Labels for unexplained language difficulties in children: We need to talk

The view from the Tower of Babel

This week saw the publication of a special issue of the International Journal of Language and Communication Disorders, focusing on labels for children with unexplained language difficulties. Two target articles, one by Sheena Reilly and colleagues, and one by me, are accompanied by an editorial by Susan Ebbels, twenty commentaries, and a final paper where Sheena and I join forces with Bruce Tomblin to try to synthesise the different viewpoints. These articles are free for anyone to access.

Terminological battles are often boring and seldom come to any consensus, so why are we putting time into this thorny issue? Quite simply, because it really matters. As we argue in the articles, having a label affects how a children are perceived, what help they are offered, and how seriously their problems are taken. 'Specific Language Impairment' has very poor name recognition compared to dyslexia and autism, despite being at least as common. Furthermore, unless we can agree on some common language, it's difficult to make progress in research, and to discover, for instance, the underlying causes of language difficulties, how common they are in different parts of the world, or what interventions work.

I was first confronted with the full extent of the problem when I tried to analyse the amount of research and research funding associated with different developmental disorders (Bishop, 2010). There are other conditions, notably autism and dyslexia, where there is plenty of debate about diagnostic criteria, or even about whether the condition exists. But even so, the terminology is reasonably consistent. For children's language difficulties, this is not the case - they can be described as cases of language difficulty, disorder, impairment, disability, needs or delay, with various prefixes such as 'developmental', 'specific' or 'primary'. Some researchers will use such labels with precise meanings, often excluding children who have co-existing conditions, whereas others use them more descriptively. This made it extremely difficult to do a sensible internet search to estimate the amount of research funding associated with children's language difficulties.  

The confusion over labels has, I think, also contributed to the lack of public recognition of language difficulties in children. A couple of years ago, I joined together with Courtenay Norbury, Maggie Snowling, Gina Conti-Ramsden and Becky Clark with the goal of remedying this situation. We started a campaign for Raising Awareness of Language Learning Impairments (RALLI) (Bishop et al., 2012), and set up a YouTube channel to provide basic information. We spent some time debating what terminology to use: "Language learning impairment" was our preferred choice, but many of our videos talk of Specific Language Impairment, simply because that is a more familiar label. The lack of an agreed label proved a real stumbling block for our attempts at public engagement, and we decided that, as well as producing videos, one of our goals would be to get the terminology issue discussed more widely, in the hope of achieving some consensus. It was a very happy coincidence that Sheena Reilly and colleagues were crystallizing their own position on this question in an article in IJLDC, and that they, and the Editors, were willing to include my article, and the commentaries of other RALLI founders, in the published debate.

One thing that came across when reading commentaries on our articles was the disconnect between research and practice. One point on which I agree with Sheena and colleagues is that there is no justification for drawing a distinction between children whose language problems are comparable with below average nonverbal ability, and those who have a mismatch between good nonverbal skills and low language. Research has failed to find any difference between children with uneven or even nonverbal-verbal profiles in terms of responsiveness to intervention or underlying causes. Such a distinction is, however, widely used in educational and clinical settings to decide which children gain access to extra support in school.  Another issue raised by the Reilly et al paper is whether it is logical to use other exclusionary criteria, and to distinguish, for instance, between children who do and don't have autistic features in association with a language problem.  

As Susan Ebbels noted in her editorial, in everyday settings "diagnostic labels and criteria were being used creatively in disputes over access to services both by those seeking to obtain services for children (often parents and their lawyers) who could be accused of ‘diagnostic shopping’ and also by those seeking to deny services (often due to financial constraints) who may use particularly restrictive criteria in order to reduce the number of children qualifying for services". 

We can't afford to ignore this confused situation any longer. The time has come to have a wider debate on these issues, with the aim of reaching a consensus about how terms are used. The Royal College of Speech and Language Therapists has set up a moderated discussion forum where people can give their views on the best way forward. Please do consider adding your voice: it is important that all those affected by this issue have a say, whether you are a speech-language therapist/pathologist, psychologist, teacher, health professional, legal expert, policymaker, a parent of a child with language difficulties, or someone who has experienced language difficulties. We'd also love to hear from those outside the UK - whether English-speaking or not. You can access the discussion forum here.

Finally, to raise awareness of this debate, during the week of 24th-31st August I will be taking over  the @WeSpeechies Twitter handle as guest curator. On Tuesday 26th at 8.a.m. BST there will be a live twitter debate on this topic. Feel free to join in, even if you aren't a regular tweeter.

References
Bishop, D. (2010). Which Neurodevelopmental Disorders Get Researched and Why? PLoS ONE, 5 (11) DOI: 10.1371/journal.pone.0015112  
Bishop, D., Clark, B., Conti-Ramsden, G., Norbury, C., & Snowling, M. (2012). RALLI: An internet campaign for raising awareness of language learning impairments Child Language Teaching and Therapy, 28 (3), 259-262 DOI: 10.1177/0265659012459467

Slides on this topic are available here.

Addendum Friday 29th August 2014

We've had a great week of interactions on Twitter. A transcript for the week is available here.

I'll look through this and aim to organise the material in due course, but meanwhile would encourage anyone who is interested to continue the discussion on Twitter. I'm appending below some tweets that I generated throughout the week to generate debate.

As noted above, the chat links in to a special issue of the Internat. J Lang. Comm Dis which is free to access here http://t.co/ncTUaYvyoI.  NB it is not all that obvious but there are 10 commentaries after each target article.

If you want to join the discussion on Twitter, feel free to comment at any time, but, please include the #WeSpeechies hashtag, so we can aggregate comments easily. Also if your comment relates to a numbered question, please add Q1, etc so we can relate them.

Monday started with my attempt to summarise each of the  twenty commentaries in a Tweet-length message.

Summaries from commentaries

Paediatricn Gillian Baird: ICD &DSM classifications talk of 'language disorder'; implies distinct from normal variation.  Disorder’ used for conditions without obvious aetiology; functional effect described separately in ICFDH.

Lauchlan/Boyle, ed psych view. Must ask: ‘Will label change the child's life for the better? Aetiology often irrelevant

Bellair et al: community SALTs. No one label works for both research & clinical. SLI has problems but we can manage them.

Mabel Rice: "SLI has yet to receive widespread adoption in clinical practice, in spite of the great need for it." critical of DSM5: excluded "well-researched category of SLI", included SCD, "with a minimal research base"

Kate Taylor SLP. SLI underidentified. Changing the term won't resolve the issue, which is one of measurement rather than label.

Conti-Ramsden: Any Consensus Panel on terminology must be international and include voices from different languages,

Hansson et al: ICD10 labels don't map on to use by researchers in Sweden . : Sweden: phonological & grammatical difficulties seen as part of language impairment. Soc comm probs separate

Clark & Carter: Survey:Scottish SALTs unclear re terms & diagnostic criteria. Move from exclusionary to inclusionary criteria.

Hüneke & Lascelles http://t.co/9rVKJzoBZV. Concern that watering down terminology will mean kids lose scarce resources. Prefer medical term 'developmental dysphasia' that gets problems taken seriously

Strudwick/Bauer http://t.co/GSY5Xwz283 Concern that labels don't capture comorbidities; most ch with 'SLI' have other problems

Michael Rutter, psychiatrist "both clinical & research classifications needed but they require a different approach"

Rutter: Specific’ implies ‘pure’ language impairment; "not supported by any of the available evidence"

Larry Leonard: Many researchers already use broader definition of SLI: do not use term to mean children have a pure profile. communicatn with the public/other disciplines will be even harder if we adopt generic label ‘language impairment.

Snowling: DSM5 treats Communication Disorders separately from Specific Learning Disorders, yet they often co-occur

Aoife Gallagher,SALT; ethical issue:"who owns diagnosis once it has been given.. who ultimately has the right to take it away"

Andrew Whitehouse: ‘SLI’ provides neat criteria for researchers but label hides behavioural & aetiological heterogeneity

Dockrell/Lindsay Educational perspective re SLI is missing yet day-to-day support of learning/development provided by teachers. in England ‘speech, language & communication needs’ (SLCN) indicates primary need is with language & communication

Grist & Hartshorne: http://t.co/QKeQbQFsdy Children & young people we work with rarely describe selves as having SLI or SLCN

Norbury @lilaccourt Relaxing diag criteria will increase demand for services.SALTs shld focus on severe & persistent impairmts

Parsons et al @wordaware Shockwaves through SALT profession if nonverbal IQ criteria and delay/disorder distinction removed .Use of marketing approaches to development of a new term, including consultation with parents & young people.

Wright: legal perspective Much time spent in tribunal appeals arguing re labels: eg is it delay or disorder, is it specific?

Questions for debate

On Tuesday we had a live twitter chat with four question topics, and later in the week, I added further numbered question. Here is the total list – we'd love to hear your thoughts on any or all of these:

Q1 What is your view on use of the diagnostic label SLI? Does it reflect a medical model and is this appropriate.

Q2 is What are appropriate criteria for identifying children's language problems

Q3; Should IQ, ASD features, hearing loss determine whether language-impaired children can access services?

Q4 What terminology is most appropriate for children who have unexplained language problems?

Q5 ICD11 will use'Developmental Language Disorder' and DSM5 uses 'Language Disorder'. What do people think of these terms?

Q6 In research SLI still widely used but without requiring IQ discrepancy. Should we retain SLI but with this broader meaning, or is it just confusing?

Q7 In UK education, Speech, Language and Communication Needs (SLCN) is popular term. Is it used outside UK? Is it useful?

Q8 In UK clinical practice, distinction between language 'delay' & 'disorder' is used, but it has no research support.  Where does delay/disorder distinction come from? How defined?

Q9 Is there any support for a return to the more medical term 'developmental dysphasia'?

Q10. Reilly et al and several commentators suggest we drop 'Specific' and use the term 'Language Impairment' instead .What wld be advantages (e.g. avoids unfair exclusion) and disadvantages (e.g. too broad)?

Q11 What do people think of terms 'Language Learning Impairment' or 'Primary language impairment'? '

Q12 Do diagnostic labels actually help children and families?

Q13 Shld terminology/diagnostic criteria be responsibility of speechies, or shld other professions & families have a say? Assumptions/practices seem v. different in education/medicine/psychology vs speech-language therapy/pathology

Q14 In yr area, who does intervention with kids whose language problems are associated with autism?

Q15 Some  people take pride in identifying themselves as dyslexic. Does this ever happen for kids with language problems? If not, why not?

Q16 Has anyone encountered situation where child not offered intervention bcs language problems attributed to social deprivation?

Q17 Insurance considerations seldom important in UK, but affect label use elsewhere. Do US insurers just require DSM?

Overhyped genetic findings: the case of dyslexia

A press release by Yale University Press Office was recently recycled on the Research Blogging website*, announcing that their researchers had made a major breakthrough. Specifically they said "A new study of the genetic origins of dyslexia and other learning disabilities could allow for earlier diagnoses and more successful interventions, according to researchers at Yale School of Medicine. Many students now are not diagnosed until high school, at which point treatments are less effective." The breathless account by the Press Office is hard to square with the abstract of the paper, which makes no mention of early diagnosis or intervention, but rather focuses on characterising a putative functional risk variant in the DCDC2 gene, named READ1, and establishing its association with reading and language skills.

I've discussed why this kind of thing is problematic in a previous blogpost, but perhaps a figure will help. The point is that in a large sample you can have a statistically strong association between a condition such as dyslexia and a genetic variant, but this does not mean that you can predict who will be dyslexic from their genes.

Proportions with risk variants estimated from Scerri et al (2011)

In this example, based on one of the best-replicated associations in the literature, you can see that most people with dyslexia don't have the risk version of the gene, and most people with the risk version of the gene don't have dyslexia. The effect sizes of individual genetic variants can be very small even when the strength of genetic association is large.

So what about the results from the latest Yale press release? Do they allow for more accurate identification of dyslexia on the basis of genes? In a word, no. I was pleased to see that the authors reported the effect sizes associated with the key genetic variants, which makes it relatively easy to estimate their usefulness in screening. In addition to identifying two sequences in DCDC2 associated with risk of language or reading problems, the authors noted an interaction with a risk version of another gene, KIAA0319, such that children with risk versions in both genes were particularly likely to have problems.  The relevant figure is shown here.

Update: 30th December 2014 - The authors have published an erratum indicating that Figure 3A was wrong. The corrected and original versions are shown below and I have amended conclusions in red.

Corrected Fig 3A from Powers et al (2013)

Original Fig 3A from Powers et al (2013)

There are several points to note from this plot, bearing in mind that dyslexia or SLI would normally only be diagnosed if a child's reading or language scores were at least 1.0 SD below average.

1. For children who have either KIAA0319 or DCDC2 risk variants, but not both, the average score on reading and language measures is at most no more than 0.1 SD below average at most.
2. For those who have both risk factors together, some tests give scores that are from 0.2 to 0.3 SD below average, but this is only a subset of the reading/language measures. On nonword reading, often used as a diagnostic test for dyslexia, there is no evidence of any deficit in those with both risk versions of the genes. On the two language measures, the deficit hovers around 0.15 SD below the mean.
3. The tests that show the largest deficits in those with two risk factors are measures of IQ rather than reading or language. Even here, the degree of impairment in those with two risk factors together indicates that the majority of children with this genotype would not fall in the impaired range.
4. The number of children with the two risk factors together is very small, around 2% of the population.

In sum, I think this is an interesting paper that might help us discover more about how genetic variation works to influence cognitive development by affecting brain function. The authors present the data in a way that allows us to appraise the clinical significance of the findings quite easily. However, the results indicate that, far from indicating translational potential for diagnosis and treatment, genetic effects are subtle and unlikely to be useful for this purpose.

*It is unclear to me whether the Yale University Press Office are actively involved in gatecrashing Research Blogging, or whether this is just an independent 'blogger' who is recycling press releases as if they are blogposts.

Reference
Powers, N., Eicher, J., Butter, F., Kong, Y., Miller, L., Ring, S., Mann, M., & Gruen, J. (2013). Alleles of a Polymorphic ETV6 Binding Site in DCDC2 Confer Risk of Reading and Language Impairment The American Journal of Human Genetics DOI: 10.1016/j.ajhg.2013.05.008
Scerri, T. S., Morris, A. P., Buckingham, L. L., Newbury, D. F., Miller, L. L., Monaco, A. P., . . . Paracchini, S. (2011). DCDC2, KIAA0319 and CMIP are associated with reading-related traits. Biological Psychiatry, 70, 237-245. doi: 10.1016/j.biopsych.2011.02.005
 

Neuronal migration in language learning impairments: a suggestion

Specific language impairment (SLI) and dyslexia are related developmental disorders in which a child has difficulty learning to talk (SLI) or to read (dyslexia). Many children have both problems, although they can occur separately (Bishop & Snowling, 2004), and they are sometimes grouped together as ‘language learning impairments’. There's good evidence that genes are implicated in causing these conditions (Bishop, 2009).

A popular account maintains that the genes implicated in language learning impairments affect a very early process in the developing brain known as neuronal migration (Galaburda et al., 2006). It’s an attractive theory that has the potential to provide a link from genes to behaviour. However, when I looked at the evidence, I found myself not entirely convinced. Here I’ll briefly review research on this topic, explain my reservations, and conclude by proposing a study that needs doing. I’m not an expert in neuroanatomy or neuroimaging, so I’ll be interested to see if others think this proposal is sensible.

Abnormalities found in 1979 case report. Solid circles show ectopias/dysplasias, and shaded area shows micropolygyria (based on Galaburda et al, 1985) .

Over thirty years ago, Galaburda and Kemper published a post mortem study of the brain of a man with developmental dyslexia who died from an accidental fall at the age of 20 years. He’d had delayed language development, and was diagnosed with dyslexia in the first grade. His Stanford-Binet IQ of 105 was well in advance of his reading attainments. He developed epilepsy at 16 years of age. His brain showed areas of displaced neurons (ectopias) in the left cerebral hemisphere, especially around the left planum temporale. There was also an area of polymicrogyria, i.e. excessive number of small convolutions, giving a lumpy appearance to the cortex. This raised the possibility that we might find the origins of dyslexia not in the gross features of brain structure, but at the microscopic level, in the organisation of neurons. However, as the authors noted: “It is not possible to tell from a single case whether or not the anatomical findings have any causative relationship to the clinical findings – much less whether the malformation is responsible for the seizure disorder, the learning disability, both, or neither” (p. 99). They also noted that the kinds of neuroanatomical abnormality that they found in their patient were probably too rare to explain dyslexia in general, which has a prevalence of around 5-10% in the population.

A subsequent report added further evidence for a link to dyslexia (Galaburda et al, 1985). Similar abnormalities were found in three further post-mortem cases, and in none of these was epilepsy described, though one had delayed speech and one had “notable language difficulties”. Three additional cases, this time of female dyslexics, were reported by Humphreys et al (1990), but these were less compelling: the evidence for migrational abnormalities was less strong, and other pathologies could have been implicated.

There’s a general problem with the methodology of these studies, which is that they were not conducted blind. The cellular abnormalities that were described require an expert eye and clinical judgement, and you wouldn’t necessarily see them unless you were looking for them. Could they just be spurious findings? Galaburda and colleagues noted that similar anomalies are sometimes reported as incidental findings in unselected autopsy brains, and so a key question was whether the findings in dyslexic brains were really unusual. Accordingly, Kaufman and Galaburda (1989) analysed ten control brains using identical procedures to those used for dyslexic brains. They found abnormal cells in three control brains, but the anomalies were far less numerous than those seen in the dyslexic brains. This provides useful context, but ideally, we need a study where the neuroanatomist is given both dyslexic and control brains and asked to analyse them without knowing which was which, to avoid the perceptual and cognitive biases that can affect even the most scrupulous of observers.

The anomalies described by Galaburda and colleagues reflect disruption at an early stage of brain development, when neurons are being formed and organised into coherent structures. This website from Pasco Rakic has some nice animations showing how a brain is formed when neurons are first generated in the foetus. Neurons formed in the ventricular zone travel out to the surface of the cortex along radial glial fibres, gradually building up six distinct layers of the cortex from the inside out. Studies with rodents, and evidence from humans with developmental disorders, indicate that this process can be disrupted in a range of ways. In some people, a proportion of cells fail to migrate at all, and can be seen as clusters of abnormal cells around the ventricles. This condition, known as periventricular heterotopia, does not normally impair cognitive function but does cause epilepsy. In other cases, there is partial migration followed by arrest, leading to lissencephaly, typically associated with epilepsy and severe intellectual impairment(Guerrini& Parrini, 2010). In mice, a naturally-occurring genetic mutation leads to the phenotype of the reeler mouse, which has severe motor co-ordination problems linked to disorganisation of the usual laminar structure of the cortex, because the migrating neurons fail to penetrate to the surface of the brain. The cases studied by Galaburda and colleagues had a range of anomalies, described as ectopias, dysplasias, heterotopias, ‘brain warts’ and polymicrogyria, associated with disruption affecting different stages of neuronal migration and postmigrational development (Barkovich et al, 2012).

What makes this work exciting is a potential link to genetic studies of dyslexia. There are replicated associations of dyslexia with several genes, including DYX1X1, KIAA0319, DCDC2 and ROBO1. As Galaburda et al (2006) noted in their review, mutations of these genes have been linked to migrational anomalies in rodents. It looks, therefore, as though the route from brain to behaviour could be neatly explained by postulating a genetic influence on neuronal migration that leads to a brain that is not optimally connected.

Some puzzles, however, remain. First, the genetic variants associated with dyslexia are not mutations. They are common in the general population. Associations with dyslexia are found in studies with very large samples, but they are not very strong. For instance, one can deduce from the published data on the KIAA0319 locus that there is a low-risk version of the gene that is found in 39% of normal readers and 25% dyslexics, and a high-risk version that is found in 30% of normal readers and 35% dyslexics. If the dyslexic risk variant causes anomalies of neuronal migration, then we should see lots of people with those anomalies, many (most) of whom will not be dyslexic. Of course, it is all a matter of degree; it is possible that each risk variant has only minor effects on neuronal migration, and causes problems only if it occurs in conjunction with other genetic or environmental risks. Neuronal migration can be affected by environmental factors, such as toxins, nutrition, and disease or trauma affecting the brain. So the ubiquity of these risk alleles does not rule out a causal route via neuronal migration mechanisms, but it does make the story more complicated.

What if we look at the association between neuronal migration disorders and dyslexia from the other direction, i.e. assessing reading ability in individuals with known migrational abnormalities? Chang et al (2005) did this in people with periventricular nodular heterotopia - a disorder in which a proportion of neurons fail to migrate from the ventricular zone. Most of their participants had normal range IQ. On the Wide Range Achievement tests of reading and spelling, their mean scores were average or above-average. Many of them did, however, do poorly on the Nelson-Denny reading test and on this basis, the authors concluded they were dyslexic. But this test, which stresses speed, was designed for college students, not for the general population. The fact that most participants were older than college students, and all were on anti-epileptic medication, makes the claim of dyslexia in these people far from convincing. Minimally, this study should have included a comparison group to control for age, background and medication status.

A final issue is why migrational abnormalities haven’t been noted in MRI studies of dyslexia. In studies of children with specific language impairments, a Brazilian group has reported remarkably high rates of polymicrogyria (De Vasconcelos Hage et al, 2006). However, this does not seem to be a general explanation for SLI. My colleagues tell me there were no cases of this in people with SLI who participated in a recent MRI study that we published, and none was mentioned in a series reported by Webster etal (2008). MRI studies of dyslexia have been considerably more numerous, yet, as far as I can establish, none has mentioned migrational anomalies. Of course, many MRI studies focus on averaged data, which would mask individual variations. So, a key question is whether the failure to report migrational abnormalities in MRI studies is because (a) no-one was looking for them, (b) they are too subtle to see on regular MRI scan, or (c) they aren’t involved in most cases of language learning impairments.

I was intrigued by this question, so I looked for literature on detectability of neuronal migration anomalies on MRI scan. My impression is that these wouldn’t necessarily be detected unless you were looking for them, and if you were, detectability depends on the type and location of anomalies. Wagner et al (2011) devised an automated method of MRI analysis that was successful in picking up 82% of Type IIA cortical dysplasias and 92% of Type IIB, compared to 65% and 91% detected by an expert neuroradiologist. Periventricular nodular heterotopia seems a more obvious pathology that is routinely detected on MRI scan.

On this basis, I’d say there’s a study out there crying out to be done. There are plenty of reports of MRI scans comparing dyslexic vs control brains. We could revisit those scans using the automated methods developed by Wagner et al to test the hypothesis that the rate of neuromigrational anomalies is higher in the dyslexic vs control samples. It’s clear that MRI scans won’t pick up everything, and subtle anomalies may be missed. However, if the neuronal migration account of language learning impairments is correct, we should nevertheless expect to see a measureable difference in the rates of anomalies between cases of dyslexia/SLI vs. controls. And if genetic information is available as well, then a comparison could be done between those with and without risk variants.

References

Barkovich, A. J., Guerrini, R., Kuzniecky, R. I., Jackson, G. D., & Dobyns, W. B. (2012). A developmental and genetic classification for malformations of cortical development: update 2012. Brain, 135(5), 1348-1369. doi: 10.1093/brain/aws019

Bishop, D. V. M. (2009). Genes, cognition and communication: insights from neurodevelopmental disorders. The Year in Cognitive Neuroscience: Annals of the New York Academy of Sciences, 1156, 1-18.

Bishop, D. V. M., & Snowling, M. J. (2004). Developmental dyslexia and Specific Language Impairment: Same or different? Psychological Bulletin, 130, 858-886.

Chang, B. S., Ly, J., Appignani, B., Bodell, A., Apse, K. A., Ravenscroft, R. S., . . . Walsh, C. A. (2005). Reading impairment in the neuronal migration disorder of periventricular nodular heterotopia. Neurology, 64(5), 799-803.

De Vasconcelos Hage, S. R., Cendes, F., Montenegro, M. A., Abramides, D. V., Guimarães, C. A., & Guerreiro, M. M. (2006). Specific language impairment: linguistic and neurobiological aspects. Arquivos de Neuro-Psiquiatria, 64, 173-180.

Galaburda, A. M., & Kemper, T. (1979). Cytoarchitectonic abnormalities in developmental dyslexia. Annals of Neurology, 6, 94-100.

Galaburda, A. M., Sherman, G. F., Rosen, G. D., Aboitiz, F., & Geschwind, N. (1985). Developmental dyslexia: four consecutive cases with cortical anomalies. Annals of Neurology, 18, 222-233.

Galaburda, A. M., LoTurco, J. J., Ramus, F., Fitch, R. H., & Rosen, G. D. (2006). From genes to behavior in developmental dyslexia. Nature Neuroscience, 9, 1213-1217.

Guerrini, R., & Parrini, E. (2010). Neuronal migration disorders. Neurobiology of Disease, 38, 154-166.

Wagner, J., Weber, B., Urbach, H., Elger, C., & Huppertz, H. (2011). Morphometric MRI analysis improves detection of focal cortical dysplasia type II Brain, 134 (10), 2844-2854 DOI: 10.1093/brain/awr204

Webster, R. I., Erdos, C., Evans, K., Majnemer, A., Saigal, G., Kehayia, E., . . . Shevell, M. I. (2008). Neurological and magnetic resonance Imaging findings in children with developmental language impairment. Journal of Child Neurology, 23(8), 870-877. doi: 10.1177/0883073808315620