The reading process involves a combination of eye movements, including convergence, binocular saccades and fixations. Fixations are needed to focus each word on the fovea whereas saccades are required to change fixation between words or from the end of one line to the beginning of the next (Figure 1).

[CaptionComponent="3015"]

Introduction to Dyslexia

Dyslexia is a type of specific reading disability (SRD) which causes the ‘selective impairment of reading skills despite normal intelligence, sensory acuity (visual awareness of surroundings) and motivation’.2 Its presentation is variable, with inter-individual differences in type and severity of symptoms throughout life (Figure 2). Dyslexia is currently diagnosed using the ‘discrepancy model’, where reading is more than two standard deviations below that expected on the basis of intelligence, as typically assessed by intelligence quotient (IQ).3

[CaptionComponent="3016"]

There are various conflicting theories regarding the causes of dyslexia, but most scientists agree that it is inherited through families. The most often cited theory explaining the condition is known as the phonological processing impairment theory 5 where the sufferer is unable to derive the sound (and consequently unable to extract the meaning) from the word.

Rival theories have, however, identified a link between various aspects of the visual system and poor reading skills symptomatic of dyslexia. A popular, though highly controversial, theory related to the visual system is ‘the so-called magnocellular theory of developmental dyslexia 3 which will be considered in context throughout this series. A range of literature will be reviewed and a balanced argument on the validity of the aforementioned theory will be put forth. The overall aim of the series is to allow the eye care practitioner, firstly, to decide whether the evidence exists to support the view that deficits in the magnocellular processing pathway cause the visual symptoms which manifest in dyslexia.

Dyslexia is a term often used to explain a wide range of presenting reading concerns and it is important that the practitioner firstly understands the underlying processes that may influence such presentations (and even to establish whether these are organic or structural) and secondly to understand the strength of the evidence base underscoring research into the causes of dyslexia as a whole. It is hoped then that they might be in a stronger position when advising patients on a common and often very troubling range of symptoms.

Overview of Visual Processing

In evolutionary terms, the visual system provides a description of the location (where?) and identification (what?) of objects that have a ‘survival value to the species’.6 This interpretation is undertaken in a dynamic visual sensory world where an individual’s focus of gaze and attention is constantly shifting. An excellent example of this fluidity is the process of reading.

Primary Visual Pathway

The primary visual pathway (as represented schematically in Figure 3) begins in the retina where two main types of retinal ganglion cells (parvocellular or P cells and magnocellular or M cells) collect specific visual information from photoreceptors. These ganglion cells then condense the visual signal information into a manageable size to be then transmitted to separate and related M and P layers of the lateral geniculate nucleus (LGN).

[CaptionComponent="3017"]

The LGN is used as a relay point on the way to the area of the brain designated as V1 (the primary visual cortex) and is needed to regulate the flow of visual signals that reach this area. Area V1 processes the visual information and codes aspects of local image features including:

  • Size

  • Orientation

  • Direction of any movement

  • Disparity between the signals from each eye

All of these individual measures of stimulus quality are essential in order to build the overall perceptions of identification of objects (‘what’) and spatial relations (‘where’).7

Both P and M systems are functionally specialised to collect different types of visual information. The P system (what) makes a slower, more sustained response to identify the object whereas the M system (where) makes a rapid, more transient response to draw our attention to new or changing objects in our visual field. A summary of the M and P pathways are described in Table 1.

[CaptionComponent="3018"]

Higher Cortical Pathways

Two pathways emerge from area V1; a dorsomedial (DM) stream and a ventral stream. The DM stream projects onward to various sections of the posterior parietal cortex (PPC) including the area labelled as V5 in the middle temporal region, so usually described as V5/MT (see Figure 3). This is dominated by M rather than P input and is involved in motion vision (seeing moving objects or maintaining vision whilst the eyes themselves are moving).

The DM stream also plays a significant role in the visual guidance of eye and limb movements and it ‘projects onwards to the frontal eye fields and down to the superior colliculus and cerebellum, which are all very important for controlling saccadic eye movements’.9 In contrast, the P system projects via mainly to the inferior temporal cortex. This ‘ventral stream’ is responsible for object and colour perception.

Physiological Data of Dyslexic Brains

In 1991 Livingstone and colleagues published a groundbreaking study for the first time identifying physiological and anatomical abnormalities affecting the M visual pathway in the dyslexic brain.10 The study was divided into two parts:

  • Assessing the responses of subjects identified as dyslexic to various tasks using visually evoked potential (VEP) recordings

  • Observing the anatomy of postmortem dyslexic brains

In the VEP studies, five dyslexics and seven control patients were tested. Each dyslexic patient was stated as being ‘beyond average intellect’ and had been professionally diagnosed as having dyslexia. All control subjects were normal readers and had been matched for age, intelligence, educational background and professional status. Both the transient and steady-state VEP were measured. A rectangular checkerboard stimulus was reversed at 15Hz for the transient VEP and at varying frequencies for the steady-state response. The more noteworthy findings were obtained from the steady state response. Single subject analysis of a dyslexic subject compared with a control subject is demonstrated in Figure 4.

[CaptionComponent="3019"]

The results reveal that the dyslexic subject generates a very diminished, though not absent, cortical evoked response to low-contrast stimuli (1% contrast) yet normal responses to high contrast stimuli (15% contrast). Low contrast is sampled by the M system thus suggesting that dyslexic subjects have deficits in M processing.

Comparison of the various response amplitudes (using the technique of Fourier spectral analysis) showed that this result applied to dyslexic subjects as a group (Figure 5).

[CaptionComponent="3020"]

Figure 5A shows that dyslexic subjects, when looking at a 15Hz stimulus frequency produce a significantly reduced VEP response to low (1% and 2%) contrast but a normal response to higher contrast (15%). Additionally, Figure 5B shows that dyslexic subjects, when looking at low contrasts (0.01) produce a significantly reduced VEP response to a 15Hz stimulus frequency but normal responses to slower stimulation frequencies (e.g. 10Hz or 5Hz).

The researchers hypothesised that the normal response of dyslexic subjects to the 15Hz stimulus at higher contrast could be accounted for either by a speeding up of their M systems or to a high frequency response of the P system. However they were unable to differentiate these two possibilities due to an overlap in responses by the M and P system to high contrast and high spatial frequency which had been assumed after from research on monkeys. In other words, the monkey M system would respond more rapidly to the high contrast stimulus than the low contrast stimulus (suggesting the M system speeds up) but at high (15%) contrast the P system would also respond to frequencies as fast as 15Hz (suggesting a high frequency response of the P system).

In 1993, Galaburda and Livingstone published a further which further identified the marked physiological and anatomical abnormalities in the dyslexic brain.

They had extended the physiological test as performed in the 19 1 study,11 which measured the averaged VEP response to the contrast reversal of a binocularly presented checkerboard pattern, but this time used 4% and 40% contrast. This is significant because 4% contrast was sampled by M cells and 40% contrast was sampled by P cells. At 4% contrast, the VEP between 40 and 90ms differed significantly between in the two populations (Figure 6). Galaburda and Livingstone explained this finding as being due to a ‘small broad negative wave being delayed by 20 to 40ms in the dyslexic sample’.11

[CaptionComponent="3021"]

The physiological results of these two studies illustrate that dyslexic subjects have a reduced sensitivity to ‘low-contrast, fast visual stimulation’ which are suggestive of defects in M visual processing. Reversal stimuli give a big positive gain response at 100ms post stimulus (known as the P100). The longer latency, or in other words delayed P100 to low contrast stimuli seen in dyslexic subjects is consistent with this finding (Figure 6). Furthermore, when looking at low contrast stimulus, abnormalities in the dyslexic VEP response can be seen from 50ms post visual stimulation (Figure 6), something which Livingstone and colleagues concluded to be consistent with early on defects in the M pathway. This would seem to suggest a definite underlying physiological variation among dyslexics.

Anatomical abnormalities in the dyslexic brain

For the second part of their study, Livingstone and colleagues compared the autopsy specimen LGNs taken from five diagnosed dyslexic subject cadaver brains to five control brains that had undergone sufficient assessment throughout life to rule out a diagnosis of dyslexia.10 Images of both the M and P sections of the LGN were taken and digitalised for analysis of neuron cell body sizes.

Figure 7 demonstrates a statistically significant reduction of approximately 25% overall in mean cell area in M layers of autopsy dyslexic LGN when compared to the control but no reduction in mean cell are of P layers of LGN.

[CaptionComponent="3022"]

Additionally it was noted that the M layers were anatomically more disorganised than in the non-dyslexic brains. Conversely, P layers of the LGN of dyslexic brains did not differ in appearance from the P layers of the control brains.

Livingstone and colleagues theorised that the smaller sized M neurons in dyslexic LGNs (as found in histological studies) were likely to have functional consequences; the ‘smaller cell bodies are likely to have thinner axons, which should have slower conduction velocities. These abnormalities might be magnified if there were also defects at earlier or later stages in the M pathway.’

Significance of results

It has been argued that slower conduction between the LGN and area V1 cannot be the only dissimilarity in the visual pathway between dyslexic and non-dyslexic subjects. This is because around a 30% decrease in the M cell axon diameter would result in only about a millisecond delay in geniculocortical conduction time, and even a two- or threefold decrease in diameter would result in only a few milliseconds delay.10 Nevertheless, there is a possibility that the M division of the visual pathway could be effected in dyslexics at many levels and, if so, then the processing abnormalities or delays might be cumulative, resulting in observed delays in the evoked potential and reduced cortical evoked response to low contrast stimuli.10

Pathophysiological abnormalities

The anatomical and physiological evidence for anomalies in the M subsystem of the LGN in dyslexia has more recently been linked to the pathophysiology of cortical motion processing in a study using functional magnetic resonance imaging (fMRI) data.12

Eight adult male controls were matched to six male dyslexics using IQ, age and other behavioural measures. To assess whether there was a detectable visual motion deficit in the dyslexic subjects, responses were assessed using a stimulus velocity judgment task in which subjects saw low contrast dots moving horizontally with 100% coherence. Two such stimuli were presented in succession for one second each, the second differing from the first only in velocity. The first stimulus velocity was the same for all trials (7.0 deg s-1) and the second stimulus velocity ranged from 5.0-9.6 deg s-1. Subjects were instructed to state whether the second motion stimulus was slower or faster than the first. The results (Figure 8) demonstrated a significantly poorer performance in the dyslexics across the entire range of velocities tested (paired t-test, one tailed, P<0.03.).

[CaptionComponent="3023"]

fMRI data

This behavioural deficit was also associated with reduced activation in the higher level motion processing area V5/MT which was measured when the dyslexic subjects viewed the same low contrast moving random dot stimuli while positioned in the MRI scanner (Figures 9 and 10).

[CaptionComponent="3024"]

 

[CaptionComponent="3025"]

Figure 10 shows group analysis of variance (ANOVA) in brain activity of the dyslexic group compared to the control group when responding to low contrast (5%) random-dot stimulus (M stimulus) and when responding to stationary, high contrast (40%), patterned stimulus (P stimulus). Local blood oxygenation level-dependent contrast signals were measured by fMRI.The moving stimulus task generated functional activation in area V5-MT in each control subject (10a) but failed to do so in each dyslexic subject (10b). There is a link between area V5/MT and motion sensitivity; proposing therefore that dyslexics possess a reduced sensitivity to motion. In contrast, presentation of stationary patterns (10 c and d) resulted in equivalent activations in area V1/V2 and extra striate in both groups.

[CaptionComponent="3026"]" width="650" height="902" /> Figure 10: fMRI comparison brain activity of dyslexic males with control males in response to moving, low contrast (5%) random-dot stimulus (a,b) and stationary, high contrast (40%), patterned stimulus (c,d). Red shows areas of increased brain activity. Note areas of brain activity in control subjects which were absent in dyslexic subjects. [simulation based on 12][/caption][CaptionComponent="3027"]" width="650" height="907" /> Figure 10 fMRI comparison brain activity of dyslexic males with control males in response to moving, low contrast (5%) random-dot stimulus (a,b) and stationary, high contrast (40%), patterned stimulus (c,d). Red shows areas of increased brain activity. Note areas of brain activity in control subjects which were absent in dyslexic subjects. [simulation based on 12][/caption]Conclusion

This article has offered a brief outline of dyslexia and shown how there is published evidence for both physiological and anatomical variation between the brains of those classed as dyslexic as compared with those assessed to have normal reading ability. This might serve as some reassurance to those nervous of their stigmatizing label when declared to any sceptical of any underlying physical basis to the condition. However, this measured difference is not considered in typical practical appraisal and assessment systems. In the next article we will delve more deeply into the saccadic inhibition theory and the unstable binocular fixation theory of dyslexia and assess the evidence based on visual testing, such as contrast sensitivity.

Jaskiran Sandhu is a hospital optometrist based in Surrey. The author would like to thank associate professor Simon Grant of City University for his review and input of this paper

References

1 Snowden, R, Thompson, P & Troscianko, T, 2006. Basic Vision: An Introduction to visual perception. 1st ed. Oxford: Oxford University Press.

2 Livingstone, M & Hubel, D, 1987. Psychological evidence for separate channels for the perception of form, colour and depth. The Journal of Neuroscience, 7(11), pp. 3416-3468

3 Stein, J, 2001. The Magnocellular theory of developmental dyslexia. Dyslexia, Volume 7, pp. 12-36.

4 Lexxic Limited, 2013. Dyslexia. [Online] Available at: htpp://www.lexxic.com/2/expertise/8/dyslexia [Accessed 1 April 2013].

5 NHS, 2009. Dyslexia. [Online] Available at: http://www.nhs.uk/Conditions/Dyslexia/Pages/Causes.aspx [Accessed 10 November 2012].

6 Casagrande, V & Marion, R, 2011. Processing in the Primary Visual Cortex. In: L. Levin, et al eds. Adler’s Physiology of the Eye . Edinburgh: Elsevier, p. 586

7 Livingstone, M & Hubel, D, 1988. Segregation of form, colour, movement and depth; anatomy, physiology and perception. Science, New Series, 240(4853), pp. 749-749

8 Evans, B J, 2001. Dyslexia & Vision. 1st ed. London: Whurr Publishers Ltd

9 Goodale, M A, & Milner, A D. (1992). Separate visual pathways for perception and action. Trends in Neuroscience, 15, 20-25.

10 Galaburda, A & Livingstone, M, 1993. Evidence for a magnocellular defect in developmental dysleexia. Ann N Y Acad Sci, Issue 682, pp. 70-82

11 Livingstone, M S, Rosen, G D, Drislane, F W & Galaburda, A. M, 1991. Physiological and anatomical evidence for a magnocellular defect in developmental dyslexia. Proceedings of the National Academy of Sciences of the United States of America, Volume 88, pp. 7943-7947.

12 Talaraich, J & Tournoux, P, 1988. Co-Planar stereotaxic atlas of the human brain. New York: Thieme