DYSLEXIA:   A  MULTIDIMENSIONAL  APPROACH  

     BY  MARIA CHAPA


Between 2 and 10% of the population of the English
speaking world have great difficulty with learning to
read in spite of educational resources, a normal  or
above IQ, and no obvious sensory deficits.  This disorder
is known as dyslexia and is the subject of much
cognitive,  and biological research.  The ambitious goal
of achieving an integrated neuroscientific explanation of
disorders like dyslexia, is beginning to be realized
because of  the collaborative efforts of scientists from
different disciplines working at several different levels
of analysis.  Studies from the fields of cognitive
neuropsychology, neuropsychology, neuroanatomy and
genetics  with their contributing theories will be
discussed here.  Each has made significant contributions
to our somewhat limited knowledge of this  complex 
disorder.

To understand dyslexia, we must first have a basic
knowledge of the underlying processes involved in
reading.  For this basic knowledge, we turn to the
discipline of  cognitive psychology,  which studies
normal mental processes such as perception, memory,
language, and reading.  Neurology  and cognitive
psychology come together to form cognitive
neuropsychology, the study of disorders of cognitive
function that arise as a consequence of brain injury. 
Study of the deficits caused by brain injury has
contributed much to our understanding of the way we read. 
One of the goals of cognitive neuropsychology is to draw
conclusions about normal, intact cognitive processes from
the study of patients with acquired dyslexia.  The term
acquired means that these patients had already learned to
read prior to their injury.  In contrast, developmental
dyslexia refers to reading difficulties that become
apparent  when children are learning to read.  Although
quite different in origin, there is evidence to suggest
that some parallels can be drawn between  developmental
and acquired dyslexia. 

One of cognitive psychology's most important
contributions has been that of it's word recognition
model.  These cognitive models suggest that  word
recognition is the product  of orchestrated activity
that occurs within a number of cognitive subsystems which
operate at least partially independently of one another. 
These subsystems are referred to as modules, and are very
important in the understanding of reading disorders.  For
example, if different operations for word recognition are
handled by different cognitive modules, then it is
possible  that  people with reading disorders would
exhibit some aspects of reading that are normal, while
other others are impaired, producing different patterns
of reading disorders.  This is in fact found to be true. 
Through careful analysis of these patterns, cognitive
neuroscientists can  draw their conclusions about  normal
reading processes.

The cognitive model for word recognition presented here
is a simplified version  that emphasizes the cognitive
processes involved in recognition of single written
words.  When printed word is presented , the first module
to be activated s the visual  analysis system.  It is the
job of the visual analysis system to identify letters 
and  note their position in the word.  It must encode
letter identities and positions before the next module, 
the visual input lexicon can identify strings of letters
as familiar words.  The visual input lexicon is a mental
word storage unit which contains representations of all 
familiar words, or recognition units.  Becoming familiar
with new written words involves creating new recognition
units  for them in the visual input lexicon and forming
associations between these units and representations of
meaning and pronunciations.  This is an important part of
learning to read.  Comprehension of written word requires
the activation of the semantic system.  This system
contains everything you know about a word except how to
pronounce it.  This knowledge is contained in a separate
word store, the speech output lexicon.  The function of
this system is to make the spoken word form available to
the speaker. a word such as  neuropsychology' is thought 
to be retrieved in one go from the speech output lexicon. 
But the phonemes in  neuropsychology'  cannot be
articulated all at once.  To explain how this is done, 
the presence of a phoneme level  which is a short term
store in which phonemes cam he held in the interval
between retrieval and articulation.  Studies of this
phoneme level suggest that it is capable of holding
strings of several words if need be.  An alternative
route exists in which it is thought that whole word
recognition is attained.  It is a direct route from the
visual analysis system to the speech output lexicon, 
bypassing the semantic system .  These are two ways in
which word recognition is achieved.  But both of these
routes require existing representations in the visual
input lexicon which is limited to familiar words that
have been stored previously. These paths do not account
for unfamiliar word strings or non words.  a third route
exists for this important role in the pronunciation of
unfamiliar words.   When faced with a word we don't know,
the visual analysis system will identify it's letters
making it possible for a component called the grapheme
phoneme conversion to translate unfamiliar letterings to
phoneme strings.  This output activates the phonemes at
the phoneme level, where they can be articulated.  This
third route involving this grapheme phoneme conversion is
extremely important  one.  Although  theorists believe
that this component  is probably only used occasionally
by skilled adult readers as the dominant route from print
to sound,  the young or unskilled reader however will
come across unfamiliar words which must be sounded out by
grapheme phoneme conversion.  Damage to this component of
the process is implicated in several disorders, and is
the subject of much research.  

Acquired phonological dyslexia provides us with a good
example of the significance of this cognitive theory. 
This form of dyslexia was found only after being
predicted by a cognitive theory, and has had a very
direct implications for theories of normal reading
processes.  First reported in 1979 by Beavois and
Derousne, phonological dyslexia is a disorder in which
patients are unable to read unfamiliar and non words. 
Familiar words however can be read reasonably
successfully.  As we know from our cognitive model  the
grapheme phoneme conversion component is responsible for
this task.  Phonological dyslexics read on a whole word
basis and are severely impaired at this level.  Surface
dyslexia in contrast, rely on this grapheme phoneme
conversion component heavily, and show impairment in
whole word recognition.  This is important in that this
is evidence of  what is known as a double disassociation
indicating that the two processes are mediated to some
degree separately within the mind and brain, supporting
their models.  Phonological dyslexia is also of interest
because many children diagnosed as having developmental
phonological dyslexia show symptoms very much like those
of acquired phonological dyslexia.  Although caution must
be used in drawing conclusions between the two, some
evidence suggests that there are some parallels that do
exist. 

This identification of the phonological problems in
dyslexics has spurred much research in the biological
sciences as well.  Anatomical studies attempting to
identify specific areas of the brain in which
phonological processing occurs have been somewhat
successful.  The anatomy of the phonological system is
known in broad terms, with much of the evidence stemming
from pathological correlates in  aphasic patients with
phonological deficits in speech comprehension.  These
include damage to Wernicke's area  which holds the
memories for the sequences of sounds that constitute
a word, and Broca's area which contains the memories for
the sequencing of muscle movement need for articulation. 
Paulesu et al..,1993 ,and  Price et al..1995, showed
however that Wernicke's area is not only activated by
activity to the auditory system, but also by language
tasks that do not involve any auditory stimulation.  One
study by E. Paulesu et al..1996 , combines cognitive and
biological investigations in comparing normal and
abnormal phonological processing.  The paradigm consisted
of a rhyming task, and a phonological short term memory
task in which verbal material was presented visually. 
These tasks are designed to engage the phonological
system.  The subjects were  five compensated
developmental  in which their only difficulty was in
phonological processing.  a control group of normals was
used.  PET scans  used to determine which areas of the
brain were activated during these phonological tasks,
showed that for dyslexics, a subset of the brain region
normally used  in phonological tasks was involved.
Broca's area was involved during the rhyming task, and
the temporo  parietal cortex was involved during the
memory task, but not in concert, as was seen in the
normal control group.  But perhaps  more importantly was
the discovery that in the dyslexics, the left insula was
never activated.    Paulesu et al.. theorized from their
studies that at a physiological level, performance of
phonological processing tasks would depend on the
interaction between Broca's areas and the supramarginal
gyrus.  In normals, both structures as well as the insula
were activated.  The insula is thought to act as a bridge
between these two structure and that the function of this
bridge is one involving conversion between codes. 
Perhaps the insula and the grapheme phoneme conversion
component are linked.    In addition the independent
activation  of the anterior and posterior speech areas
supports the notion that representations of unsegmented 
and segmented  phonology  are functionally and
anatomically separate.

The field of genetics has provided us with further
insights regarding certain types of dyslexics.  It is
clear from investigations of twins and families that
dyslexia is a developmental disorder with a genetic
origin.   Grigorenko et al..,1997  used single word
recognition and phoneme awareness as phenotypes, and was
able to significantly link  their single word reading
phenotype  near the centromere on chromosome 15, and very
significantly link phoneme awareness to the short arm of
chromosome 6.  The researchers interpreted this to
indicate that cognitive dissociable components of the
reading system are linked to separate genes, and that
mapping from genes to aspects of cognition may be quite
close.  This report represents a collaboration of
neuropsychology and genetics, and is one of the first
independent replications of genetic results for a complex
human behavior.  a basic point here is that dyslexia was
a cognitive science success story  before it became a
success story for modern molecular genetics.  Careful
analysis of the behavioral phenotype was undertaken
before recent advances could be made in understanding the
genetics.  If the linkage studies of dyslexia had used
letter reversals, or just reading comprehension as a
phenotype it is very unlikely that as much progress would
have been made.  It is important to note that few of the
complex behavioral phenotypes now being tested for
linkage have an equally mature cognitive science analysis
on which to build.  

In conclusion, this study is  exemplary in that it
combined state of the art of cognitive science with that of
genetics to replicate and extend previous linkage
results for dyslexics.  By utilizing techniques from
different disciplines, we are able to integrate theories,
and further research.    

                                                    
REFERENCES

1.Human Cognitive Neuropsychology    Andrew W. Ellis and
Andrew W. Young, 1994

2.Dyslexia: A Neuroscientific Approach  Frank H. Duffy,
and Norman Gecshwind, 1985

3.Reading, Writing and Dyslexia: A cognitive analysis
Andrew W. Ellis. 1993

4.Physiology of Behavior Neil R. Carlson. 1986

5.Is developmental dyslexia a disconnection syndrome? E.
Paulesu et al.. Brain, 1996 119, 143-157

6.Neuroanatomical and neurophysiological aspects of
dyslexia Cynthia A Riccio,Ph.D.. 

Topics in Language Disorders ,Feb. 1996

7.Using Genetics to Dissect Cognition Bruce F.
Pennington, American Journal of Genetics, 60:13-16,1997  

Copyright © 1997, Dr. John M. Morgan, All rights reserved - This page last edited March 17, 1997
If you have any feedback for the author, E-mail me