Sue Buckley, BA, CPsychol, AFBPsS
Professor of the Psychology of Developmental Disability
© The University of Portsmouth Inaugural Lectures, 9 May 1996. Acting Vice-Chancellor: Dr. Michael Bateman, BA, PhD
© The Down Syndrome Educational Trust 1997. All rights reserved.
Reprinted with permission of Frank Buckley
Operations & Finance Director
The Down Syndrome Educational Trust
The Sarah Duffen Centre
Belmont Street, Southsea
Portsmouth, Hampshire
United Kingdom PO5 1NA
+44 23 9282 4261 Fax: +44 23 9282 4265
Translations:
Background
I began investigating the mental abilities of children with Down's syndrome in Portsmouth in
1980 after receiving a letter from a father. Leslie Duffen, describing how he had discovered
that his daughter Sarah could begin to learn to read at the age of three years. Sarah was born
with Down's syndrome. At three, she was just beginning to imitate and to use single words in
her speech. Leslie taught Sarah to read on flashcards the words that he wanted her to be able
to use in her speech and he observed that she began to use the words she had learned from the
printed form at a faster rate than those she only experienced in the spoken form.
When Leslie wrote to me in 1979, Sarah was twelve years old. She had good speech, was reading and writing at her age level and being educated in a local comprehensive school. She had received all but one year of her education in mainstream schools and was considered to be exceptionally able for a child with Down's syndrome. Leslie felt sure that her exceptional progress had been the result of teaching her to read early and that other children might be helped in the same way.
I found Leslie's letter surprising and intriguing. His experience with Sarah suggested that pre-school children with Down's syndrome could learn to read and that reading might be a "way-in" to spoken language for these children. In 1979 children with Down's syndrome were not
thought capable of learning to read at all by most professionals1 and there was very little
research into the reasons for their spoken language difficulties, even though these were well
documented at that time2. Children with Down's syndrome were expected to be late to talk,
moving slowly from single words to two word utterances by five or six years of age. Very few
progressed beyond "keyword" or telegraphic speech, even as adults, rarely using more than the simplest early grammar that typically developing children are using by three to four years of
age. As Sarah had exceeded these expectations, we certainly thought that Leslie's observations
warranted further investigation.
With a grant from the National Down's Syndrome Association we were able to appoint a
teacher and set up a research study to begin to investigate these hypotheses. We followed the
developmental progress of fifteen children, of between two to four years, for three years while
they received a regular home-teaching programme from us.
Joanna, the first child that we tried teaching to read in 1980, learned thirty words in a month at
two years and six months of age. (We have this recorded on video-tape for the sceptical!) It
was immediately clear that Leslie's observations with Sarah might well apply to other children
with Down's syndrome and the Joseph Rowntree Foundation agreed to fund our work for a
further year, with the condition that we make a video and begin to disseminate our work to
teachers and parents, which we did.3,4
Insights gained from our first study
Reading progress and reading strategies
In the 1980 research project, we recorded the children's progress with their reading and while
initially we were most surprised by the speed and accuracy of their performance (as clearly
illustratrated on the video record of Joanna mentioned above), we were even more surprised by
some of the errors we began to see.
We expected the children to make visual errors i.e. to confuse words which look similar such
as hair and rain or this and shoe. These are the sorts of errors seen in the early performance of
all beginning readers taught in this "look and say" way and we found them in our children's
performance.
We did not anticipate the other consistent type of error that we observed when the children
were reading single words, the semantic error. Here the word the child says has the same
meaning as the one they are looking at but has no visual similarity. For example, the child
looks at the printed word shut and says closed or looks at harbour and says ship. These
semantic errors excited us at the time for two reasons.
Firstly, they suggested that the children were definitely decoding the print for meaning and not
just "barking at print" in a meaningless way as some of our critics were suggesting. They were
reading single words on flashcards so had no available clues to meaning such as might be
provided by pictures or the rest of a sentence. They must have decoded the printed word for
meaning and then thought of word which was linked by meaning to the target word on the
card.
Secondly, the errors demonstrated that the brain could go straight from print to meaning,
without changing the visual image of the word to its spoken form first and then accessing the
meaning. In 1982, reading theorists were still arguing about whether the developing brain
could actually do this as semantic errors had only been observed in the reading of brain
damaged adults described as "deep dyslexics".5,6 Our children were demonstrating that it
could and our videotapes caused quite a lot of excitement at the time. The first published examples of non-disabled children producing semantic errors when learning to read were reported in 1986.7
Signing
In 1982, signing was just beginning to be introduced to the younger children ill special schools
in the UK and some of the children we were working with learned sign at school, after we had
taught them to read. We noticed two interesting effects of learning sign.
Firstly, that some of our children were able to sign the correct responses to flashcards without
any extra teaching. We felt this was another clear illustration that they were not simply
"barking at print" but were reading the print for meaning and able to replace the spoken
response we had taught them with a sign entirely on their own initiative.
Secondly, that for some of the children signs were an easier and faster response mode than
speech. They would look at a flashcard and appear to be concentrating on producing the
correct spoken response, meanwhile their hands were already making a correct sign. This
additional time needed to produce speech suggested to us that the children might have some
sort of specific production delay with speech. Even when they knew what they wanted to say,
they had difficulty in saying it.
This was one of the fist clues we had to indicate that not all the language delay typical of
children with Down's syndrome could be blamed on general cognitive delay. In fact the way in
which the children substituted similar meaning words and signed when reading encouraged us to feel that they were more intelligent and had more understanding
of language that they were being given credit for.
Mime
When we watched the video-tapes on which we recorded the children's reading progress, we
noticed that we had captured a number of sequences in which a child was trying to describe an
experience or explain something to his mother for which his spoken language skills were
inadequate. The child would resort to the combined use of single keywords and mime to try to
convey his message. We were convinced again that the children knew what they wanted to say
but could not express themselves in speech.
There are a number of possible hypotheses which could be generated to explain these
observations. For example, the child may not have mastered enough vocabulary or grammar to
be able to construct the sentences that were being implied by the combination of words and
gesture. Alternatively, we could postulate that the child was thinking in sentences but could
not execute them in speech.
Either way, we were fairly confident, from watching these sequences, that the children's
understanding of their world was more advanced than their expressive language development
indicated. What was beginning to emerge was the possibility that the children's language development was being delayed by a series of specific language learning difficulties in addition to the effects of any more general learning difficulties.
Can reading enhance speech?
Our own observations of our early readers and those of many other parents and teachers8,9 who followed our methods suggested that Leslie Duffen's original hypothesis was right and
that reading is a "way-in" to their first language for these reading children with Down's
syndrome. We collected case studies of which those of Daniel, Louise and Alistair are typical.
Daniel's progress recorded by his home-teacher illustrates the effect of the early reading on his
speech development. Daniel's first spoken words were Daddy, teddy at fourteen months and he
was using some fifty single words at two years three months. Daniel was introduced to reading
at two years and four months when he had a production vocabulary of about fifty single words.
He learned to read ten words in two months and these were chosen to build two word phrases
for him to read at two years eight months. These were rapidly transferred to his speech,
according to the observations recorded by his mother and his teacher8. By two years eleven
months Daniel was reading six three-word phrases, which again he rapidly began to use in his
speech. At three years four months he could read sixty-six flashcards and many two and three
word combinations of these. The next month he was reading simple books and the words 'and'
and 'a' appeared in his speech. At three years six months he was reading four word sentences
and at three years eight months he could read one hundred and six words and was speaking in
six word sentences. Daniel's rapid progress continued and, at eight years of age, his reading
age was twelve years four months and spelling age nine years nine months on the school
assessment.10 At nine years and six months of age Daniel wrote this following a school
outing:-
"When we arrived at the villa it was more like a swimming pool to me but as I got closer I saw it was a villa. I thought it was very good at first. When we got in there was a wooden building with staff in it and I thought what is this building doing in the middle of the villa. Then next to it was the ruins of a Roman villa.. I wanted to explore so we split up in our groups I was in Mrs Blacks group. First we worked on the arch way then we moved on to the cellar then my favourite part the dining room. Then we met Stephen he was just about to go upstairs then Mrs Gilbert said if I was afraid of heights then Katie and Hannah could go upstairs with them. I did a wonderful picture of a Roman farm. Then we had lunch and then me and Stephen went to choose a postcard but it was to busy and Stephen had to go and finish his pictures. I bought a postcard of a Roman Castle I loved it very much so I paid for it. Then it was time to go back to school The other coach went off without us but we followed it and got back to school and we found a short cut."Daniel finds handwriting difficult and is still printing. He prefers to use a word-processor for writing and wrote the above story on his computer. He reads punctuation correctly but leaves much of it out when writing. He is being educated n a mainstream school and is average in his class for all subjects. His mastery of grammar exceeds that predicted to be achievable by persons with Down's syndrome even in 1990.11
Christmas storyWhile case studies such as these are valuable, they do not provide definitive evidence of the effect of reading on speech development as these children are at a stage when improvement in speech and language would be expected and, perhaps these particular children were going to be the "brighter" ones anyway. In the mid-eighties, it was not possible to establish a longitudinal study of a group of children with Down's syndrome large enough to represent the typical range of learning difficulty, in order to study the effect of literacy teaching, as most of the special schools that the children went to did not teach them to read. We had to wait until 1993 to set up such a study11, and I will describe that study later. Meanwhile, we tried other ways of collecting the evidence.I am the shepherd one very cold night I was looking after my sheep. My fiends told me about a special baby that had just been born. Some of the shepherds were fast asleep. I woke them up and told them about the baby. A angel said you must go to Bethlehem and find the baby. He will be wrapped in a warm cloth and lying in a manger. I found the baby Jesus and it made me feel very happy.10
Teaching grammar to teenagers
The study was designed to improve the comprehension and production of grammar for a group
of twelve teenagers. Teaching which used print to support the learning was more effective over
six different sentence structures than speech and picture only teaching. All the teenagers did
better in the reading condition, but there were large individual differences. The teenagers who
gained the most were those with little or no reading ability at the start of the study and
considered to be the "least able" by the schools. At the end of the training year, the teenagers
demonstrated a significant gain in comprehension of grammar compared to a previous baseline
year of no intervention beyond ordinary school practice and a significant increase in the length
of the utterances that they used in everyday conversation. Individual differences in the size of
the gains were linked to differences in auditory short term memory function and in speech production difficulties.12,13
As I was teaching the teenagers, I realised that some had such limited auditory short-term
memory spans that they could not copy a six or seven word sentence if they only heard it
spoken, even though they could produce the "keyword" version of tile sentence from a picture
prompt without a model. For example one teenager, Rebecca, could look at the picture of a
boy sitting on a chair and say "He sit chair", but she could not repeat "He is sitting on the
chair" immediately after I had said it while we still looked at the picture. She, like several other
teenagers in the group had digit spans of only two (digit span is the measure of how many
digits a person can recall immediately after hearing them spoken at the rate of one per
second).14 Rebecca was fifteen years old and a non-disabled teenager of her age would have a
digit-span. of six to seven digits. My group of teenagers with Down's syndrome had spans of
two to four digits. Work published by Charles Hulme and Susie Mackenzie of York University
in 1987 confirmed that the auditory short-term memory spans of my teenagers were typical of
teenager's with Down's syndrome.15
My training study had demonstrated that reading could be used to teach grammar and improve
the conversational skills of teenagers with Down's syndrome. It has also drawn our attention
to poor auditory short-term memory as another possible reason for their difficulty with learning
grammar in the way most children do, that is by simply listening to the language around them.
Memory studies
1n 1990, with one of our colleagues in the Psychology Department, Dr. John McDonald, we
set up a memory training study with a graduate student, Irene Broadley, and investigated the
effectiveness of two types of memory training programmes.16,17,18 Both programmes proved
effective in increasing the children's short-term memory spans and the gains were sustained for
the first year after training when compared with an untrained control group but did not last
over a three year period for all the children.19 In addition, the children who had increased their
memory spans also showed a significant gain in their understanding of grammar, supporting
our view that limited memory span would restrict access to grammar. However, we could not
be certain that the language gains were due only to the memory gains as those who had made
most progress were in mainstream schools and learning to read.
When Dr. Glynis Laws assessed a subset of fourteen children from this study in 1995, she
found that we had some longitudinal evidence of the beneficial effect of reading on the
children's cognitive development. Those children who had learned to read since they were first
assessed for the memory study in 1991 had significantly better language and memory scores in
1995 than those who had not learned to read, even though there had been no significant
differences between the groups on these measures in 199l. Unfortunately, Glynis noted that all
but one of the readers were in main stream schools, while the non-readers were in special
schools so the positive effect of being in a more normal language environment in the
mainstream schools could not be ruled out. She then looked at the cognitive scores of readers
and non-readers in the special schools at the start of the study in 1991 and the readers had
exactly the same advantage on their language and memory measures, despite being in poor
language environments, so it looks as though reading is the important variable in the
longitudinal study20. The results of a longitudinal study of mainstreamed children which we set
up in 1993 with another graduate student, Angela Byrne, will give us better data on this issue.
Current studies
Angela is following the progress of twenty-four children with Down's syndrome and
comparing their progress with a group of their mainstream classmates who are matched with
them on reading age, as well as classmates who arc average readers for their age. Since 1990,
the majority of children with Down's syndrome in our local area have been able to attend their
local mainstream schools with special needs support so we are now able to study
representative groups of children all in the same social environment. This mainstreaming
programme has been largely due to the work of my colleague Gillian Bird1O. Gillian has been
working with me since 1983, contributing to the research programme and providing the
professional support to the schools. Without her practical work in the schools and with
parents, we would not have this group of readers with Down's syndrome to study.
Angela's study is charting the reading, writing and spelling progress of all the children, looking
at the cognitive strategies they are using to read and the links between reading, language and
memory skills. The children's ages ranged from five to twelve years at the start of the data
collection in 1994. All the children with Down's syndrome are learning to read and their
reading ages ranged from five years to eight years five months. The typically developing
children identified by their teachers as average readers for their age demonstrate age-
appropriate cognitive profiles over all the measures, whereas the slower readers for their age in
the same classes turn out to be significantly delayed relative to the average readers on all these
measures. The children with Down's syndrome, while matched with the slower readers on the
reading measures, are significantly behind them on the number, language and memory
measures. In other words, the children with Down's syndrome show advanced reading ability
compared to all their other cognitive skills at this time.
One year later, the children with Down's syndrome do not differ significantly from the slow
readers in reading progress, in other words both groups have made similar progress. However,
Angela has been assessing their reading strategies. Research on children's reading development
which indicates that all children go from a logographic stage (when words are recognised by
'sight' only) to an alphabetic stage (when words can be 'sounded out' letter by letter) and then
to an orthographic stage.12 At the end of the first year of her study Angela has shown that the
children with Down's syndrome are not progressing to an alphabetic strategy as fast as the
slow readers despite keeping up with them on reading performance. They are still relying on
logographic or visual strategies.
To be good at alphabetic or phonological recoding, a child must be able to hear all the sounds
in the words and given the hearing loss and auditory processing problems that children with
Down's syndrome often have22,23 they are likely to be less able to use this route with ease than
their typically developing peers. Our case studies to date suggest that despite these very real
additional difficulties experienced by the child with Down syndrome compared to typically
developing children, they do slowly progress to being able to use phonological recoding for
reading and spelling. Research on the links between typically developing children's reading
progress and other aspects of cognitive development suggest reciprocal interactions. The more
language knowledge and the better the phonological awareness and working memory skills
children bring to the task of reading, the faster they will learn to read in the first year of reading
instruction. In the second year, reading success appears to develop working memory and
phonological awareness skills.24,25 Being able to read opens up access to knowledge and the
biggest vocabulary explosion for children is between the ages of about seven and sixteen, when
children are typically learning on average three thousand words every year.26 Reading and
writing also teach children correct grammar.27 So, the benefits of reading that we are seeing for
our children with Down's Syndrome are those that would be predicted from research with non-disabled children.
Children with other disabilities
We believe that other children with language learning difficulties might also benefit from
reading instruction in the way we have demonstrated for children with Down's syndrome. With
another graduate student, Michele Appleton, we have begun to study the influence of early
reading on the progress of deaf pre-school children, comparing their progress with twenty
children with Down's syndrome and twenty non-disabled children. After one year, the children
with Down's syndrome and the non-disabled children are progressing at the same rate, some
children in each group have established sight vocabularies of sixty to seventy words and some
only a few words, with no significant differences between them. However the deaf children are
finding the task more difficult, with little success so far, though we would not wish to draw any
conclusions as to the reasons for this at present. It is well-known that severely and profoundly
deaf children have great difficulty with literacy and that less than half leave school able to read,
so that this is an important area for future research.28,29,30
We plan to extend our work to autistic children in the next year, as studies in America suggest
that they also can be taught language through reading but that they learn language differently
from most children, by learning language in chunks via echolalia, and make more and different
use of visual imagery.31,32
In conclusion
We now know that the speech and language delay of children with Down's syndrome is not
just due to some global mental impairment as was assumed in 1980. Put another way, it was
thought that their language reflected their ability to understand the world and that little
improvement could be expected as their ability was fixed by their genes1. We would argue that
the reverse is true; that children's understanding of the world is heavily dependent on their
ability to learn their native language. Every word a child learns draws his or her attention to
some aspect of the world or the people in it. The size of a child's vocabulary reflects the extent
of their knowledge and in addition, language becomes a tool for thinking, reasoning and
remembering. Any child with a language delay will therefore be at risk for delay in all areas of
cognitive or mental development.
I would like to end with a parent's view as this is how the whole project began.
"I started to teach Emma to read after hearing you talk in Bristol seven years ago. She was then two years and four months of age. Emma is now nine years old and an able and avid reader. She attends our large local mainstream primary school and holds her own well in the second year junior class. She seems to develop in leaps and bounds. Being able to read has done so much for her.33
It helped her speech. For example when she began to read at age two, she spoke understandably but imperfectly as she left out the definite and indefinite articles, prepositions etc. The change came when she was able to sentence build in flashcards. Today her speech is mature and her teacher commented at the last parents evening that the extent of her vocabulary and her turn of phrase would leave many in the class standing.
It helped in the way other children regarded Emma and not least her own self-esteem. They knew and she knew that in reading she was amongst the best in the class. This apparently less able child wasn't so less able after all!
Emma is now an independent reader and books give her so much. She wakes early and reads for at least one hour every morning. She makes her own choice of book but everything she reads fulfils her - she chuckles when reading 'The Twits' and cries over 'Heidi'. These are her two favourite books at the moment and she reads them over and over again. Equally however she will read poems and her atlas, history book, nature book etc. from which she teaches herself. She loves her Bible. She is very proud when her five year old sister carries the newspaper to her and asks "What time is ......... on the television?" She is always able to tell her and I feel Sarah, who I feel senses rather then knows Emma's differences, is thrilled with the sense of her big sister having the 'big sister' image for once."
Revised: August 11, 2006. |