Under review: Physicists Albert Le Floch and Guy Ropars developed a now-disputed theory about dyslexia.
Maxppp / Samuel Nohra

Controversial dyslexia study marred by methodological and ethical problems, researchers say

Subsequent studies have also failed to validate a range of products based on the work.

For years, scientists have been raising concerns about a 2017 paper that put forward a novel theory about dyslexia and spawned a range of products for people with the condition.

A new commentary published today in the Proceedings of the Royal Society B says there is no evidence to support the theory that dyslexia arises from symmetry in the distribution of photoreceptors in the left and right eyes.

That theory, first proposed by two physicists in the same journal, was the basis for patent filings on flickering eyeglasses, screens and lamps marketed for dyslexic children. The battery-powered glasses, dubbed the Lexilens, received an Innovation Award at the Consumer Electronics Show in 2020. That same year, France’s National Academy of Medicine awarded the researchers a prize for their work.

The new commentary, however, calls the 2017 study “highly problematic” and “likely not reproducible.”

The authors point out that the study failed to conduct an objective assessment of the dyslexic students it recruited, such as evaluating their baseline reading ability. They assert that the published methodology is confusing, contradictory and potentially biased because the participants were not properly randomized.

It is also unclear whether the study was approved by an institutional review board as required by the journal, the commentary says. The original article failed to disclose any details of the informed-consent procedure or the name of the institutional review board, noting only that it had been “conducted according to the principles expressed in the Declaration of Helsinki.”

“This was a failure of the peer-review process,” says Florian Naudet, a psychiatrist and professor of medicine at the University of Rennes who co-wrote the commentary. “It should not have been published.” Naudet also notes that several independent trials have so far failed to find a benefit from the related products.

The originators of the theory, Albert Le Floch and Guy Ropars, also at the University of Rennes, did not respond to specific questions about their work but say they stand by their findings.

“Our physicists’ approach to dyslexia is robust,” they wrote in a joint email to The Transmitter, noting that the clinical studies conducted to date have been flawed. “Naudet’s comments do not discredit any of the scientific results from the six figures in our 2017 paper.”

T

wo commercial sponsors of the products that sprang from the 2017 paper have blocked researchers from publishing or sharing results from their own double-blind clinical trials, the investigators who oversaw those trials under a contract with the university have revealed to The Transmitter.

The researchers, who were not paid by the companies, say they signed contracts suggesting that the results would be published and shared with participants. But when the trials were completed in 2021, the companies refused to waive a confidentiality clause in the contract.

“I was very upset by this,” says neurologist Catherine Allaire, who led one of the trials at Rennes University Hospital before her retirement. She found no evidence that a lamp being developed by the university’s technology transfer company, Ouest Valorisation,  helps dyslexic children. “They never allowed us to publish these results.”

Lili Light for Life, which licensed the technology from Ouest Valorisation, provided The Transmitter with a customer satisfaction survey and noted that it is running a new trial in the United Kingdom.

Frederic Mouriaux, an ophthalmologist at the same hospital who ran a study on the Lexilens glasses, says that he was not even permitted to see the results, including whether the glasses could be leading to adverse outcomes.

In response to a formal request from Rennes University Hospital to publish the data and share it with the trial participants and their families, the company sponsor of the glasses, Abeye, wrote in a letter shown to The Transmitter that it would not share the data because Mouriaux’s study was flawed.

“The first study did produce some positive results, but not positive enough for my tastes,” says Michael Kodochian, Abeye’s CEO. In his view, a proper clinical evaluation of the Lexilens would include monitoring children’s reading ability over time as they develop their reading skills using the glasses. “The eyewear does not teach how to read or spell,” he says. “You cannot erase five years of struggling.” He says the company is planning a new study of the Lexilens but declined to provide any details about it.

“I can hear the sound of goal posts being desperately relocated,” says Dorothy Bishop, emeritus professor of developmental psychology at the University of Oxford and a co-author on the commentary. “If they need more time points to evaluate the Lexilens, then they should provide data to confirm that is the case.”

L

e Floch and Ropars came to the field without any special expertise in dyslexia, but they did have knowledge about lasers and an interest in an optical phenomenon known as Maxwell’s spot: If an observer stares at a purple screen, a reddish spot with a halo around it will appear in the center of their field of vision.

Maxwell’s spot arises because the macula, the region on the retina with the highest concentration of photoreceptors, has a pigment that absorbs blue light. In fact, at the center of the fovea, the depression within the boundaries of the macula where vision is the sharpest, there are no blue cone photoreceptors, only red and green ones.

In their 2017 study, Le Floch and Ropars used their own invention, dubbed the foveascope, to create tracings of this spot in the left and right eyes of 30 children with dyslexia and 30 without. In all of the children without dyslexia, the spot was round in their dominant eye and somewhat elliptical in their non-dominant eye. By contrast, 27 of the dyslexic participants had nearly symmetrical round spots.

A photograph of black sunglasses
Blind spot: The company that sponsored Lexilens glasses blocked researchers from seeing the results of a study that tested the glasses at Rennes University Hospital.
Courtesy of Lexilens

Le Floch and Ropars suggested that that lack of ocular dominance could explain why dyslexic people might confuse the letters b and d: They are simultaneously seeing both the letter and its mirror image. They further reported that they were able to suppress the mirror image in several study participants using a special lamp that flickered at 70 hertz.

“Our observations lead us to believe that we indeed found a potential cause of dyslexia,” Ropars told The Guardian at the time.

Although some people with dyslexia have visual deficits, for the majority the condition arises from problems within the language-processing regions of the brain, affecting their ability to read and write, most psychologists and neuroscientists say.

A

fter the initial paper was published in October 2017, several well-known language experts took to PubPeer, a post-publication review site, and to their personal blogs, to criticize it.

Mark Seidenberg, a cognitive psychologist at the University of Wisconsin-Madison, pointed out that mirror-image mix-ups, despite their frequent portrayal in popular accounts, are not specific to dyslexia.

Elisabeth Bik, a Dutch microbiologist and image sleuther, noted that the researchers had not declared any competing interests, even though they had apparently been granted a related patent in France in 2015.

That patent had evidently been annulled at the time of publication, according to Naudet, but the university submitted a separate patent application related to the glasses in November 2017.

Naudet, who does not study dyslexia but says he is passionate about scientific integrity, was unaware of the controversy surrounding the paper until Mouriaux contacted him to express his disappointment at being unable to publish his research on the glasses.

Naudet began to dig into the original study and says he was troubled by what he found. He soon reached out to several dyslexia experts, including Seidenberg and Bishop, collaborating with them to submit a letter to the journal’s editors in May 2023.

After the journal editors sent inquiries to the initial authors and the scientific integrity officer at the University of Rennes, they concluded that the initial paper should not be retracted, but they invited Naudet to submit a commentary.

Naudet says he understands their decision. “I think the journal handled the case correctly.”

Not only have subsequent studies failed to verify the value of the products that are based on the 2017 paper, the scientific council of the French Ministry of Education has also issued a statement that Le Floch and Ropars’ theory “cannot be considered validated” and that no studies have demonstrated the effectiveness of the special lamps or glasses.

Naudet is satisfied with the outcome, he says. “There’s nothing that proves that the devices work.”