1. The document summarizes the 3rd Oxford-Kobe Symposium which brought together over 30 world experts to discuss the latest research on dyslexia. 2. Key findings included evidence that dyslexia is caused by both genetic and environmental factors and deficits in both the auditory and visual pathways. Research also showed differences in how dyslexia manifests depending on the writing system, like differences found between English, Chinese, and Japanese scripts. 3. Speakers also discussed promising avenues for intervention including computer-based training programs and games to improve skills like phonological awareness, rapid naming, and visual attention which could help remediate dyslexia.

1. 1/5
Good progress and an emerging
consensus in the scientific understanding
of dyslexia as well as an appreciation of
how skilled and appropriate intervention
can help people with dyslexia were the
salient themes in this outstanding
event.
The organizers, Dr Maki Koyama of Rutgers
University, USA, Professor John Stein of
the University of Oxford, UK, and Clarice
Davies of the Dyslexia Research Trust,
UK, had assembled more than 30 speakers
and other world experts to discuss the
latest research into the causes of learning
difficulties.
Dyslexia International attended the event,
which was divided into two parts:
1. Academic – The Neurobiological Basis
of Dyslexia in Alphabetic, Syllabic, and
Logographic Scripts
and
2. Public Day – The Neurobiology &
Treatment of Dyslexia
In a brief report it is not possible to do
justice to all the papers presented but
the slides for the 4th day are available on the
web site of the Symposium
(http://www.oxfordkobe.com/program.
html).
It was clear that dyslexia was an extremely
complicated bundle of conditions, that
is to say it is multivariate, and that both
genetic and environmental factors were
intertwined. Although it may have seemed
in the last decade that there existed two
predominant views opposed to each
other which, on the one hand, posited a
phonological deficit arising from disorders
in audition and, on the other, weaknesses in
visual processing, it is now more apparent
how intimately these two systems are
linked.
The acquisition of a high level of
literacy – that is the ability to read fast and
accurately - depends on the establishment
of self-reinforcing loops of neural
communications between different parts of
the brain. These circuits may not be built
reliably if one or both of the two principal
pathways are not functioning properly
and/or they are then not bound together.
A deficit in the auditory system will not
Report – Oxford-Kobe Symposium
Progress and Hope
Reporting the 3rd Oxford-Kobe Symposium,
11 – 14 April 2013

2. 2/5
adequately distinguish patterns of sounds,
segment them correctly and then further
process these patterns into a phonological
lexicon. An impairment in the visual
system, which itself has multiple paths, will
cause failures in the accurate tracking of
a line of print and discerning at speed the
different shapes of letters or characters; in
this case an orthographic lexicon will not be
established.
Over four days we heard evidence about the
failures in these systems and their interplay,
causes both genetic and impoverished
exposure to experience, but also about
effective methods of remediation based on
the brain’s remarkable plasticity.
Nina Kraus from Northwestern University,
US, showed that people with
dyslexia had selectively inconsistent neural
encoding of fast-moving speech,
in particular the differences between
consonants. And yet these children could
be helped with hearing devices which
decreased the distraction of background
classroom sounds. In an impassioned plea
she also urged the merits of early music
training which helped the discernment
of rhythm, timbre and pitch, as well as
the training of auditory memory. Using
different techniques, April Benasich of
Rutgers University, US, identified the ability
to perform fine-grained acoustic analysis
as a predictor of later disorders in language
learning. Franck Ramus of the Ecole
Normale Supérieure, France, illustrated a
difference between the two hemispheres of
the brain in responses to sounds. In what
is called the gamma range (25 – 35 Hz),
which is critical for phonemic processing
(the smallest segments of speech), there
was a deficit in the left hemisphere whereas
in adults there was a deficit in the longer
intervals (at 2 Hz).
Turning now to the visual pathway, Trichur
Vidyasagar of the University of Melbourne,
Australia, said in a striking phrase that the
brain was a ‘hacker’. It had to learn a skill
for which it had not evolved. The searching
mechanism for survival was a random
search in space with rapidly flicking eye
movements.
But to read one had to use the eyes ‘like a
spotlight’ to travel accurately and swiftly
in a linear direction. The manager of this
system was the dorsal (‘upper’) stream in
the cortex towards the back of the brain and
it was critically involved in visual attention.
The neural pathway in this system was
composed of magnocells, one of three
principal paths from the retina to the
brain. It responded primarily to moving
stimuli and it used those stimuli as both
information in itself and information
about how to move the eyes. In reading
this enabled the fovea of the retina to bring
its parvocells for the detection of fine
shape and colour unto the target. This is
the ventral (‘lower’) stream. The correct
formation and location of magnocells could
be compromised by genetic factors leading
to incorrect development, insufficient
nutrition and immune disorders, as
explained by Stein. It is important to note
that the auditory pathway is also composed
of magnocells.
In support of Vidyasagar was Sylviane
Valdois of Pierre Mendèz-France University,
who stated that not all children with
dyslexia exhibited a phonological disorder
but some showed a reduction in visual span
attention, limiting the number of visual
elements that could be processed in parallel.
The evidence was therefore for subtypes of
dyslexia with different cognitive disorders
and different brain dysfunctions. Having

3. 3/5
looked at particular sensory systems we
turn to wider aspects of the integration
of these systems, which introduces yet
another complicating factor, namely varying
trajectories induced in the developing brain
as it is exposed to experience. And, because
reading is such a complex skill, it is not
surprizing that developmental disorders can
be attributed to deficits in both individual
systems and/or their intricate intermeshing.
Stanislaus Dehaene of the Collège de
France is famous for indentifying what is
called the visual word form area, in the
left hemisphere in the ventral pathway.
Dehaene showed how this area was deeply
transformed as one learned to read, with
the intriguing result that activations in
this area in response to faces became
reduced but more prominent in the right
hemisphere. This is an example of brain
plasticity. Furthermore, reading expertise in
alphabetic scripts also impacted the earliest
stages of processing in the visual cortex.
This is called ‘top-down’ influence. New
studies showed that the exact site of this
early effect differed between French and
Chinese subjects, which must be promoted
by their different scripts.
Fumiko Hoeft of the University of
California San Francisco, US, used a new,
different kind of technique to predict
reading outcome. This study centred on the
‘white’ matter of the brain, more properly
the fascicles, which connected different
regions of the cortex. Laurie Cutting of
Vanderbilt University, US, also examined
cortical – cortical interconnections as
well as cortical – subcortical connections.
Subcortical processing is an emerging field
of research as scientists are addressing
the contributions of structures like the
thalamus, basal ganglia, brainstem and
cerebellum. In other words, reading
and reading failure, is not just about the
‘thinking’ part of the brain. The script you
are reading now is an alphabetical script.
But there are other methods of encoding
language on paper such as Chinese
and Japanese. One of the aims of the
symposium, in addition to gathering experts
from all over the world, was to discuss
whether dyslexia was an innate condition
potentially affecting any person or whether
its cause lay in different scripts. The short
answer is that it did not but there are subtle
nuances. For example, preliminary studies
showed that in Japanese there appeared to
be a different ratio of dyslexics in Katakana,
the simplified syllabic system, and Kanji
the logographic system used by the more
literate, with a higher rate in the latter.
It would not appear that similar numbers
of children in Japan and China have been
tested as in Europe and America nor are
the research paradigms exactly the same
but it certainly seems that children in these
countries are susceptible to similar early
processing disorders.
Japanese offers the possibility of teasing
out to some extent phonological awareness
(PA) and visual cognition (VC) as children
deployed the different scripts. Akira Uno of
the University of Tsukuba, Japan, presented
results which separated the effects of deficits
as children matured. PA and VC were
important predictors for ideographic Kana
reading in preschool children. But as they
grew older the degree of the phonological
contribution to learning difficulties
decreased and visual cognition also ceased
to be a significant predictor in later grades.
Degree of automatization remained an
important predictor in Kana but vocabulary
was the most important predictor for
Kanji word reading [which emphasizes
the relevance of cognition and ‘top-down’
processing].

4. 4/5
‘Is Dyslexia Really Uncommon in Japanese
Children?’ was the title of the talk by
Tatsuya Koeda of Tottori University of
Japan. In 2011 his team had developed the
Hiragana Reading Test, which was officially
adopted the Ministry of Health, Labour and
Welfare.
However this test was not adequate for
early detection – considered essential –
because of the reading level required. So
they developed a new reading test and a set
of guidelines, from which they concluded
that the prevalence of dyslexia in Japanese
children was between 0.82 % and 2.1 %,
with a bias to the larger value. [By Western
standards this does seem low. However
they had used information provided by
teachers and acknowledged that criteria for
identifying reading and writing difficulties
in dyslexic children were not well known
amongst school teachers.]
Hua Shu confirmed that Chinese children
could suffer from impaired auditory
processing. Dyslexic children were
found perceive lexical tone contrasts less
categorically and less precisely than age-
matched controls in a discrimination task
[thus arguing for a universal basis and not a
culturespecific source of auditory deficits].
These observations were supported by
James Booth of Northwestern University,
US, who talked about the neural basis of
lexical processing in Chinese. His research
pointed to a universal network involved in
the reading process but subtle differences
across languages. In Chinese, the visuo-
spatial demands of processing characters
was greater than, for example, in English, in
which there where many fewer basic forms
and their mapping to phonology was semi-
regular whereas in Chinese it was arbitrary.
The conclusions were that the unique
characteristics of the different orthographies
resulted in cross-linguistic differences.
Two speakers gave wide-ranging
perspectives on dyslexia. Ken Pugh of
Haskins Laboratories, US, spoke in both
parts of the Symposium. He said that
specific instruction in reading was essential;
good treatment in the form of multi-modal
training over time, definitely evoked reliable
improvements. He also said, bluntly, that
‘many children can become dyslexic because
of lousy teaching’. Pugh laid great stress on
the plasticity of the brain. In line with other
speakers, he said that children with reading
disability failed to develop a coherent left
hemisphere circuitry. In a masterly survey
Albert Galaburda of the Harvard Medical
School, US, explored numerous themes.
‘Dyslexia research has gone further than
research in any other learning disorder.’ His
own well-known early findings of unusual
patterns of neuronal migrations in people
with dyslexia had held up well. Cross-
disciplinary research would be needed,
from geneticists, through molecular,
cell and developmental biologists, to the
numerous forms of neuroimaging and
other measurement systems, right up to
psychologists.
Otherwise we would never get from genes
to behaviour. He warned against excessive
attention to ‘modularity’ - a small pathology
in one region could have huge consequences
in others. Much work remained to be
done in finding the more precise timing
of developmental stages but new, very
powerful, techniques were becoming
available. In response to his own question,
‘Where next?’, he replied ‘Southwards’ or
downwards from the cortex. For example,
the cochlear nucleus had a large number
of receptors sensitive to androgen, which
meant that the debate about the prevalence
of dyslexia in males as opposed to females
was a live issue.

5. 5/5
What about genetics? Silvia Paracchini of
the University of St Andrews, UK, gave
us an update on genome-wide association
studies. [Some of this work was initiated
by the EU in the Europe-wide ‘Neurodys’
programme.] The sample sizes were not
large enough to be able to reach robust
conclusions.
Also, assessment was specialized and time
consuming; the languages were different;
and the exhibited behaviours were complex.
That said, a few very interesting biological
markers had been singled out. For example,
one gene called PCSK6 played an important
role in a pathway determining left/right
body asymmetries. The association
between the gene found and individuals
with dyslexia appeared to be specific – as
measured by tests of hand skill – but there
was no correlation between dyslexia and
handedness.
We have considered ‘progress’. What about
‘hope’? There was much cause for optimism.
As well as Kraus’s positive message about
the value of music training three other
speakers spoke about practical forms of
remediation.
Following long-term studies which
predicted children at risk, Heikki Lyytinen
of the University of Jyväskylä, Finland,
developed Graphogame, a computer game
which was enjoyable for children whilst it
specifically trained the association between
basic letters and sounds, permitting graded
levels of competence. It was proving very
effective when it was combined with an
optimal phonics approach. It was available
in Finnish and English. Trials were running
in more than 15 countries in four continents
including four countries in Africa.
Kristen Pammer of the The Australian
National University (after confirming the
importance of the dorsal/magnocellular
path) introduced preliminary findings from
the use of computer games to improve
spatial processing. More work was needed
but early results were encouraging. Andrea
Facoetti of the University of Padova, Italy,
said that letters had to be precisely selected
from irrelevant and cluttering letters by
rapid orienting of visual attention before the
correct letter-to-speech sound integration
applied. He was also using action video
games to improve attention and the tracking
of moving objects. Improved reading
abilities had been shown in groups of
children using the games.
On the final day in a panel discussion there
was a question about the communications
between researchers and those shaping
education policy in administrations, and
also whether the research presented was
yet strong enough to convey unambiguous
information to all governments.
There was a mixed response, signalling a lot
of variation between countries. However
at State level in the USA for example, there
were numerous initiatives and there
was certainly a lot of research funding from
federal sources.
The event was sponsored by St Catherine’s
College, Oxford, the Kobe Institute and the
Dyslexia Research Trust.