Evidence Check 1: Early Literacy Interventions - Science and Technology Committee Contents

Memorandum submitted by Professor Julian Elliott (LI 33)


  A key difficulty that surrounds the use of evidence in respect of dyslexia is not that there is a dearth of high quality research studies operating at different levels of analysis, ranging from genetics to neuroscience, to cognitive and educational psychology, to classroom practice. Rather, it is the weakness in progressing from one level of analysis to the next, and the misconceptions and misunderstandings that result, which are highly problematic. I try to represent this difficulty by producing a general proposition that would seem to represent the most widely held understandings held by the general public and by many education and medical practitioners.


  Among the wider population of people with reading and other literacy difficulties, there is a subset that has a condition called dyslexia. Genetic and brain functioning studies have secured the argument that this condition has a biological basis. Dyslexia can be identified by clinicians on the basis of specialized tests and assessments. Having diagnosed that an individual is dyslexic, it is then possible to set in train intervention programmes that are geared to remediate the individual's problem. These programmes differ significantly from best practice for "non-dyslexic" poor readers. Failure to recognise the condition is likely to result in incorrect forms of intervention (or no intervention) and ultimately, unrealized potential.

  It is possible to unpick the validity of this widely held view by considering each of the links in this claim. This can be achieved by examining a series of specific propositions that, cumulatively, underpin the logic of the general claim above.


  3.1  Many individuals experience reading difficulties for reasons other than bad teaching, adverse home environment, poor motivation, emotional factors, or severe sensory/physical/mental impairments.

    Comment: This seems be incontrovertible although contrary claims can be found from time to time.

  3.2  For this group, the origin of their reading problems is essentially biological. Genetic factors result in a predisposition or susceptibility to reading failure. The realisation of this failure depends significantly upon environmental factors. This interaction means that searching for a direct genetic link to reading difficulty is unlikely to prove successful. However, this does not rule out the promise of important gains for understanding and intervention in the future.

    Comment: This again is incontrovertible.

  3.3  This underlying biologically-based reading problem can be labelled "dyslexia". Dyslexia can be understood as a reading difficulty that has a genetic basis (or, more accurately, genetic bases, as these are likely to be multiple and heterogeneous).

    Comment: While appropriate for scientific research, this precept cannot translate directly into clinical practice as there is currently no clear genetic test that can be employed to make a differential diagnosis for any given individual.

  3.4  This genetic predisposition has often been held to be linked to problems relating to phonological factors.

    Comment: Even proponents of the phonological theory (eg Torgesen and Snowling) accept that this does not offer a full account There are different accounts of biological foundations of dyslexia and there are likely to be different genetic bases of these foundations.

  3.5  Reading disability has been held to affect 20% of the population (Shaywitz, 2003). Estimates of dyslexia range from 5% to 15%+.

    Comment: Is the there a biological difference between the dyslexic group and a larger reading-disabled group? How does one reconcile these differing estimates with the tendency of some writers (such as Shaywitz, Wagner) to treat reading disability and dyslexia interchangeably? Is it likely that a relatively high proportion of the population has a genetic problem that underpins their difficulty?

  3.6  Dyslexia involves more than decoding and it is likely that the problem transforms itself; from reading slowly to making spelling errors, to having difficulty comprehending. In addition, other weaknesses, not directly related to reading (eg planning and organising oneself) are sometimes seen as a form of dyslexia even if the individual's literacy skills are only mildly problematic.

    Comment: As the range of difficulties that are seen as characterising dyslexia increases, the proportion of the population with the condition expands to the point that the diagnosis becomes meaningless and practitioners increasingly respond in a cynical and unsympathetic fashion. Where should one draw the line?

  3.7  The shift from structural to functional brain studies has resulted in greater understanding of those areas of the brain involved in reading. "Dyslexics" demonstrate reduced activation in some of these areas and hyperactivation in others.

    Comment: These studies typically contrast brain functioning in good and poor groups of readers. They do not differentiate a dyslexic subgroup from within a broader population of individuals with reading difficulties. Thus, it is ingenuous to suggest that brain imaging techniques have diagnostic utility. Currently, one cannot yet use fMRI with individuals for clinical (diagnostic or intervention) purposes. This crucial point is not widely understood by many practitioners or lay audiences.

    There are also other difficulties in using brain studies as indicative of a condition (dyslexia) as, this work involves correlational data. It has been shown, for example, that intervention results in increased activation and, therefore, fMRI merely shows what is happening at any given time rather than an underlying problem that causes reading difficulty.

  3.8  Dyslexia (ie biologically-based reading difficulty) can be accurately diagnosed on the basis of a clinical interview and educational assessment.

    Comment: There is a disingenuous sleight of hand operating here. The link between identifying biological explanations and conceptualisations and offering a differential diagnosis on the basis of presenting symptoms and various cognitive and academic test scores is not at all straight forward.

  3.9.  Dyslexia is typically defined as unexpected poor performance in reading. Thus, it is based upon a comparison with the individual's functioning in other areas. Dyslexics often demonstrate a "sea of strengths" (Shaywitz) in other areas and this can be a signpost for diagnostic purposes.

    Comment: Does this adequately accord with a biological explanation of reading difficulty? Is it not possible that someone with dyslexia would be just as likely to be weak, as to be strong, in other areas? In addition, are there some areas of strength that are necessary/sufficient for a meaningful diagnosis? This seems unclear at present.

    Can you be dyslexic if you are from an unstimulating environment, have a low IQ and demonstrate few, if any, strengths? Surely, the answer must be in the affirmative but, if so, does this not demolish the use of terms such as "unexpected performance" for diagnostic purposes?

  3.10  It is possible to diagnose dyslexia on the basis of the presence/absence of a number of symptoms.

    Comment: We are in horoscope country here. The lists are very long—look hard and you'll invariably find some confirming features; multiple permutations are possible. There is no essential distinguishing element (eg the presence of working memory difficulties) and two identified dyslexics can have almost wholly different sets of symptoms (other than the reading difficulty itself).

  3.11  Differential diagnosis is important because it will lead to appropriate forms of intervention.

    Comment: In my opinion, this is a key issue in respect of the Select Committee's inquiry. It is recognised, even by dyslexia pressure groups, that the forms of intervention typically recommended for dyslexics are equally valid for any child encountering reading difficulties. Thus, a diagnosis of dyslexia (irrespective of its questionable criteria) adds nothing in respect of guidance for intervention.

    What is important is that there are not faulty attributions made about the child's intelligence or motivation. However, it is still possible that the dyslexic child is cognitively weak and that they may be unmotivated—hardly surprising, given the struggle to cope in school that they will typically encounter on a daily basis.

  3.12  It is clinically and educationally useful to differentiate dyslexics from other poor readers.

    Comment: Is it possible to identify significant numbers of poor readers (excluding those with obvious physical/sensory/mental difficulties) who would not be classified as dyslexic? How would they present? What would be the key discriminating features? How would they be treated differently as a result of the diagnosis? Clearly, there are many poor readers who do not get this label.

    Is this differentiation more helpful than calling all poor readers dyslexic until such time that science provides a more valid means of differentiation? In my opinion, prolonging a dyslexic/non-dyslexic distinction within the wider pool of poor readers is arbitrary, does not inform differential intervention, and is potentially harmful to those many poor readers who, for many various reasons, may not be labelled as "dyslexic".

November 2009

previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2009
Prepared 18 December 2009