The widely supported phonological processing impairment theory hypothesises that dyslexics are unable to derive the sound, and consequently unable to extract the meaning, from words.1 This theory proposes that specific reading difficulty (SRD) children have a ‘poor ability at distinguishing phonemes’, so for example ‘tend to confuse sounds such as s, sh, th, f and v’. 2 However, these problems dyslexics face with recognizing and distinguishing ‘speech sounds’ do not explain the vast array of visual problems that they also suffer from. These visual problems are summarised in Figure 1.
[CaptionComponent="3041"]
For example, the superimposition of words is difficult to explain by phonological processing impairment. There have been, however, suggestions for it to be explained by reference to the transient (M) visual pathway 3 (as introduced in the first article in this series). 4 This theory was widely accepted as an explanation for the visual problems dyslexics face and has been mentioned in various credible papers about deficits in the M system and dyslexia.5,6
Transient Pathway Differences
Response in ‘normal’ readers
It is possible to study the M and P channel responses during the eye movements made when a non-dyslexic subject reads. 11
[CaptionComponent="3042"]
Figure 2 illustrates the pattern of eye movements made during reading. In a hypothetical response, the reader fixates on each word for 250ms and then makes a rapid saccadic eye movement (2A).
2B shows the P channel response which occurs in fixations and can last for several hundred milliseconds to provide details of what the eye is seeing. In comparison the M channel response, as seen 2C, is very sensitive to stimulus movement and gives a much shorter duration response. It inhibits the P channel response which would persist from a preceding fixation and potentially interfere with the succeeding one. Consequently, both channels are involved in normal reading (2D).9
Response in dyslexic readers
Any problem in either the P or M response channel or in the way they interact may have harmful consequences for reading. The M system was thought to be ‘important in refreshing the image each time your eyes move; if this replacement of one image by the next is not working properly the successive images from one fixation to the next become confused and reading is difficult.’ 10
Breitmeyer’s Saccadic Inhibition Theory
Breitmeyer proposed that deficits in the M response channel imply readers will suffer visual persistence, with words superimposing and accounting for a common dyslexic reading problem. Using Breitmeyer’s theory, Figure 3 illustrates what can occur when reading a line requiring up to three fixations.
[CaptionComponent="3043"]
This theory is dependent on the notion that saccadic eye movements have two functions; changing visual fixations as well as signaling the M channel response to inhibit the P channel response 11,12 as demonstrated in Figure 2C.
Support for Breitmeyer’s explanation
The importance of the M channel, and how deficits in this channel can cause symptoms found in SRD people, is summarised below. Consequences of M channel deficits include:
- Weakened saccadic suppression. This results in at least a partial temporal overlap, rather than a clear temporal segregation, of successive frames of retinotopic sustained activity from sustained fixations
- Increased retinal image smear during saccades
- Reduced constancy of visual direction
- Weakened images produced on the retina when scanning the page
These symptoms have been found in about 60 to 70% of dyslexics.13 In 1993, Breitmeyer suggested that it may be ‘more than coincidental’ that this is also the proportion of SRD children noted by Lovegrove 9 also suffering from a M channel deficit. However, this means that the remaining 30 to 40% of dyslexics suffer a different problem. This will be discussed later.
Manipulations to visual processing
If Breitmeyer’s saccadic inhibition theory is correct, then manipulations to visual processing that slow the P system more than the M system (for example low contrast) should theoretically ‘restore the proper temporal relationship’ thus alleviating the dyslexic visual symptoms. Evidence of this can be seen in results from certain studies.
Giddings and Carmean 14 tested comprehension skills when reading a page of text among 51 college students, 21 of whom had learning disabilities. They found that ‘comprehension of the control group was little affected by the contrast’ of black letters on a white page; however, the mean score for the learning disabled students was 10% higher on the pages with reduced contrast thus supporting Breitmeyer’s theory.
Evidence against Breitmeyer’s explanation
Dyslexics tend to confuse neighbouring letters in a word and commonly complain that ‘letters appear to move around in words’. The importance of the first symptom is that the confused letters are neighbouring and, in other words not ‘separated by 6-8 characters or about 2 degrees of visual angle’, which is approximately the average saccade length when reading.15 Breitmeyer’s theory could potentially explain why letters in adjacent words in sentences are confused. However, on closer inspection, it is noted that dyslexics do not confuse neighbouring words together, only neighbouring letters, so suggesting that the theory is inaccurate.
Strong evidence, in the form of a study by Burr,16 further disputed Breitmeyer’s theory. In the study, Burr and colleagues used four normal subjects (two of whose results are shown in Figure 4) to discriminate the luminance of a horizontal strip during a saccadic eye movement.
[CaptionComponent="3044"]
As seen in Figure 4, for both observers, luminance discrimination was strongly impaired at the beginning of the saccades, steadily recovering to normal levels about 150 milliseconds later. But equi-luminant chromatic red/green stimuli showed increased sensitivity during saccades, reaching maximal levels at latencies of around 50 to 100ms. Burr noted that the magnitude of this saccadic enhancement was variable but clearly evidenced by all four observers (two of whose results have not been shown in Figure 4) for both discrimination and detection.
These results suggest that saccades may ‘selectively suppress the M pathway’, which is involved in luminance discrimination (possibly at the level of the LGN) while ‘sparing or even enhancing the P pathway’ which is involved in colour discrimination, for example between red and green, the complete opposite to Breitmeyer’s theory.
Contrast Sensitivity Studies
The theoretical contrast sensitivity curve (Figure 5) describes our ‘window of visibility’; any information that lies above the curve is invisible, so we can only see the visual information below the curve. It is made up by an overlap by the M and P pathway and represents the ‘sensitivity of whichever of the two systems is most sensitive at any given spatial frequency’.17
[CaptionComponent="3045"]
One way of assessing whether dyslexic individuals show a reduced sensitivity in M processing would be via plotting their contrast sensitivity (CS) curves because any defects in the M or P pathway would alter the shape of the CS curve. Livingstone et al revealed that dyslexics have reduced sensitivity to low contrast indicating a depression in the M mediated section of the CS curve.5
Shape of CS curve in dyslexics
There are a lot of conflicting studies and controversy regarding the shape of the CS curve in dyslexics and hence whether dyslexics have an M processing deficit. Skottun, an advocate against the M pathway, compiled a review of the various studies, counting studies which measured temporal and spatial contrast frequency.17
The results of the 22 studies on spatial contrast sensitivity collected by Skottun are summarized in Table 1. Skottun had grouped the studies into eight categories depending on the nature of the sensitivity loss, or lack thereof.
[CaptionComponent="3046"]
The conclusions drawn from Table 1 depend on the criteria employed to include any particular outcome as evidence for M deficits. All four studies in category 2 show ‘sensitivity reductions at both low and medium frequencies and deficits which are most pronounced at the lowest frequencies’, suggestive of M deficits. Out of the remaining 18 studies, seven studies (category 7) found ‘no contrast sensitivity loss associated with dyslexia’ whereas eleven studies were ‘inconclusive’ (categories 4, 5 and 6).
However, zero studies showed a reduction in sensitivity of equal to or less than 1.50 cycles per degree (c/deg). According to Legge 18 the spatial frequency at which detection switches from M cells to P cells is 1.50 c/deg. Therefore all 22 studies in Skottun’s review fail to meet this criterion and are unable to truly count as evidence for M deficits.
The mixture of results is represented graphically in Figure 6.
[CaptionComponent="3047"]
Findings A and B in Figure 6 are consistent with Livingstone and colleagues’ physiological findings on CS in dyslexics;5 there is a clear depression of the CS curve under low spatial frequency (SF) as sampled by the M system. Interestingly, findings F, G and H are in direct opposition to the M theory. At high SF, it is expected that the P pathway takes over so dyslexics should yield a normal response. However, this is not the case in study D, F, G and H.
Finding C shows that disabled readers tend to have reduced contrast sensitivity than controls. The reduction in sensitivity tended to be ‘largest at 4 c/deg and less at both 2 c/deg and at the higher frequencies’; however, the pattern of loss of sensitivity is inconsistent with an M processing deficit. Finding D shows that losses in dyslexia are relatively independent of spatial frequency.
Skottun 17 noted that the various studies which he reviewed had conflicting criteria for including something as evidence for a M deficit. In particular, with study C in Figure 6, ‘all the sensitivity reductions [recorded] were in the range of spatial frequencies where the P system normally mediates detection’, ie above approx 1.0 to 1.5 c/deg.18 All tested frequencies were in this range so the potential of this study to uncover M deficits is improbable.
A pro M deficit supporter, Lovegrove, claimed that there are ‘consistent differences’ in contrast sensitivity, between dyslexics and controls and that there is a ‘consistent pattern of lower sensitivity of specifically reading disabled to low spatial frequencies (1 to 4 c/deg) than the controls’.19 These statements are not supported by Skottun’s analysis. Reading through the literature, it is becoming increasingly clear that the M processing deficit theory has polarised opinion, with many scientists and studies giving biased views as to its validity.
Conclusion
Having found a theory that seems to better explain some of the visual findings reported by dyslexics, and established underlying organic changes in the brain, it is clear that some psychophysical data still fail to confirm the theory while other offer support. In the final article in this series, we will take a closer look at evidence for another theory, the unstable binocular fixation theory, and then be able to offer a reasoned evidence-based overview of the existence and nature of dyslexia.
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 on this paper
References
1 NHS, 2009. Dyslexia. [Online] Available at: www.nhs.uk/Conditions/Dyslexia/Pages/Causes.aspx
2 Stein, J & Walsh, V, 1997. To see but not to read; the magnocellular theory of developmental dyslexia. Trends Neuroscience, Volume 20, pp. 147-152.
3 Breitmeyer, B, 1984. Visual Masking: An Integrative Approach. New York: Oxford University Press
4 Sandhu J. Eye movements and reading disorders in dyslexia – part 1. Optician 29.01.2016.
5 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
6 Lovegrove, W., 1993. Do dyslexics have a visual deficit?. In: S. Wright & G. R, eds. Facets of Dyslexia and its Remediation. Wollongong, Australia: Elesvier Science Publishers B.V., pp. 33-59
7 Breitmeyer, B., 1980. Unmasking visual masking: A look at the 'why' behind the veil of the 'how'. Psychology Review, Volume 82, pp. 52-69.
8 Breitmeyer, B., 1993. The Role of Sustained (P) and Transient (M) channels in Reading and Reading Disability. In: Facets of Dyslexia and its Remediation. Houston: Elesvier Science Publishers B.V., pp. 13-31
9 Lovegrove, W., Martin, F. & Slaghuis, W., 1986. A theoretical and experimental case for a visual deficit in specific reading disability. Cognitive Neuropsychology, Volume 3, pp. 225-267
10 Snowden, R., Thompson, P. & Troscianko, T., 2006. Basic Vision: An Introduction to visual perception. 1st ed. Oxford: Oxford University Press
11 Mason, A., Cornelissen, P., Fowler, S. & Stein, J., 1993. Contrast Sensitivity, Ocular Dominance, and specfiic reading disability. Clinical Vision Sciences, Volume 8, pp. 345-353
12 Breitmeyer, B. & Ganz, L., 1976. Implications of sustained and transient channels for theories of visual pattern masking, saccadic suppression, and information processing. Psychological Review, Volume 83, pp. 1-36
13 Stein, J., Riddell, P. & Fowler, S., 1989. Disordered right hemisphere function in developmental dyslexia. In: v. E. In C, I. Ludenberg & G. Lenerstrand, eds. Brain and Reading. New York: Srockton Press, pp. 139-157
14 Giddings, E. & Carmean, S., 1989. Reduced brightness contrast as a reading aid. Perceptual and motor skills, Volume 69, p. 383
15 Rayner, K. & McConkie, G., 1976. What guides a reader's eye movements?. Vision Research, Volume 16, pp. 829-837
16 Burr, D. C., Concetta Morrone, M. & John, R., 1994. Selectuve suppression of the magnocellular visual pathway during saccadic eye movements. Nature, 371(6946), pp. 511-513
17 Skottun, B., 2000. The magnocellular theory of dyslexia; the evidence from contrast sensitivity. Vision Research, Volume 40, pp. 111-127