severe difficulties in
prospects with negative consequences for adult learning to decode print. Children with dyslexia
well-being. When we consider problems in learning find it hard to recognize printed words, have great
to read, it is important to make a clear distinction difficulties ‘sounding out’ unfamiliar words, and
between decoding (the accuracy or fluency of read- often also read slowly. In European languages, which
ing aloud) and comprehension (the adequacy of have more regular writing systems than English, the
understanding text). Problems in learning to decode main symptoms of dyslexia are poor reading fluency
(developmental dyslexia) and problems in learning and spelling but the predictors of reading (and
to comprehend text (reading comprehension
impairment) are distinct forms of difficulty, both
a
of which appear to depend principally upon impair- Division of Psychology and Language Sciences, University College
ments of oral language development. As we will London and bSt John’s College, Oxford, UK
outline below, Dyslexia is related to early problems Correspondence to Professor Charles Hulme, Division of Psychology
and Language Sciences, University College London, Chandler House, 2
in oral language development, with persisting prob-
Wakefield Street, London WC1N 1PF, UK. Tel: +44 20 7679 4285;
lems in the development of speech-sound (phono- e-mail: [Link]@[Link]
logical) skills being a particularly important obstacle Curr Opin Pediatr 2016, 28:731–735
to learning to decode print. In contrast, reading
DOI:10.1097/MOP.0000000000000411
comprehension impairment depends critically
This is an open access article distributed under the Creative Commons
upon broader oral language difficulties; particularly Attribution License 4.0 (CCBY), which permits unrestricted use, distri-
problems with understanding word meanings, and bution, and reproduction in any medium, provided the original work is
problems with grammatical skills. properly cited.
1040-8703 Copyright ß 2016 The Author(s). Published by Wolters Kluwer Health, Inc. [Link]
Neurology
Similarly, children who are poor readers are less likely
KEY POINTS to seek out opportunities for reading than good read-
There are two main types of reading difficulty; ers and hence will have less exposure to print. Low
problems with word-level decoding (dyslexia) and levels of print exposure and reading practice will, in
problems with reading comprehension. Some children turn, compromise reading development.
experience both. For many years, studies of the likely cognitive
causes of dyslexia were dominated by case–control
There is a heightened prevalence of dyslexia in
children with a dyslexic parent and being at family risk studies of highly selected clinical groups. Such stud-
of dyslexia is the strongest risk factors for poor reading ies are subject to referral bias and often children with
from the preschool years. comorbid conditions are excluded from study
samples. Arguably a more robust approach is offered
Many children with dyslexia experience language
by prospective longitudinal studies, which follow
delays and difficulties as preschoolers.
the development of children at high risk of dyslexia
Dyslexia is the outcome of multiple risk factors. from the preschool years, to examine the character-
istics of those who go on to be dyslexic to identify
Interventions for poor reading should promote oral
language skills in the preschool years and phoneme causal risk factors.
awareness in the early school years to provide a strong Studies of children at family risk of dyslexia by
foundation for later literacy development. virtue of having a first degree affected relative have
highlighted the crucial importance of language to
literacy development. Several studies of children at
family risk of dyslexia have been completed and
dyslexia) are the same, namely letter knowledge, &&
others are ongoing [6 ]. These studies show a
phoneme awareness and rapid naming (RAN) skills heightened prevalence of dyslexia in the offspring
[1]. of affected parents, with some 44% developing dys-
It is important to emphasize that reading skills, lexia. These studies also show that dyslexia is not ‘all
like many other human characteristics (e.g., weight, or none’. Rather, among children at family risk of
blood pressure) show a continuous distribution in the dyslexia, literacy outcomes are distributed continu-
population. As such, the criteria used for diagnosis are ously with some children, while not qualifying for
to some extent arbitrary (just as for obesity or hyper- the label of dyslexia, nonetheless showing dyslexic
tension). A common criterion for diagnosing dyslexia symptoms, including relatively poor reading
is reading accuracy more than 1.5 SD below the mean, fluency and spelling.
which results in roughly 7% of the population being The findings of family risk studies also show that
identified as dyslexic [2]. Dyslexia is more common in single deficit accounts of dyslexia are inadequate.
males, and is frequently comorbid with other devel- Although such studies confirm that poor phonolog-
opmental disorders such as specific language impair- ical skills (e.g., poor nonword repetition in the pre-
ment (SLI), attention deficit hyperactivity disorder, school and poor phoneme awareness in the school
or developmental coordination disorder [3]. years) are primary risk factors for poor reading,
dyslexia is more likely to be ‘diagnosed’ when
broader language impairments (including difficul-
Risk factors for dyslexia ties with grammar and poor vocabulary) persist into
Dyslexia runs in families and twin studies have estab- the school years. Such findings add to a growing
lished that difficulties with reading and related skills body of evidence that a phonological deficit is one
are highly heritable. More recently, molecular of a number of risk factors for dyslexia that accumu-
genetic analyses have identified many potential can- late towards a threshold [7].
didate genes of small effect associated with individual
differences in reading [4]. Notwithstanding the
importance of genetic factors, the home literacy RELATIONSHIP BETWEEN DYSLEXIA AND
environment [5] and the quality of teaching are likely LANGUAGE IMPAIRMENT
additional influences on reading development.
Furthermore, genes act through the environment Preschool profiles
and both passive and active gene–environment cor- It is well established that children with speech and
relations can be expected to affect literacy outcomes. language difficulties are at risk of literacy problems
Parents with dyslexia not only share genes with their and there are likely to be shared endophenotypes
offspring, but also plausibly may provide a different between SLI, speech sound disorder, and dyslexia
home literacy environment to that found in homes [8]. Bishop and Snowling [9] suggested that dyslexia
where parents do not experience literacy difficulties. and language impairment are both characterized by
732 [Link] Volume 28 Number 6 December 2016
Spoken and written language Hulme and Snowling
poor phonology (a shared risk factor for poor decod- conceptual understanding, repetition, and articula-
ing) but differ in the extent to which broader tion (with repetition and articulation loading least
language difficulties (vocabulary and comprehen- well on this factor, possibly because they also tap
sion) are implicated. However, few studies have speech processes). Language at 3.5 years predicted
made direct comparisons between the disorders as the foundations for decoding: namely letter knowl-
SLI tends to be diagnosed in the preschool years, edge, RAN, and phoneme awareness with the
whereas dyslexia is typically diagnosed much later relationship between language and phoneme aware-
(after a child has been in school for some years but ness being particularly strong. In turn, letter knowl-
failed to make adequate progress in learning to edge and phoneme awareness measured at 4.5 years
read). predicted decoding skills at 5.5 years. Reading com-
&&
Nash et al. [10], Hulme et al. [11 ], and Snowling prehension at 8 years of age was predicted by decod-
&
et al. [12 ] followed the development of children at ing at 5.5 together with language at 3.5 years.
family risk of dyslexia and children with preschool These findings underline the importance of oral
language impairment in the Wellcome Language language development for reading development.
and Reading Project – a 5-year longitudinal study The impact of language development at 3.5 years
from age 3.5 to 8 years. Three groups of preschool on decoding was fully mediated by phoneme aware-
children were recruited to the study: children at ness and letter knowledge, and there was a direct
family risk of dyslexia, children considered to be long range effect of language skills at 3.5 years on
language impaired, and children at low risk of dys- reading comprehension measured some 5 years later
lexia (controls). Following parental assessment of at age 8 years. A clear implication is that children
literacy skills, children were initially classified as with poor language skills in the preschool years are
either at family risk or not at risk. Next, following at high risk of developing both dyslexia and reading
a language assessment, the children were grouped comprehension impairment. However, there is a
according to whether they fulfilled criteria for complication. It is not uncommon for children
language impairment, defined as at least 1 SD below who show delayed language development in the
the mean on measures of receptive and expressive preschool years to ‘catch up’ and for speech diffi-
language skills. This procedure yielded four groups culties to resolve. In this light, Bishop and Adams
of children: typically developing, children with SLI, [14] proposed a ‘critical age hypothesis’ – only when
children at family risk of dyslexia (FR), and children children have problems with language (or speech)
at family risk of dyslexia with SLI. It is noteworthy development that persist to the point of school
that about a third of the children at family risk of entry will they succumb to reading problems.
dyslexia also had a preschool language impairment.
These children did not differ from the SLI-only
group on any phonological or language measure, Trajectories of language development
whereas the FR-only group had difficulties only on To investigate language development in the Well-
phonological tasks that tapped speech production, come Language and Reading Project, and more
namely articulation and word and nonword repeti- specifically, to differentiate children with persistent
tion. They also showed subtle and short-lived diffi- versus resolving language impairments, Snowling
&
culties with grammar (problems with marking the et al. [12 ] examined the progress of individual
past tense – ed, and third person singular – s) and in children at three time points: preschool (3.5 years),
repeating sentences accurately. Gooch et al. [13] school entry (5.5 years), and the middle of primary
went on to show that the children in the SLI groups school (8 years). By assessing each child on tests
were more likely to have difficulties with attention tapping receptive and expressive grammar and
and motor development than children at family risk vocabulary, it was possible to determine whether
without language impairment. or not they fulfilled criteria for SLI at the different
time points. In total, 66% of the children in the
sample had normal language at all three times.
Predictors of individual differences in Among the others, it was possible to discern three
reading trajectories of development: 16% had language
&&
Hulme et al. [11 ] tracked the progress of the chil- impairments that resolved, whereas 56% had per-
dren from the three risk groups and controls at ages sisting language impairments. In addition, some
3.5, 4.5, 5.5, and 8 years and used the data in a path 28% showed late emerging language impairments;
model to investigate the causal relationships these children performed well within the normal
between language and literacy skills. At 3.5 years, range on language tasks at 3.5 years, yet scored as
a single language factor could be defined by per- poorly as the persisting group at age 8 years. It is
formance on tests of vocabulary, grammar, noteworthy that this ‘late emerging’ trajectory was
1040-8703 Copyright ß 2016 The Author(s). Published by Wolters Kluwer Health, Inc. [Link] 733
Neurology
strongly associated with familial dyslexia. In line foundations for reading will be important. There
with the critical age hypothesis, children with early is growing evidence from randomized trials that
language difficulties, which resolved by the start of educational interventions for reading and related
formal literacy instruction tended to have good learning disorders are effective when delivered by
literacy outcomes, but children with both persisting trained practitioners [17]. In general terms, inter-
and later emerging difficulties tended to have read- ventions to promote decoding (and hence remedi-
ing difficulties with more than 40% of each group ate dyslexia) comprise training in phoneme
affected. awareness linked to letter knowledge, together with
systematic phonic instruction in the context of
book reading. In contrast, interventions to promote
Screening and assessment reading comprehension involve work to improve
Given that dyslexia has its precursors in preschool, it oral language skills (including work on vocabulary
may be possible to identify children early with a view and narrative skills) and emphasis on the use of
to providing early intervention. However, screening inferences and metacognitive strategies to ensure
is not straightforward because risk and protective coherent understanding of text [18].
factors interact during learning to read. Using data Fricke et al. [19] evaluated an oral language
from a Finnish longitudinal study, Puolakanaho et al. programme aimed at improving vocabulary, gram-
[15] showed that being at family risk of dyslexia mar, narrative and active listening skills, and also
increases the probability of reading disability. How- assessed its impact on reading. Children receiving
ever, if letter-naming skills develop early, the risk is the intervention made greater gains in oral language
considerably decreased. Similarly, for a child with and narrative skills than a waiting-list control group
poor letter-naming skills at 4.5 and 5.5 years, the who received ‘business as usual’ and their phoneme
probability of dyslexia is lower if he or she has good awareness and letter knowledge also improved.
phonological awareness or efficient RAN. Although the programme contained no activities
In a similar vein, data from the Wellcome directed towards reading, and no gains were made
Language and Reading Project confirmed a chang- in decoding skills, an important outcome was that
ing pattern of prediction of dyslexia at different ages the intervention group showed significant gains in
[16]. Although family-risk status was a stronger pre- reading comprehension 6 months after the inter-
dictor of dyslexia from preschool than language vention had ended. Moreover, these gains were fully
difficulties were, at the time of school entry, poor mediated by gains in oral language.
language skill was a significant risk factor.
Additional predictors in the preschool years
included letter knowledge, phonological awareness, CONCLUSION
RAN, and executive skills and, at the time of school Dyslexia represents the lower end of a continuous
entry, motor skills slightly increased the prediction distribution of reading skills in the population. The
probability. These findings underline the fact that oral language weaknesses that are the precursors of
dyslexia is the outcome of multiple risk factors and dyslexia can be observed in the preschool period.
confirm that children with language difficulties at Being at family risk of dyslexia is an important risk
school entry are at high risk. However, screening for factor and a major proximal cause is in speech
dyslexia does not reach an acceptable clinical level processing (phonological) difficulties. Although
until close to school entry when letter knowledge, dyslexia is primarily a decoding difficulty, many
phonological awareness, and RAN, rather than fam- children also experience reading comprehension
ily risk, together provide good sensitivity and speci- problems associated with co-occurring language dif-
ficity as a screening battery. ficulties. We believe that screening for language
difficulties should be conducted at school entry to
identify children who are at risk of difficulties in
INTERVENTIONS FOR CHILDREN’S learning to read. There is good evidence from
READING AND LANGUAGE DIFFICULTIES randomized trials that language interventions can
We began by arguing that there are two types of ameliorate language weaknesses detected in the
reading disorder and we showed that the predictors early school years. Similarly, specialist teaching
of decoding (and hence dyslexia) differ from the methods are effective in ameliorating the problems
predictors of reading comprehension. It follows that in learning to decode print that are seen in children
different forms of intervention are required for chil- with dyslexia.
dren with dyslexia compared with ‘poor compre-
henders’. The evidence reviewed so far suggests that Acknowledgements
early interventions to strengthen the language None.
734 [Link] Volume 28 Number 6 December 2016
Spoken and written language Hulme and Snowling
9. Bishop DV, Snowling MJ. Developmental dyslexia and specific language
Financial support and sponsorship impairment: same or different? Psychol Bull 2004; 130:858–886.
The study was funded by Wellcome Trust grant 10. Nash HM, Hulme C, Gooch D, Snowling MJ. Preschool language profiles of
children at family risk of dyslexia: continuities with SLI. J Child Psychol
WT082032MA. Psychiatry 2013; 54:958–968.
11. Hulme C, Nash HM, Gooch DC, et al. The foundations of literacy development
in children at family risk of dyslexia. Psychol Sci 2015; 26:1877–1886.
Conflicts of interest &&
This study presents longitudinal data examining the predictors of individual
There are no conflicts of interest. differences in reading in children at high risk of dyslexia. It shows that early
language skills are the foundation for later literacy and their impact on word
decoding is via phonological awareness and letter knowledge
12. Snowling MJ, Duff FJ, Nash HM, Hulme C. Language profiles and literacy
REFERENCES AND RECOMMENDED & outcomes of children with resolving, emerging or persisting language impair-
READING ment. J Child Psychol Psychiatry 2015. doi: 10.1111/jcpp.12497.
Papers of particular interest, published within the annual period of review, have This study examines the developmental language trajectories of children at high
been highlighted as: risk of dyslexia. There are three trajectories: children with persistent language
& of special interest impairments through the preschool and early school years; children whose early
&& of outstanding interest language difficulties resolve and a third group whose difficulties emerge late
around the time of school entry. This latter profile is associated with familial dyslexia
1. Caravolas M, Lervag A, Mousikou P, et al. Common patterns of prediction of and strongly associated with poor literacy.
literacy development in different alphabetic orthographies. Psychol Sci 2012; 13. Gooch D, Hulme C, Nash HM, Snowling MJ. Comorbidities in preschool
23:678–686. children at risk of dyslexia: the role of language ability. J Child Psychol
2. Peterson RL, Pennington BF. Developmental dyslexia. Lancet 2012; Psychiatry 2013; 55:237–246.
379:1997–2007. 14. Bishop DV, Adams C. A prospective study of the relationship between
3. Hulme C, Snowling M. Developmental disorders of language, learning and specific language impairment, phonological disorders and reading retarda-
cognition. Chichester, UK: Wiley-Blackwell; 2009. tion. J Child Psychol Psychiatry 1990; 31:1027–1050.
4. Paracchini S, Scerri T, Monaco AP. The genetic lexicon of dyslexia. Ann Rev 15. Puolakanaho A, Ahonen T, Aro M, et al. Very early phonological and language
Genomics Hum Genet 2007; 8:57–79. skills: estimating individual risk of reading disability. J Child Psychol Psychiatry
5. Sénéchal M, LeFevre J. Parental involvement in the development of children’s 2007; 48:923–931.
reading skill: a 5-year longitudinal study. Child Dev 2002; 73:445–460. 16. Thompson P, Hulme C, Nash HM, et al. Developmental dyslexia: predicting
6. Snowling MJ, Melby-Lervåg M. Oral language deficits in familial dyslexia: a individual risk. J Child Psychol Psychiatry 2015; 56:976–987.
&& meta-analysis and review. Psychol Bull 2016; 142:498–545. 17. Hulme C, Melby-Lervag M. Educational interventions for children’s learning
This is a metareview of all current studies of children at familial risk of dyslexia difficulties. In: Thapar A, Pine D, Taylor E, et al., editors. Rutter’s textbook of
showing that oral language delays and difficulties are the earliest signs of dyslexia child and adolescent psychiatry, 6th ed Chichester: Wiley-Blackwell; 2015.
in the preschool years. pp. 533–544.
7. Pennington BF, Santerre-Lemmon L, Rosenberg J, et al. Individual prediction 18. Snowling MJ, Hulme C. Interventions for children’s language and literacy
of dyslexia by single vs. multiple deficit models. J Abnorm Psychol 2012; difficulties. Int J Lang Commun Disord 2012; 47:27–34.
121:212–224. 19. Fricke S, Bowyer-Crane C, Haley A, et al. Building a secure foundation for
8. Pennington BF, Bishop DV. Relations among speech, language, and reading literacy: an evaluation of a preschool language intervention. J Child Psychol
disorders. Ann Rev Psychol 2009; 60:282–306. Psychiatry 2013; 54:280–290.
1040-8703 Copyright ß 2016 The Author(s). Published by Wolters Kluwer Health, Inc. [Link] 735
