To test the efficacy of Irlen colored overlays for alleviating reading difficulties ostensibly caused by Irlen syndrome, a proposed perceptual disorder with controversial diagnostic status.
Sixty-one schoolchildren (aged 7–12 years) with reading difficulties were assessed by an Irlen diagnostician. We used a within-subject study design to examine differences in reading rate across 3 conditions: using an overlay of a prescribed color; using an overlay of a nonprescribed color; and using no overlay. In a subset of 44 children, all of whom had a diagnosis of Irlen syndrome, we also used a between-group design to test the effects of Irlen colored overlays on a global reading measure.
The Irlen diagnostician diagnosed Irlen syndrome in 77% of our poor readers. We found no evidence for any immediate benefit of Irlen colored overlays as measured by the reading-rate test or the global reading measure.
Our data suggest that Irlen colored overlays do not have any demonstrable immediate effect on reading in children with reading difficulties.
Comments
Re: Rebuttal: Factors that would account for lack of immediate changes
Ms. Irlen takes issue with the design of our study on colored overlays for reading. We are perplexed and saddened by this. Even a casual reader would notice that our study relied on extensive co-operation and collaboration of an Irlen Institute diagnostician, who modified her normal practice to facilitate our double-blind design. More than this, we involved the Irlen Institute in our design from the outset, as this was essential to mounting a fair test of Irlen practice. We sent our draft design to Ms. Irlen, and received a reply on 9 October 2009 from Sandra Irlen, who described our study as ' very exciting' and 'well-designed', and made several recommendations. One was that we should include a measure of reading comprehension as well as speed, which we accordingly did (we are happy to provide copies of this correspondence on request). It is, to say the least, inconsistent for Ms. Irlen to now describe our study as 'significantly flawed in terms of methodology', and it is hard not to view this as an opportunistic response to the negative outcome of the study.
Ms. Irlen identifies two factors that may have accounted for our negative results: the fact that we focused on immediate effects, and the presence of uncorrected vision problems in our sample. On the first issue, she states that 'we would never have expected to see immediate effects' of the overlays. This is contrary to the Irlen Institute's core diagnostic procedures, which are based upon an evaluation of a person's immediate response to colored overlays. It is also contrary to numerous published statements of Ms. Irlen's. For example, in her book, 'The Irlen Revolution' [1], she describes her original discovery of the treatment: 'For the first time, all the struggling readers were able to see words on the paper that were not jumping, swirling, dancing, or all jumbled together' (p. 20). Of one adult reader, Ms. Irlen states 'she had struggled to read for so long, and was suddenly able to without such a great effort' (p. 20).
Ms. Irlen suggests that the lack of developed reading skills in our sample would preclude immediate effects of the overlays. It was precisely to avoid this problem that we used a test of reading speed consisting only of very simple words. All of the children in our sample knew all of these words, and this was verified before testing. It is illogical to suggest that, if colored filters improve text clarity, they should not lead to faster reading of such text that is within the reader's competence. In any case, we are also interested in long-term outcomes, and have completed a one-year follow-up study of the original cohort, which is now under review at another journal.
Ms. Irlen then suggests that the visual problems identified in 54% of our sample 'would have had a significant affect [sic] on the test results'. This point does not reflect well on Irlen Institute practice, as the children received an orthoptic assessment only due to their involvement in our study (the assessment was paid for by us). Had our study not been taking place, the children would have had the Irlen assessment, but not the orthoptic assessment. Moreover, there was no overlay benefit even in the sub-sample without orthoptic problems. Ms Irlen has been able to verify this personally, since we provided our raw data to the Irlen Institute on study completion.
Ms. Irlen ends her rebuttal by stating that we made 'strong recommendations... against the use of colored overlays altogether'. This is false; in the conclusion of our article, we merely urge interested parties to 'carefully consider the totality of the evidence' (p. e937). She then provides a list of studies showing positive effects of colored filters (one of which she cites twice). Given that she does not predict immediate overlay effects, it is curious that several of these studies looked only at immediate effects. Moreover, several concerned methods of colored filter provision other than the Irlen Method. In any case, citing only those studies that support a treatment, when several negative studies exist, gives a biased impression. In our article, we cited four recent reviews, all of which identified significant methodological problems in many studies in this area. A constructive approach for people interested in this treatment would be to address these problems by carrying out research designed to avoid them. This was our objective in carrying out our study, and we consulted with the Irlen Institute to ensure that the assessment was fair to them. We stand by our design, and the results that it produced, even if Ms. Irlen does not.
Reference
[1] Irlen H. The Irlen Revolution: A Guide to Changing Your Perception and Your Life. New York, NY: Square One Publishers; 2010
Conflict of Interest:
None declared
Rebuttal: Factors that would account for lack of immediate changes
The recent article published in Pediatrics (2011; 128:4 e932-e938), Irlen Colored Overlays Do not Alleviate Reading Difficulties, was significantly flawed in terms of methodology, interpretation, and presentation of results. In this response, we would like to bring to light some of the factors that might have accounted for the lack of findings reported in this study (all of which were failed to be discussed in the study itself).
To begin, the title of the study does not accurately reflect the results and conclusion presented by the authors. In fact, the authors themselves clearly state they were only studying the immediate effect of color; yet the title of the article denigrates the entire process they were testing, not just its ability to have immediate effects. There are several factors, from methodological flaws to basic cognitive development, any of which could have accounted for the lack of positive results when it came to color's ability to result in immediate changes in reading ability in this study. However, none of the factors mentioned below were discussed by the study's authors as potential limitations of the study.
The most basic oversight is that the research chose to look at "immediate" changes in reading ability when using overlays with young students who had reading difficulties and who, therefore, had not yet developed reading skills. Given that the children in the study were, on average, 9 years of age and identified as poor readers, it is highly unlikely that they would have already mastered or even acquired many of the basic skills required for reading. Irlen Colored Overlays/Filters remove the barrier preventing these skills from being acquired but are not a replacement for instruction and practice which develop the necessary foundation required to read. There are 25 years of research showing positive effects involving Irlen Colored Overlays/Filters. For those with reading skills, the overlays provide print clarity, stability, and comfort so that they can efficiently and effectively utilize their reading skills. For those without adequately developed reading skills, Irlen Colored Overlays/Filters remove the barrier preventing them from acquiring these skills at a normal rate.
Past studies looking at the use of colored overlays to improve reading in young children generally involve an intervention period, thus allowing children to use their overlays while acquiring their reading skills. These studies compare children who were allowed to use their overlays during the intervention period with a similar group of children who were not provided with colored overlays/filters to use during the same period of instruction. Specifically, in a study by Jeanne Noble, Michelle Orton, Sandra Irlen PhD, and Greg Robinson PhD (2004), those who needed colored overlays made significant progress in all aspects of reading; whereas, those who needed overlays and were not allowed to use them made no progress. Importantly, after the intervention period, the control group was also given overlays; and they made the same significant progress as their counterparts in all aspects of reading. We would never have expected to see immediate effects with this population of young children who were identified as poor readers.
A second issue with the recently published Pediatrics article is that visual problems were identified but not corrected prior to testing. Fifty -four percent (54%) of the small number of subjects in this study were identified as having uncorrected visual problems which were not corrected prior to participating in the study nor were these students eliminated from the research. This would have had a significant affect on the test results as uncorrected visual problems will continue to stand in the way of reading improvement, even if issues related to Irlen Syndrome are eliminated through the use of color.
In conclusion, this was a poorly designed and limited study of the Irlen Method, especially when dealing with such a complex issue as visual processing problems related to reading problems. We do find it unfortunate that the authors of the study have made such strong recommendations cautioning parents, educators, and professionals against the use of colored overlays altogether when there are numerous controlled studies confirming the improvements in reading with Irlen Overlays and Spectral Filters (Kyd, Sutherland & McGettrick, 1992; O'Connor, Sofo, Kendall & Olsen, 1990; Robinson & Miles, 1987; Williams, Le Cluyse & Rock-Faucheux, 1992; Jeanes, Busby, Martin, Lewis, Stevenson, Pointon et al., 1997; Northway, 2003; Scott et al., 2002; Tyrrell, Holland, Dennis & Wilkins, 1995; Wilkins & Lewis, 1999; Williams, Le Cluyse & Littell, 1996; Robinson & Conway, 2000; Bouldoukian et al., 2002; Jeanes et al., 1997; Robinson & Foreman, 1999) and ongoing medical research looking at physiological changes in both body and brain through fMRI, SPECT scans, VER, and biochemical analysis.
Conflict of Interest:
As Executive Director of the Irlen Institute, I receive financial compensation and have a commercial interest in the Irlen Method.