Dyslexia has been variously described as:
Why, though, is there so much disagreement about dyslexia? In this article, I will suggest that it is due to two things: the way dyslexia is defined, and the fact that reading is a complex skill.
By the end of the 19th Century, researchers knew that damage to specific parts of the brain could impair specific functions. This finding led psychiatrist Emil Kraepelin to devise a classification system for mental and behavioural disorders that still forms the basis for diagnosis today. His system was based on two assumptions:
Doctors had noticed that patients with damage to a particular part of the left hemisphere of the brain had problems with reading. This deficit was named “dyslexia”, meaning “impaired reading”. Teachers noticed that some children with otherwise normal intelligence found it difficult to learn to read. The same descriptive label was applied to these children but a distinction was made between their “developmental dyslexia”, and “acquired dyslexia” in people who could previously read.
Children with reading difficulties often had other problems, for example, with spelling, writing or arithmetic. Later, brain scans revealed structural differences between the brains of people with developmental dyslexia and those of normal readers. The term “dyslexia” is now rarely used to denote simply impaired reading, but refers to a supposed syndrome along the lines set out by Kraepelin: a single disorder with a range of symptoms, caused by atypical brain development.
This concept will be reinforced by the inclusion of dyslexia in the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) due out in 2013. Dyslexia doesn’t look like a syndrome in the DSM, but the standardised measures of dyslexia it refers to are likely to include symptoms additional to reading difficulties. The big problem with syndromes is that it is impossible to tell without further research whether a group of symptoms that co-occur are linked in some way, or whether they just happen to co-occur by chance. It is also challenging to work out whether they originate in the brain or elsewhere, which brings us to the second factor in the dyslexia disagreement, the complexity of reading.
Reading is often referred to as a basic skill, but to the human brain it is complex. In order to learn to read English, children’s brains need to:
None of these sub-skills are simple and together they engage several brain areas. Phonemes are processed in temporal lobes (auditory pathway); auditory processing deficits could make detecting or discriminating between phonemes difficult. Graphemes are processed in occipital lobes (visual pathway); visual processing deficits could impact on graphemes. Blending involves parietal lobes and comprehension the frontal areas of the brain. Blending and comprehension could both be impaired by working memory deficits. Working memory retains information for only a few seconds, so slow information processing could mean that by the time the end of a word or passage was reached, the beginning would be forgotten.
Because reading is complex, a problem with any part of any sub-skill or any relevant area or function of the brain could cause some reading impairment. Minor problems with one or two sub-skills could explain the variation in reading ability amongst “normal” readers – even those with similar home backgrounds and the same teacher. A significant problem with one sub-skill or minor problems with several could result in a diagnosis of dyslexia. The complexity of reading also means that children who have problems with different sub-skills could all end up with the same broad-brush diagnosis.
The most obvious explanation for developmental dyslexia is that if damage to a particular part of the brain causes reading difficulties in adults who can already read, then difficulty learning to read is probably due to damage or abnormal development in the same part of the brain. Unfortunately, it isn’t quite that simple; because children’s brains are still developing, there could be other reasons why they struggle to learn to read.
Reading difficulties could originate at different levels:
Dyslexia tends to run in families, implying that genetic factors are involved. It is often assumed that genes affect only the brain’s hard-wiring. That’s one possibility. Genes and environmental factors can also affect the structure and function of sense organs and the biochemistry of sensory processing – the brain’s source of information. Information itself can affect brain structure. To understand how, we need to look at what happens in the brain during learning.
Babies are born with the basic structure of the brain in place, but many connections between brain cells (neurons) have yet to be formed. Neuroscientists have identified three main changes to neurons during learning: synaptogenesis, synaptic pruning and myelination 1,2. This process appears to occur in all learning, including reading.
Sensory information is transmitted through the brain in the form of electrical impulses that pass from one neuron to another. Electrically activated neurons activate neighbouring neurons via chemical neurotransmitters that pass across junctions called synapses. Neurons develop new synapses (synaptogenesis) when they are activated by novel patterns of sensory stimuli, such as letters or words. If a pattern is repeated, as the letters or words become familiar, some connections between neurons are strengthened and others weaken and die off (synaptic pruning). This leads to the formation of neural pathways. If a pathway is used frequently, its neurons develop a fatty myelin sheath (myelination) that speeds up electrical signals, and relevant tasks (in this case reading) become fast and automatic.
For this process to work efficiently, frequent, consistent sensory input is required. Anything that makes sensory input infrequent or inconsistent can impair or delay learning. Conditions such as recurring glue ear (common and often undetected in young children) could affect the ability to detect and discriminate between phonemes; amblyopia (lazy eye), strabismus (squint) and nystagmus (involuntary eye movements) could affect the ability to detect and discriminate between graphemes. These conditions can clear up spontaneously, but a five-year-old who previously had one of them could already have developed atypical neural pathways that don’t support efficient reading. A child who hears and says “th” as “f”, for example, might need to develop a new pathway for “th” that is robust enough to override the “f” response. By the time a child is diagnosed with dyslexia, the cause might have vanished, leaving a reading difficulty in its wake. Another area of controversy is also explained by the complex nature of reading: interventions.
Heated debate rages over reading interventions – whole language versus analytic phonics versus synthetic phonics. Anecdotal evidence suggests that the Dore system, tinted lenses and overlays or auditory training programmes are effective; studies don’t support those claims. The reason for the controversy is, I suggest, because reading difficulties have different causes.
One thing brains are good at is spotting patterns, so many children can learn to recognise words and understand how spelling works with little adult support. Others need explicit training in recognising phonemes and graphemes and the patterns they form. We know that many children with reading difficulties find it hard to identify and discriminate between phonemes 3. However, it is important to bear in mind that this finding doesn’t rule out other factors. My son can discriminate between the phonemes “i” and “e”, except when they are within words (for example, “pin” or “pen”). He also confuses visually similar letters, such as “h” and “n” or “j” and “i”, and transposes and reverses letters and numerals and can’t blend more than three phonemes. This suggests that he has auditory, visual and working memory issues.
The changes to neurons during learning imply that intensive, systematic synthetic phonics training helps with decoding because it exposes children to frequent, consistent information about phonemes and graphemes, enabling them to form the “correct” neural pathways. However, it might not improve processing speed, working memory capacity or eye movement anomalies, so difficulties with blending, comprehension and stability of visual input could remain.
A similar caveat applies to therapies. Exercises that focus on balance and hand-eye coordination might result in improvements if reading difficulties are due to poor visual tracking, but not if they are due to poor phonological awareness. One would predict the opposite outcomes for auditory training programmes. The issues of coloured lenses and overlays are complex too. However, we now know, for example, that the neurotransmitter dopamine is affected by the amount of blue light entering the eye 4. Dopamine helps maintain muscle tone, so the level of blue light is likely to affect eye muscle function; anecdotal evidence suggests colour therapies are most effective in children who report visual problems with reading. However, because studies tend to lump together all children with dyslexia, regardless of what’s causing their reading difficulties, it is hardly surprising if assessments of teaching programmes and therapies show mixed results.
Research evidence suggests that reading difficulties have multiple causes and that genetic and environmental factors are involved. Impairments could originate in the structure and function of the brain, the sense organs, in sensory processing or in inconsistent or inadequate sensory input. Interventions will be effective only if they address the specific causes of reading difficulties in individual children. “Dyslexia” is still useful as a shorthand term for “impaired reading”, but the assumption that it is a single syndrome has caused considerable confusion.
Sue Gerrard is a researcher with a consultancy specialising in knowledge modelling. She is a former primary teacher and for four years home-educated her son who has autism and difficulties with reading, spelling, writing and arithmetic:
1. Clancy and Finlay, B. (2001). Neural correlates of early language learning. In M. Tomasello & E. Bates (Eds.), Language development: The essential readings. WileyBlackwell.
2. National Research Council (2000), From Neurons to Neighbourhoods: The Science of Early Childhood Development, Ch. 8: The developing brain. Washington D.C.
3. Goswami, U. (2000). Phonological representations, reading development and dyslexia: towards a cross-linguistic theoretical framework, Dyslexia, 6, (2), 133-151.
4. Cowan R.L., et al. (2000). Sex differences in response to red and blue light in human primary visual cortex: a bold fMRI study. Psychiatry Research, 100 (3), 129-38.