Wednesday, June 13, 2007


Liz Ditz alerted EoR to Cellfield. Aimed at dyslexics, Cellfield states:

This is why Cellfield has developed a program that targets phonological awareness in a unique way, freeing working memory for better comprehension and enhanced language skills. This is why after only two weeks, decoding skills can improve by almost two years and comprehension can improve by one year, whether your child is normal or dyslexic. (See The Australian Journal of Learning Disabilities Vol 10 issue 2).

Cellfield was developed by Dimitri Caplygin, a Science and Engineering graduate of UNSW. In a worrying example of how people with little experience in a specific area suddenly "see" the solution that the experts fail to notice, he states:

A chance encounter with severe reading disorders provided that connection in an emotive way, which led to that familiar inventive 'flash'. Dimitri was moved by the widespread suffering of dyslexics and was bewildered by the positions of exclusivity taken by many scientists as to the causes of dyslexia. With the fresh eyes of an outsider, Dimitri thought their positions were largely not contradictory, but part of a continuum of causes that could be tied together through computer science.

Such an insight is not, of course, impossible, but it is also one of the cardinal signs of "theories" and therapies lacking evidence. The FAQ is more marketing-oriented, rather than providing information.

Why does Cellfield work?

Brain scanning research shows a 'bottleneck' in areas where 'cross-communications' between the auditory, visual and motor functions normally take place. Addressing this neural abnormality is critical to the development of language skills. Cellfield is the first to target these 'cross communications' by simultaneously activating visual, auditory and motor pathways. Cellfield's research based design also induces attention, expands working memory and provides novelty with reward.

EoR was a little concerned about this sentence though, on success rates:

For illustration, somebody claiming a 90% success rate targeting a subgroup that represents only 20% of the reading disorder population is only an 18% success rate.

No. It's a 90% success rate in a specified subgroup. Extrapolating that success rate unmodified to different groups is unwarranted. It may be more successful in a different population group. It might be less successful. It might have the same success rate. This is the sort of error journalists fall into ("a study shows 90% of us are in danger of..." when it's actually 90% of middle-aged executives with prior heart problems, or so on) but EoR wouldn't expect a trained Engineer and Scientist to fall for it.

The only published evidence appears to be a Peer-reviewed and published study of 262 subjects who completed Cellfield Intervention, Australian Journal of Learning Disabilities Volume 10 - Number 2, 2005 which Cellfield makes a copy available of on its site.

The study provides this description of the Cellfield Intervention:

The intervention comprises ten one-hour sessions, each consisting of ten exercises. Some of these target phonological processing, requiring the concurrent activation of visual and auditory processing. Other exercises involve decoding and encoding activities using tasks such as finding text embedded in continuous random text without spacing. Motion graphics designed to stimulate the magnocellular pathways and other visual exercises requiring eye/hand coordination are also incoroporated into each session.

The children involved in the study were 187 males and 75 females ranging in age from 7 to 17 "who undertook intervention at the Cellfield Clinic at some time during a 24-month period". There was no control group. While impressive gains were made, the Discussion at the end of the study (EoR wonders if any of the parents considering using the Cellfield Intervention will bother to read the report, or make it to the end) is cautionary:

Notwithstanding these impressive results, there are several limitations to this study that should be taken into consideration. Firstly, a convenience sample of those seeking intervention for reading difficulties was employed. Thus, the generalizability of these results is uncertain. The students who undertook the Cellfield treatment during the span of this study may be peculiar to the population of Australian students who experience difficulty learning to read. The Cellfield Intervention is a commercial venture that requires a certain monetary investment on behalf of parents and thus the participants in the present study's sample may be representative of those who have reading difficulties but who have the financial resources to ameliorate their predicament.

The study also notes that longterm benefits, if any, have not been established. Furthermore, some of the scales used to measure improvement were based on norms established in US and British populations.

On its links page, Cellfield claims support from authorities such as Professor John Stein at Oxford University and Professor Max Coltheart at the Macquarie Centre for Cognitive Science. EoR could locate no reference to Cellfield at either of the relevant Professors' sites.


  1. As someone who works with the generally scientifically uneducated and/or illiterate facing the effects of neurological injury, something that the general public understands about as well as it understands research design or possibly less:

    This sort of crap really pisses me off. It's like the Dore program. Is the benefit as stated, or is the benefit from paying attention to someone, or is the benefit from actually just pursuing a structured, regulated exercise program that may reasonably be expected to train concentration anyway?

    This sort of insidious marketing is probably as detrimental to a public understanding of science as more explicit woo is, because it's so deceptive.

    Oh well, maybe we can put Dore, Cellfield and in a ring and let them fight it out with peer-reviewed papers. My money's on Cogmed, actually, because working memory training is something that does have a decent range of evidence behind it.

  2. It's Liz from I Speak of Dreams. Thanks, EoR!

    And Dubito, I was under the impression that the program was undertaking serious scientific evalution.


Note: only a member of this blog may post a comment.