Visual Dyslexia: New study is all sizzle, no steak
A new research report about dyslexia and vision has been drawing outsized media attention, with hyped up headlines that suggest a breakthrough discovery as to the cause and treatment of dyslexia.
But the reality is far more modest: a brief report by two French physicists who have not previously studied dyslexia, exploring a previously unexamined quirk of visual perception that seems to be associated with dyslexia – at least among the thirty adult dyslexic students that participated. 1
Mark Twain is credited with saying, “A lie can travel half way around the world while the truth is putting on its shoes.” And this certainly seems true of internet memes and clickbait.
I was skeptical of language apparently drawn from press releases – dyslexia “caused by an eye deformity?” (Daily Mail) A “cure” for dyslexia? (Today’s Parent) A new “‘relatively simple’ method of diagnosis …by simply looking into a subject’s eyes.? (Medicalxpress.com)
I don’t think anyone who works in the field of dyslexia or who has followed research over the years would fall for this. Genuine research is careful, painstaking, incremental … and most important, confirmed through replication.
I was even more baffled when I read the abstract of the journal article. The researchers said they used a “so-called foveascope”? “So-called” or otherwise, there appeared to be no such thing. The appropriate and standard tool for obtaining objective measurements of the retina would seem to be optical coherence tomography (OCT). How can results be quantified, compared, or replicated without the use of standardized and accepted equipment?
Fortunately, although initially it appeared to be behind a paywall, the full text article now seems to be available for online reading or download. And truthfully, I think it is quite a nice little bit of research, quite good as a first effort for two physicists who seem to be complete newcomers to the field of dyslexia, though apparently quite knowledgeable about laser technology. (see Curriculum Vitae Albert Le Floch).
But that all it is — the report of preliminary experiments, limited in scope, and documented in fairly well-written report containing a detailed and readable presentation of data. I liked reading the study because of its fresh and creative approach.
But there already is a large body of research going back decades reporting visual quirks and traits associated with dyslexia. So this is something like the discovery of one very tiny piece of a 10,000 piece jigsaw puzzle. We have a vague idea of which section of the puzzle this piece will go in (the eye-part, or maybe the cerebral dominance part). But because of the novelty of the research, no way as yet to know exactly where this new bit fits, or even whether it is documents anything other than an artificially induced phenomenon.
That “foveascope”? Turns out that was a device that the researchers invented. Why did they invent it? Because direct comparison of the outline of the right and left fovea centralis was “out of reach” because “the analysis has to be made post-mortem.” No, they didn’t “look into the eyes” of their subjects — instead they created an experimental protocol where they induced their living adult subjects to experience an after-image effect and to report and describe what they saw. Which would tell us nothing about what children might see, or think they see, in a similar setting. Logically it would seem to be no more or less reliable for diagnostic purposes than asking a person whether the letters seem to move on the page.
This is where the naivité of the researchers peeks through. Their audacity seems to be borne in part of ignorance of the range of symptoms and interventions that have long been explored and in use for dyslexia. For example, they wrote that the impact of adding a specialized LED lamp to their experimental protocol on the dyslexic subjects’ perception was “amazing” – but colored lenses and filters and specialized lamps have been in use as assistive technology for dysexics for decades. 2 Research is of course mixed on how well these interventions work – but it is hardly surprising that an effect would be observable in an experimental context. Especially one in which the researchers were inducing a faux visual effect rather than having their subjects actually look at or read printed text on a page or screen.
And of course those of us who work with Davis methods are well aware of the impact of orientation and disorientation on perception.
In sum, the French researchers did nothing more than attempt to measure and compare an apparent trait of dyslexia. Symptoms are not causes, and the researchers did not tease out whether the observed trait was tied to dyslexia, or a result of the developmental history of their dyslexic subjects as nonreaders, or simply something that was an artifact of their experimental protocol, gaps in their methodology, or selection bias.
I will be explaining more about this research and where it fits in with other research in a separate blog post. Certainly the researchers put in effort to set up their experiment and report on results. Given the media attention, outsized thought it may be, it means the report is worthy of more detailed analysis.
For now, suffice it to say that the research doesn’t say what the news accounts claim it does. This is a common problem with the way scientific research is reported in the media, but unfortunately increases the level of public misinformation and misunderstanding. The goal of scientific research should be to add to and enhance the collective body of knowledge. Eye-grabbing headlines and inflated claims only serve to undermine that goal.
For commentary from French dyslexia specialists who seem to mostly agree with my take, see Non, on n’a pas encore trouvé de cause de la dyslexie!
Update, 28 October, 2017: I found another blog post, this one by Mark Seidenberg, a leading research in cognitive neuroscience and an expert in reading acquisition. Key quote:
Dyslexics, their families and teachers, reading researchers and treatment specialists, and the organizations that represent them are asking: did someone just discover the cause and cure for dyslexia? (I know this: I get email.) As someone who has conducted research in the area, my question is different: how did this terrible article get published and how can its harmful impact be counteracted?
- Le Floch A, and G.Ropars. Left-right asymmetry of the Maxwell spot centroids in adults without and with dyslexia. Proceedings of the Royal Society B, Biological Sciences 2017 Oct 25:284(1865)
- Uccula A, Enna M, Mulatti C. Colors, colored overlays, and reading skills. Frontiers in Psychology. 2014;5:833. doi:10.3389/fpsyg.2014.00833.
20 thoughts on “Visual Dyslexia: New study is all sizzle, no steak”
Excellent analysis and commentary.
If you look carefully I think you will see that the steak is sizzling.
Yes the media will tend to hype up these things and I agree that it is unlikely that this will lead to a cure for dyslexia or a single intervention for helping ALL dyslexics since dyslexia affects multiple domains. The fact is however that this is a significant bit of research.
Using an after-image test the researchers demonstrate a lack of eye dominance in 90% of dyslexics (27 out of 30) compared to non-dyslexics who ALL exhibited one eye was more dominant. They then related this to the symmetry of the receptors between the eyes by having the subjects outline a circular area known as “Maxwell’s Spot” and found a perfect correlation between eye dominance (as measured by the afterimage test) and the symmetry of the spot patterns between the eyes – and that includes the 10% of dyslexics who were found to have a dominant eye which was explained on the basis of their ocular history.
This is a very clear and unambiguous finding even for this sample size (30 dyslexics and 30 non-dyslexics). Furthermore, it is not necessary to physically assess the receptors at the back of the eye since the concept they use (the entoptic phenomenon) and the relationship of Maxwell’s Spot to the retinal receptors is already established in the ophthalmic literature. The fact that they call their device a “foveascope” is neither here nor there as it simply describes a device for scoping the blue cone-free area of the fovea.
Regarding the comment that we have known about the “visual quirks and traits” associated with dyslexia for decades and that this discovery is just “one very tiny piece of a 10,000 piece jigsaw puzzle”, it needs to be kept in mind that not all pieces of the puzzle are equal. That is to say reading STARTS with vision and even phonological processing itself assumes normal fixation.
Most practitioners that work with dyslexia understand that there are many aspects to dyslexia but debate still exists around the role of vision. At present it is not clear what problems are primary and what are secondary but if it is shown that a problem exists at the level of the eye receptors themselves then this strongly suggests that vision plays a primary role at least for many (but perhaps not all) dyslexics and if so this would certainly impact on how we approach intervention.
The fact that the literature to date is mixed regarding vision in dyslexia comes down to the failure of MOST studies to account for how visual development changes with age. This particular study is interesting because not only does it show an anomaly at the retinal level in dyslexics that directly relates to eye dominance but the researchers go on to show the impact that this has on the way dyslexics spatially perceive letters between the eyes in terms of their relative position and direction – hardly a “visual quirk”. Furthermore, the study provides a possible mechanism to explain why the neurodevelopment for visual processing may be delayed in many dyslexics.
Finally, the afterimage test described in this research provides clinicians with a simple test that they can easily use when testing students with dyslexia (more reliable than the hole-in-the-card test) for demonstrating eye dominance – or lack thereof. Optometrists could use this information to consider new ways of helping students with dyslexia which may or may not involve using a pulsating light. At minimum it gives further support for therapies which target the development of visual pathways in the brain which are likely to be affected by the apparent lack of eye dominance.
Sorry, but I think there are too many methodological flaws in this study to draw any conclusions. The researchers worked with a non-random, non-representative group of dyslexics, so the study has no statistical validity (it can’t be generalized to other groups). There have already been large-scale studies of eye dominance in dyslexic children in the past. The perceived differences reported in one large scale study could not be replicated in a second, single-blind study (where the testers did not know which students were dyslexia and which weren’t.)
The particular after-image test for dominance used in this experiment appears to be novel and has never been validated in any other context, so it cannot be assumed to be an accurate test of dominance. I think what the researchers have found is a test that may measure changes to the foveal pit which take place in early childhood. It’s known that that the foveal perceptual span is influenced by reading practices– young children who are beginning readers have smaller spans than adults, and proficient readers show a bias in span direction from the fixation point that corresponds directly with the directionality of the written language — a market rightward bias for languages read left-to-right; leftward bias for languages read right-to-left.
The French researchers noted that early childhood up to about age 8 is a critical time for eye development, so I think the most likely explanation is that typical children who become readers at age 6-8 experience physical changes to the fovea of their non-dominant eye, whereas the individuals who do not begin reading until after age 8 are past the age of retinal malleability, so they don’t experience those changes.
That would explain why the after-image test corresponded to dominance measures with established tests for the control group, but there was no such correspondence with the dyslexic group. The researchers didn’t provide any information about age of reading acquisition for the dyslexic university students, but dyslexia is characterized by delayed acquisition of reading skills and other research on high-achieving dyslexics suggests that it is more typical for them to learn to read at age 11-12.
So this work would need to be replicated through a randomized, single-blind study, to assure collection of accurate data, simply to confirm the correspondence between the after-image test for dominance and the perception of the Maxwell’s spots. If that association is confirmed, then it would have to be followed up with tests on children to ascertain when these differences emerge, perhaps beginning with a comparison of typically developing children between ages 4 and 9.
But I don’t see it very likely that anyone will undertake the task of trying to replicate this, because I don’t see where it leads. Optometrists already know that the most common problem encountered by struggling young readers is convergence insufficiency, and developmental optometrists know how to test for that and other conditions, as well as how to treat them. (Obviously convergence insuffiency or other common childhood conditions sucy as amblyopia would supply another explanation as to different patterns of of retinal development as well).
I am Dyslexic, and I have to say that living with this disability for over 50 years and teaching students with Dyslexia for over 25 years, I think I know a thing or two about the subject.
Part of the issue is visual! Some of us with Dyslexia are capturing symbols and images incorrectly. We view these items and move them around. Yes, I am aware that our brain has a lot to do with the movement of what the eye has captured! You don’t have blind people with Dyslexia, at least none that I have come across, so this may help to prove a point.
I am in agreement that glasses, or a laser fix is not the answer, because the problem lies tied to both how we capture symbols at a glance and then how they are processed by the brain.
I am upset that we debunk the study as mythical or wrong, they are on to something, and more research needs to be done with the eye brain connection. While simply correcting the vision is not the answer, it does help in some areas.
We know that there are different types of dyslexia and different levels of severity. It’s just not right to claim it is not visual! It is, to a varied degree, if it weren’t, color overlays would not work, enlarging, bolding, framing and different font styles would not be helpful.
I’ve personally seen that all the above when used, can be helpful. I also include the use of cursive writing, because it helps differentiate how the letters are formed, pictured and written. Phonics is helpful, but not when we flip, vowels and letters around. It is also not helpful with dyscalcul1a.
People are mistaking this study to say that Dyslexia is not a visual problem and I am here to say it is a large part of the problem for many dyslexics and phonics is and not always the solution! It helps some but frustrates many other dyslexics.
Brilliant…. very well said. I say it’s common sense that needs to be used. I myself have studied as my Mother and son possibly are dyslexic…..the eye most definitely plays a part.
The study has statistical validity for the student population that was studied and whilst it cannot be said for certain that the findings apply to the wider dyslexic population the unambiguous nature of the findings (ie. a perfect correlation between the after-image test of eye dominance and the asymmetry of receptors between the eyes) makes it very unlikely that similar findings would NOT be found in the wider dyslexic population and in fact the reasonable position is to assume that it probably does. This is even suggested from the study as the researchers also tested 5 members of a family (ages not given) that were medically diagnosed with dyslexia and ALL members showed no asymmetry (ie. typical of dyslexia).
Yes, there have been numerous studies which have looked at lateral preference and eye dominance in dyslexics which come up inconclusive – but this is in agreement with the findings of this research. The issue is not that dyslexics have a particular preference but that they do NOT. It is the uncertainty of dominance that is thought to give rise to the spatial issues in dyslexia.
The after-image test as a test of eye dominance was validated in the study using the traditional hole-in-the card test in the non-dyslexic population. There is no scientific basis to directly link the after-image test with the perceptual span.
At this stage it is not known whether the symmetry of the receptors at the fovea in Maxwell’s Spot has a genetic basis for dyslexics or whether this is the result of poor reading habits. As mentioned however the researchers did some testing on other family members and the early suggestion is that it is genetic.
A randomized placebo controlled study would be good but the correspondence between eye dominance using the after-image test and the asymmetry of receptors in Maxwell’s Spot is already very convincing……so far 100%. Certainly it would be good to look at this across a wider age group.
Convergence insufficiency affects about 10% of the population, probably higher in dyslexia. The kinds of visual problems that are MOST common in dyslexia however relate to eye tracking and visual perception. These are skills that develop with age and if the research showing that dyslexics lack eye dominance is correct then this provides a possible explanation as to why we might see neurodevelopmental delays in the visual pathways. From my perspective this research contributes to the visual aspects of dyslexia and I see it being taken much further.
The study does not have statistical validity because the test subjects were volunteers, not randomly selected. Hence no P-value reported in the study — you need to start with a random selection process.
The standard measures for eye dominance showed that roughly slightly over half of the dyslexic group were either left or right dominant. So the afterimage test that the researchers used could not be validated.
I’m not saying that the researchers conclusion is necessarily false — just that it isn’t as of yet supported by evidence. It is essentially an anecdotal report using small group comparisons and previously untested forms of assessment or measurement, in an unblinded context. So it makes sense as the sort of preliminary research that might be done as a precursor to a larger study, but it really can’t support any conclusions other than that ther seems to be a correlation between the afterimage dominance test and the Maxwell’s spot test. But that could simply be because both are testing the identical thing. And there’s still no showing of any sort of causal relation with dyslexia, and certainly no evidence of any treatment.
It’s not necessary to randomly select the groups. If one was to successfully replicate this study for a much larger group and across a wider age range are you still going to argue that the findings are invalid? If however you wanted to look at the efficacy of treatment (eg. using a pulsating light to help with reading) then it certainly makes sense to use a double blind randomized process but this is not the main point of the study.
The fact that half of the dyslexics were either right or left eye dominant on the sight test of eye dominance does not invalidate the afterimage test. What this shows is that dyslexics are unsure about which eye is dominant on the sight test and that the after-image test is a better test of dominance when uncertainty exists. We can know this because the sight test of dominance correlated 93% with the after-image test in non-dyslexics (the remaining 7% didn’t contradict the after-image test the students were just unsure which eye was dominant) and the after-image test correlated 100% with the left right asymmetry of the Maxwell spot centroids for both dyslexic and non-dyslexic students (60 out of 60 students).
As a preliminary study the researchers make a very strong case for a lack of eye dominance in dyslexia. They then rightly theorize that this lack of eye dominance could lead to perceptual anomalies and proceed to demonstrate such anomalies in dyslexic students with the mirror reversal of letters in the form of afterimages but fail to replicate this with the non-dyslexic students. Based on the time delay of the reversed mirror image they are then able to cancel this out by using a pulsating light. Granted this requires more investigation but I think the REAL message here is that the lack of eye dominance in dyslexics can, and probably does, lead to problems with visual processing. That is to say, once you demonstrate an anomaly at the retinal level it sets the stage for how this might impact on reading – which is big news for those interested in dyslexia.
I suspect you are mainly coming from the point of view of treatment but in the process you seem to be discrediting what appears to be an excellent piece of research which if confirmed to be accurate has significant and profound implications regarding the role of vision in dyslexia.
Stuart, here’s a good article that explains why random selection is essential for a study to have external validity (i.e., to be generalizable to a larger group): http://www.statisticssolutions.com/difference-between-random-selection-and-random-assignment/
This isn’t my opinion; it’s a basic precept of research design. It doesn’t mean that the research can’t be considered, but without randomization, it’s impossible to calculate statistical probability (P value).
The reports about the reversed mirror impression with the afterimage test were anecdotal- mentioned in the research only in passing, and apparently only experienced by three subjects (or at least only 3 examples were given). But I would take the delay between the initial image and after image to be very clear example of disorientation, and the protocol for the afterimage test with the switching back and forth holding hands over each eye seem almost designed to cause disoriention. So I could say that this experiment confirms Davis theory – see https://www.dyslexia.com/davis-difference/davis-theory/the-cause-of-dyslexia/ (“4. Disorientation produces false sensory perceptions.
The different views and perceptions the individual is examining mentally are being registered in the brain as actual perceptions.”)
But disorientation has already been studied in different contexts — so to me the main point is simply that they failed to recognize or control for the possibility of disorientation.
I understand your argument around randomization and I agree in principle, especially if we are talking about proving a treatment but that’s not the main purpose of the study.
When it comes to the issue of “random sampling”….
If your concern was generalizing the finding to other age groups (ie. not just uni students) then I agree this requires taking a random sample across a wide age range (eg. a random sample of 100 dyslexics ages 7 to 20 yrs) OR you could achieve the same end by taking a larger sample size across each of the age groups (eg. 100 dyslexics ages 10-12, 100 dyslexics ages 13-15, 100 dyslexics ages 16-18 & 100 dyslexics ages 19-20). Personally I prefer the latter because I like hard data. As the findings in the current study are so strong it is reasonable to presume that you would also see a similar finding in the younger age groups – but that still needs to be tested.
If your concern is that the student population tested were volunteers and therefore self-selecting then one needs to consider whether this is likely to make a difference. If the study was measuring personality traits in dyslexia then this would be an issue. In this case the study is looking at eye dominance and correlating it with a physical parameter in the eye. It is unlikely that any of the volunteers had any prior knowledge of the symmetry of their Maxwell Spots or that this would realistically affect the outcome. Certainly it was acceptable by the journal review process.
My main point is that when taken in context as a preliminary study then it is actually a great piece of science. The researchers may not be dyslexia experts but they know their physiological optics and sometimes it takes people from outside the field to get things back into perspective. If it can be replicated in the wider dyslexic population then the implications are huge and therefore in my opinion this study does not justify your level of criticism (except to acknowledge caution that these findings are preliminary and still need to be replicated like all good science).
It’s hard for me to comment on the Davis theory but as far as I can see I don’t think the study necessarily contradicts this – so maybe we can agree on that!
In fact you don’t need to make alternate hand movements to create the disorientation you describe. The researchers essentially argue that this exists by virtue of the uncertainty which comes from the lack of eye dominance. They reason that this may affect the neurodevelopment of the visual pathways and brain lateralization and then proceed to give an example. The resulting perceptual anomalies (mirror imaging, shadowing and spatial shifting of letters) would then lead to confusion and further disorientation which seems to be consistent with Davis theory.
I would add that this probably also affects the ability to visualize letters and words which is also highly spatially dependent. The pulse light may address some of the immediate sensory perceptual issues but speaking as a clinician I would say that it is not likely to address the years of faulty visual attentional patterns that may have arisen as a consequence of the “disorientation”.
I don’t really want to belabor the point, but we don’t know anything about how the volunteers were recruited. If a reasearcher posts a notice at the university saying they are lookng for volunteers for research into dyslexia and vision, that is going to potentially draw a different set of students than another notice might. The University of Rennes 1 is a university that specializes in science and technology; dyslexics tend to be overrepresented in fields such as engineering, but dyslexic engineering students might be quite different than dyslexic art students. (No art students at Rennes 1- they would all be at Rennes 2). All of the dyslexics in this group reportedly experienced left-right confusion, which is a common but not universal symptom of dyslexia, and which is not ordinarily part of the diagnostic criteria for dyslexia — so was that just a happenstance? Or were the researchers using that as part of their screen, and excluding the students who didn’t have that problem? All of these dyslexic students had ongoing reading problems and qualified for 30% extra time on exams. What about students with a childhood history of dyslexia who have overcome their difficulties and read well enough that no accommodations are needed?
There is nothing wrong with doing a study of 30 French dyslexic students at a single university- but whatever the findings are, they only can apply to those 30 students. It’s a preliminary, first stage study. In my view, it doesn’t merit a press release anouncing that a new cause of dyslexia has been found, much less hinting at a new way to diagnose or some sort of cure. There are thousands of studies into dyslexia– this perhaps offers a new tidbit of information, but that’s all it is. And I’m very skeptical of the assessments they used, because both the “foveascope” and the afterimage dominance test are new inventions – so no external standard to measure against.
The vast majority of new research findings cannot be replicated — in fact, something like 2/3 of all reported research later turns out to be wrong.
I think research is a good thing. Small, large, every kind of research, including novel research.
But I don’t like the hype. I think what happened in this case was that the researchers overstated their work in a press release, and because it happened to be during Dyslexia Awareness month the news media picked it up and ran with it. So it makes a good story, but it is misleading and also disrespectful of the many dedicated researchers who have spent years doing high quality research into dyslexia.
Here’s a link to Google scholar for research on dyslexia that has been published in 2017 — https://scholar.google.com/scholar?as_ylo=2017&q=dyslexia&hl=en&as_sdt=1,5&as_vis=1 — the search produces 5880 results.
If I narrow the search to include both the words “dyslexia” and vision” – again only for reports published in 2017 – I get 2450 results:
So why all the media hype about what at best is a preliminary study?
The affiliations claimed by the authors do not exist.
* 20 square Marcel Bouget at Rennes is a home, not a laboratory.
* There is no « Laboratoire de Physique des Lasers » at Université Rennes 1 – look at https://www.univ-rennes1.fr/.
* Université Bretagne Loire is a federation of universities for french regions “Pays de la Loire” and “Bretagne”. It does not hire any academics.
A. Le Floch is not member of Université Rennes 1, see university directory.
It is possible that the results claimed by the authors will not be discussed at any international conference, nor laboratory seminars, as with previous results for that 2-people team. So that results are not discussed within the scientific community.
The “foveascope” instrument performances should be asserted: accuracy, repeatability, etc, as any scientific instrument.
Why in the world are you so against this study as a possible explanation for the cause of dyslexia in certain individuals? Isn’t the point to find out possible causes so that we can find solutions? Further study is certainly needed, but it’s no different than you using the Davis Method for dyslexia. Has that been shown to be effective in a randomized, double-blind study? If not, then according to your logic, then the Davis Method shouldn’t be used.
Joe, I am in favor of good science, not bad.
I am in favor of honest representations of research findings and results, not lies and exaggerations.
I am in favor of standardized research protocols and attempts to replicate. The first published study of Davis methods was commenced in the mid 1990’s and published in 2001. It was a longitudinal study of 86 primary school children in 6 classrooms among two separate schools, with demographically matched comparison and treatment groups.
“Double blind” is impossible in the context of a study of an educational intervention, because of course the students will know what they are being taught, and teachers will have to know what they are teaching. But blinded assessment of pre- and post treatment effects is an easily achievable goal and an indicia of reliability – in the Davis Learning Strategies study, they followed this protocol:
The results were then subject to statistical analysis, which is standard for any group research:
I know that all sounds like gobbledygook for anyone who does not have the benefit of a college level statistics course, but it is standard that there will be statistical analysis presented in any sort of group research. When I read a study that is devoid of any such analysis, it raises questions.
The Davis methods have been used for more than 35 years with tens of thousands of program completions, and there are multiple published research studies by independent researchers, working in different parts of the world, in different languages and different age groups. Formal research is listed here – https://www.dyslexia.com/davis-difference/about-davis/inside-the-davis-program/research-studies/
But we haven’t any press releases overstating or misrepresenting any study. The 2001 study I referenced above concluded with the statement, “Additional research is needed in other settings with larger groups of children. Testing should be conducted on a broader range of skills including qualitative reading samples and comprehension.” So no claim was made from the one study other than what the study showed — among the children who were tested, the children in the research treatment group performed signficiantly better for word recognition than the comparison group.
We aren’t trying to supplant the work of others but instead we respect the work of legitimate researchers and the value of building on existing scientific research.
Again – good science is careful, and incremental. One study is done to provide the foundation for subsequent studies, so that data can be gathered and evaluated in multiple contexts. Good researchers take extra measures to avoid bias in the collection of their data.
So yes, I’ll continue to promote good science when I see it and I won’t hesitate to point out when research falls short of accepted scientific standards.
I can think of a few good reasons why the media would pick up on this story….
* It is rare for a study on dyslexia to get such unambiguous findings
* It is rare for a study on dyslexia to show clear anatomical differences
* The role of vision in dyslexia is a hotly debated topic
* The implications of an anomaly in the eye are huge
* The authors offer a possible solution, albeit somewhat tentative
How many of the 5880 studies published in 2017 on Google Scholar compare with that?
Even if you argue that the study was affected by lack of random sampling (as per our previous discussion) it is STILL a remarkable finding. Obviously further investigation is required but this point was NOT overlooked by the media reports.
Regarding left right confusion (and specifically letter reversals) not being present in all dyslexics, I can only give my opinion but I can support it with evidence. The underlying spatial anomaly that potentially comes from lack of eye dominance is a LOW LEVEL sensory deficit and I believe that many students find ways to compensate for this using higher cognitive strategies. As a consequence the problem may not show up on certain cognitive tests or they may no longer exhibit symptoms such as letter reversals however they may still exhibit other anomalies such as tying their shoe laces or visualizing a word in their minds eye. At the very least if you spatially load the text they will struggle more than non-dyslexics.
Yes, and that is a huge red flag for fraud in research. It’s practically unheard of – in any research on human subjects you expect to see a distribution of response– research reports are full of scattergrams and bell curves, not lists showing the results for every subject. When results are too “perfect” it usually indicates cherry-picking of data or research bias.
I find it especially odd given that the shape of the foveal and associated perception has been described in literature as “highly variable”. (see, for example, http://www.sciencedirect.com/science/article/pii/S0042698914002442 – “since foveal shape, foveal blue scotomas, and macular pigment distribution are all highly variable among subjects, it is interesting to study how they are related.”) How can you take a quality known to be “highly variable” and then not see any significant variation in the defined groups?
I mean,I find it very hard to reconcile the reported consistent size and asymmetries of the non-dyslexic group with the drawings collected from this group of subjects: https://ars.els-cdn.com/content/image/1-s2.0-S0042698914002442-gr6.jpg (some of whom could not even see the Maxwell’s spots, and none of whom reported left/right patterns even remotely similar to what Le Floch and Ropars claim for their “controls”)
And when the results are read it isn’t “unambiguous” at all. Sixteen of the dyslexic subjects showed either left or right eye dominance on standard measures — so the reported asymmetry of Maxwell’s spots for dyslexics can’t be related to eye dominance — so contrary to being “unambiguous” the reported data contradicts the underlying assumption of a relationship between symettry/asymettry and eye or cerebral dominance.
The study didn’t include any tests or measures for anatomical differences. The proper tool to show anatomical differences would be Optical coherence tomography (OCT). The study I linked to above is a clear example – they used OCT along with their perceptual measures so they could correlated the perceptual differences with measurable anatomical differences.
But at best the testing by Le Floch and Ropars can show functional differences, and even that is uncertain, given the reliance on subjective reporting of subjects and use of nonstandard testing protocols.
It’s possible but at this stage you have to take the results at face value as we don’t have any good reason to believe otherwise. Not everything in nature fits a bell curve….
The foveal blue scotoma relates to how steep the foveal pit is and does not affect Maxwell’s spot. Macular pigment does have some correlation with Maxwell’s spot but relates to the pigment found in the layer below the receptors. Maxwell’s spot is believed to be the anatomical area that is devoid of blue cone receptors. Variations in the foveal anatomy, foveal blue scotoma or macular pigment cannot be used to the infer the same applies to Maxwell’s spot.
The dyslexics that showed eye dominance on the sight tests did so because of uncertainty of dominance (I use a sight testing method to measure eye dominance for patients myself and the results are not always clear). This uncertainty was greatly reduced for the dyslexic group by using the after-image test which showed that in fact they lacked eye dominance – the validity of which I discussed previously.
I also use an OCT device and it is good for measuring the anatomy of the foveal pit. It can also be used for measuring the auto fluorescence of pigment in the macula found in the retinal pigment epithelium layer but it cannot distinguish between the different types of cone receptors which is required for demarcating Maxwell’s spot.
The anatomy of the foveal receptors has been studied separately by other scientists, which is why we know there are no blue cones present (which is thought to reduce chromatic aberration and thereby enhance acuity). This is believed to give rise to the “entoptic images” when looking at a bright light source using a dichotic (eg. blue/green) filter. This is not a new or novel concept as Maxwell’s spot has been known about for decades and the explanation for it is well accepted in the ophthalmic literature so this is not something the researchers need to try and prove. Below is a study dating back to 1972 which utilizes a similar set up which uses an overhead projector as a light source and which also acts like the tablet in the current experiment for tracing the image. It shows the method to be accurate and repeatable and could also be used by children under 10 years of age….the youngest being 4 years old.
The main “novelty” of the research is correlating the asymmetry of the spots to eye dominance…which relates structure (receptor symmetry between the eyes) to function (eye dominance). The correlation of these findings COULD be key to providing a neurobiological basis for dyslexia…….or at the very least help explain the visual aspects of it. Certainly it raises some significant questions, especially if replicated in younger students.
But I don’t have to take data at face value; I can think and analyze critically, look at the methodology and consider what measures were taken to assure objectivity and integrity of data. Bias in data collection, manipulation of data, and misrepresentation of data are unfortunately common issues- so that means it’s important to look for indications of reliability– and be alert to issues that raise red flags.
I did a brief search and Guy Ropars is listed as a faculty member of the Université De Rennes 1 in Science and Properties of Matter.
From what I could find Albert Le Floch is an ex professor of the university (so not listed) and probably retired from a formal position but is well published and has his own blog.
Regarding instrument performance, see my link in the previous reply. The methods are similar.
Speaking of retirement I am going to withdraw from making further replies as I feel that I have made my position as clear as possible. If you feel the experiment is flawed however then you could take this up with Professor Le Floch himself through his blog.
To summarize, this is an ingenuous experiment that uses methods that are known in physics and visual science and published in what appears to be a reputable journal. The study itself is easy to replicate and IF confirmed in a larger population it has enormous implications and therefore deserves the publicity it has received.
Anyone know if this study has been reproduced with a bigger and random sample?
Comments are closed.