Dyslexia: Advances in theory and practice, Dyslexia Research Fonndation and Center for Reading Research, 20-23 November 1997, Stavanger, Norway.

Dyslexia continues to baffle and intrigue teachers and researchers alike despite significant gains in understanding achieved since the first cases of 'word blindness' were published over a century ago. The ensuing century has seen research on dyslexia taking many directions, medical, psychological and linguistic. In the recent history of dyslexia research, the most important breakthrough has been associated with the phonological approach, which holds that the core deficit of dyslexia is one relating to 'phonological awareness', the ability to consciously analyze the phonological structure of words. While this is an easy task for fluent readers, young children often show only limited awareness in this respect as evinced by their limited ability to delete segments from or add them to words, difficulties in tapping to the number of segments in a word or, to a lesser extent, the number of syllables. Particularly impor tant in the development of the phono logical approach were intervention studies that showed that training in phonological awareness had beneficial effects on reading, which general lan guage training did not (reviewed in Ref. 1)

One impetus to this line of work came from research that had shown that speech is quite unlike print in that speech segments are intensely coarticu lated, leaving their boundaries hard or impossible to locate in the speech stream. The alphabet thus does not have a clear relationship to speech but to a more abstract representation of speech at the phonemic level. Reading is thus a difficult task because 'the reader must learn to listen with his eyes'. This statement by the late Icelandic educator sak J6nsson cap tures nicely the 'unnaturalness' of the act of reading when compared to speech. Speech is the quintessential human biological trait whereas writing is an artefact invented a mere 4000 or 50 years ago. Therefore it is perhaps not so surprising that some children should find it a difficult act to master.

Considering this emphasis on the phonological abilities of dyslexics it came as no surprise that a number of presentations at the Stavanger confer ence involved research on precisely this issue. Thus Charles Hulme (York, UK) described recent work on levels of phonological processing and the effect of training on reading acquisition. Numerous tasks have been employed to tap aspects of phonological aware ness, tasks which deal with differently sized linguistic units, syllables, their on sets and rhymes, and individual seg ments. Hulme's studies take the early

intervention studies a stage further, demonstrating that segmentation abil ity is a better predictor of reading suc cess than is rhyming ability and further showing that explicit linking of phono logical training to reading yields better results than training not explicitly linked to reading. A rather depressing note, though, was sounded by Richard K. Olson (Colorado, USA), who described recent intervention studies some of which show limited or even no long- term gain in reading ability.

Usha Goswami (London, UK) de scribed comparative studies of phono logical processing in a number of languages. Specifically, she asked whether phono logical development, as measured by such tasks as previously described, is identical across different languages and whether awareness at the same linguistic levels predicts read ing development in different linguistic environments. To the first question she proposed a tentative affirmative answer, to the latter a tentative nega tive answer, noting for instance that rhyming ability shows limited predic tive value for reading acquisition in German as compared to English or Swedish. Another issue concerns the status of orthographic units in differ ent languages. English with its 'deep' orthography shows the marked in fluence of orthographically familiar units in reading (e.g. conventional rhyme spellings) whereas 'shallow' orthographies show no such effect, presumably because of their greater orthographic transparency.

Speaking of different orthog raphies and their effect on reading ac quisition, the prize item for research is surely Chinese with its unique writing system which has fuelled immense speculation on the relation of speech to writing, but unfortunately much less empirical research. Thus, as discussed by Che Kan Leong (Saskatchewan, Canada), there is almost no evidence to be had about the extent of dyslexia in Chinese, which has led to claims, surely mistaken, of its non-existence. A perva sive belief about the Chinese writing system is that it is logographic, entails writing at the level of words or ideas rather than at the level of sounds. This view of Chinese was effectively chal lenged by John DeFrancis a number of years ago when he argued for the importance of phonetic aspects in Chinese writing². This point was rein forced by Che Kan Leong who also dis cussed the importance of the more fine-grained structure of Chinese char acters, in terms of radicals and strokes. The role of these units in the reading of Chinese has seen only limited re

search but there is evidence that better readers show more 'radical awareness' than poorer readers. Obviously, Chinese provides a fascinating comparative case to studies of alphabetic reading.

The biological aspect of dyslexia that has received the greatest atten tion in recent years is that of genetics, which indicates an important genetic component. Recent localization studies have pointed to chromosomes 1, 6 and 1 5, with the strongest indications for chromosome 6. Herbert Lubs (Miami, USA) described an ongoing familial study of a large Norwegian dyslexic family where intensive studies of the three chromosomal regions previously mentioned failed to show any linkage. The identification of a clear localiz ation for a dyslexia gene is obviously problematic, in part because of compli cating factors like the absence of a reli able diagnostic test, possible hetero geneity and decreased penetrance (e.g. because of various compensatory strategies open to dyslexics).

Many speakers focused on the issue of definition. Recent years have witnessed a turning away from the previous emphasis on discrepancy defi nitions where dyslexics were consid ered to be poor readers with a high IQ and were considered different from those 'generally backward' readers who showed less or little discrepancy with IQ. This definition has, however, proved problematic and arguments for a revised definition have been put for ward, emphasizing~readin~e~ardation without reference to IQ and thus, by implication, treating all reading- backward children alike³. The defini tion of dyslexia is not purely a scientific matter since it can have profound differ ences for the remedial services offered

- though not for instance in Finland where Pekka Niemi (Turku, Finland) reported that all backward children receive identical special tutoring. This is certainly not the case in the USA where states operate with different definitions of dyslexia with disagree ments in how discrepant any discrep ancy should be. Moving across a state boundary can thus easily lead to a child losing his status as dyslexic, 'a quick cure indeed' as noted by P.G. Aaron (Indiana, USA).

While it is clear that the trend in research over the past decade has been towards a more unitary approach to dyslexia, with defective phonological processing being seen as the common und~rlying cause, many researchers still entertain the hypothesis that sub typing of dyslexics is necessary. Thus R. Malatesha Josh (Oklahoma, USA) claimed that a distinction should be
made between disorders of decoding and of comprehension with a third group showing deficits in both cat egories. This research emphasizes the need to look more broadly at the lan guage deficits seen in dyslexia. Margaret Snowling (York, UK) presented re search along similar lines and
explicitly linked such a dichotomy to recent connectionist models of reading4.

The award for the most rousing presentation at the conference must surely go to Rod Nicolson (Sheffield, UK), who spoke of his and Angela Fawcett's research on 'The role of the cerebellum in dyslexia'. Speaking late on Friday afternoon, when a hard day's conference work was beginning
to af fect the attentional capacities of those present, Nicolson brilliantly managed to arouse all present to a number of provocative ideas. Whether this was entirely due to the nature of the message being delivered, or to the deft theatrical manner in which the research was presented, is unclear in this writer's mind. But consider the claims: dyslexia is not a specific linguistic deficit but rather one of generalized motor skill deficit. The focus of an ongoing research project for 12 years, Nicolson and Fawcett originally investigated a range of motor skill deficits in dyslexia, find ing these to be pervasive and wide- ranging, involving amongst other things balance and, response latencies. The most recent aspect of their work links these motor deficits and dyslexia to a common cerebellar impairment. The ef fect on reading and spelling is held to arise because 'they
require both phono logical skills and good automatization, therefore relying on two separate roles of the cerebellum in learning skill. The need for automatic, skilled eye move ments (again cerebellar-enhanced) in reading also adds a further layer of dif ficulty' (from the speakers' handout). As evidence for this they cite their recent research on the morphology of
dyslexic brains at the Beth Israel hospital in Boston where they found significant anomalies in cell size in cerebellar cortex and inferior olive.

Alas, this last finding probably raises more doubts than it dispels since the brains investi gated are the same ones which earlier had shown abnormalities in the visual magnocellular pathway. Those findings have recently been heralded as 'the most convincing evidence' for the magno-
cellular theory of dyslexia. a theory that attempts to explain dyslexia in terms of rapid sensory processing5. The question is whether even further abnormalities can be found in these brains making it very difficult to infer anything at all from them about the precise neural causes of dyslexia. In contrast, Kenneth Hugdahl (Bergen, Norway) described research confirming earlier findings of a significantly smaller planum temporale asymmetry in dyslex ics than in a group of matched controls.

Still, it is clear that dyslexia is often accompanied by other non-linguistic deficits as argued also by Ayran van der Leij (Amsterdam, Netherlands) who of fered a slightly different take on the issue of processing speed, linking it primarily to problems of automaticity. The approach of van der
Leij, Nicolson and Fawcett can in some sense be traced back to earlier medical views of dyslexia that predate the recent in tense focus on the role of phonological awareness in dyslexia. In contrast to those earlier views, which led to sug gestions that practice in finger tapping or step dancing could lead to improved reading, Nicolson and van der Leij were seen to be in agreement with propo nents of the phonological approach in recommending specifically linguistic training to help dyslexic children over come their reading handicap.

Considering that an attempt was made in the presentations of these two speakers to extend the description of dyslexia beyond the core symptom of phonological processing it was some thing of a disappointment that the nature of linguistic processing beyond the phonological was not covered
in greater detail at the conference with the exception of a talk by Carsten Elbro (Copenhagen, Denmark) on morpho logical processing in dyslexia. Elbro's research has found that morphological factors do play a role in the reading of dyslexics, going so far as to claim that dyslexics can often be seen to employ a 'morphological strategy' of reading rather than a phonological one. Margaret Snowling also mentioned her ongoing work, in
collaboration with Uta Frith and Alison Gallagher, in which children who are genetically at risk of developing dyslexia are being monitored from an early age. A number of lan guage problems are seen in these children long before reading age. This would seem to be a most fruitful area for
further research.
 

Jörgen Pind
Professor of Psychology. Deparrment of Psychology, University of Ireland, Oddi,
15-701 Reykjavik. Ireland. tel: +3545254085 fax: +354 552 6806
e-mail: [email protected]

References
 

1 Gotwami, u. and Bryant, P. (1990) Phonological Skills and Learning to Read, Lawrence Erlbaum

2 DeFrancis, 1. (1989) Visible Speech: The Diverse Oneness of Writing Systems, University of Hawaii Press

3 Lyon. G.R. (1995) Toward a definition of dyslexia Ann. Dyslexia 45,3-27

4 Snowling, M., Hulme, C. and Nation, K. (1997) A connectionist perspective on the development of reading skills in children Trends Cognit. Sci. 1, 88-91

5 Stein, I. and walsh, v.119971 To see but not to read, the magnocellular theory of dyslexia Trends Neurosci. 20.147-152