The myth of early intervention

In theory, early intervention for learning difficulties should be a good thing: catch the children before they fail, give them what they need to succeed as early as possible.

But in the field of dyslexia remediation, the focus on early intervention has given rise to an unfortunate, and very untrue, myth: that the child who is not reading by third grade is doomed to failure. Experts quote somber statistics about the percentage of children who never learn to read; the story circulates that one state or another plans its future prison construction based on the rate of reading failure in the third grade.

An intriguing new study from Italy seems to put the lie to this theory. Researchers decided to look at how older kids (6th-8th graders) compared to younger ones (4th to 5th graders) in their rate of improvement with reading fluency, with two different interventions. Both interventions had been shown to increase reading accuracy in prior studies involving kids from 2nd to 8th grade. All kids met DSM-IV diagnostic criteria for dyslexia at the outset.

One set of kids from each age group received a intervention geared to building more automatic sublexical decoding skills with specific training in syllable identification; since Italian is a phonetically regular language, this is an effective strategy for phonetic decoding. That is, the syllable pa will always be pronounced /pa/ no matter where it appears in a word.

The alternate intervention was based on a hemispheric stimulation, using a computer program that displays words in the right or left visual fields in order to enhance left hemispheric development.

The results of the study were that older children did as well as younger children in the rate of gains made in reading fluency and various tests of reading accuracy, with one exception: with the phonetic based, syllable training, the younger children did better on tests of words read in isolation.

In other words - at least with the particular interventions studied, older kids could do just as well as younger, at least for the important end goals of improved reading fluency and accuracy of text reading.

But this brings up another question: what is the source of the early intervention myth? As the Italian researchers noted, very few studies of reading interventions look at the end goal of reading fluency. A meta-analysis in 2003 showed that "90% of the studies included standardized dependent measures of real word-reading accuracy, whereas none included measures of fluency." But deficits in fluency are the most persistent, even after accurate decoding is learned (more easily in phonetically transparent languages, more difficulty in irregular systems like English or Danish). Thus, tests of single word reading accuracy provide no indication of how the child will fare as a reader over the long term; many dyslexics grow to adulthood reliant on slow and laborious decoding.

In the US, the strong emphasis on phonetic strategies may explain both the high rates of reading failure and the early intervention myth. Despite sanctimonious claims that these phonics-based programs are "research-based", the research is not geared to mature, skilled - fluent - reading. Instead, the research tests the one thing that older children are not particularly good at learning: improved ability to recognize words in isolation. And because of this deficit in the studies, the researchers have done very little to develop or study programs aimed at fluency development. In other words, they haven't bothered to study the programs that are likely to help older children.

Of course, such programs do exist-- but as they tend not to employ the phonetic strategies already labeled as being supported by research, they are viewed with disdain, seldom implemented in schools, and remain unresearched. So we end up with only programs that are neither effective to build reading fluency nor to help older children improve their skills: hence the abiding myth that the child who reaches age 8 without reading is already a lost cause.

The Study: Fluency Remediation in Dyslexic Children: Does Age Make a Difference? by Patrizio E. Tressoldi, Maria Luisa Lorusso, Federica Brenbati and Roberta Donini, Universita` di Padova, Padova, Italy [DYSLEXIA, Published online in Wiley InterScience, DOI: 10.1002/dys.359]

Questioning research methodology & other notes

The November 2007 issue of the research journal Dyslexia raises some interesting questions about the applicability of a model of controlled empirical studies to research into dyslexia treatment methods. In an article entitled What kind of evidence do we need for evaluating therapeutic interventions?, Mary Haslum of University of the West of England, Bristol, points out that pitfalls in attempting to structure a controlled study, noting that researchers do not even agree on the definition or diagnostic criteria for dyslexia. She suggests that a more realistic approach to research would be to develop more systematic criteria for evaluating qualitative research based on case studies.

Ina separate article along the same lines (Criteria for Evaluating Interventions), T.R. Miles, with his characteristic delightful style, begins by noting the perils of the placebo or Hawthorne effect, and somehow meanders to a point where he notes the failure of any studies to account for the possibility that some participants have extra-sensory perception (ESP). [Note: with few exceptions, most scientific researchers produce journal articles that are tedious at best, and often border on incomprehensible. Miles is one of the very consistent exceptions -- always lucid, and quite often entertaining and imaginative both with his use of language and the avenues he chooses to explore.] In the end, Miles provides a set of reasoned examples that similarly challenge the idea that dyslexia research can or should be limited to the "gold standard" of the controlled study, while also urging an awareness of the various factors that can lead wishful thinkers to misinterpret the significance of their data.

Citation: Dyslexia, Volume 13, Issue 4 (November 2007)

The merging of Perceptions

An interesting new study -- researchers have found that the visual perception/interpretation of gender is affected by hearing. They played pure tones within the frequency range of male or female voices at the same time that subjects were looking at digitally morphed androgynous faces, and found that the faces were identified with as female when accompanied by the feminine frequency (160 to 300 Hz) and male when accompanies by the lower male frequency (100 to 150 Hz). However, the subjects were unable to determine whether the tones themselves were female or male when presented with paired tones; in that setting they simply chose based on the relative pitch of the two tones, regardless of actual range.

The study: "Auditory-Visual Cross-Modal Integration in Perception of Face Gender," published in a recent issue of Current Biology. The study's co-authors are investigators at Northwestern University's Visual Perception, Cognition and Neuroscience Laboratory: lead author Eric Smith, graduate student, Marcia Grabowecky, research assistant professor of psychology, and Satoru Suzuki, associate professor of psychology in the Weinberg College of Arts and Sciences at Northwestern.

Web link: http://www.medicalnewstoday.com/articles/86668.php