Department of Psychology
Göteborg University, Sweden
(SLIDE) Together, my colleague Tomas Tjus and I have conducted two studies in which we have explored the possibilities of using various interactive multimedia procedures for teaching reading and communication skills to children with autism. These studies have been carried out in Sweden as a joint effort with several other colleagues both in Sweden and USA. I would especially like to mention Professor Keith Nelson in the US and Prof. Christopher Gillberg in Sweden.
We have seen quite a few different attempts and reports concerning the effectiveness of computers as tools for learning in children with autism over the last 20 years: Researchers have tried to use the computer as a remedial tool with mixed results, teachers have used the computer in the classroom with the aim to promote learning, and some parents have hoped that the computer would help their child to cope with the handicap.
There is, however, no room in this brief exposé to review all the attempts that have been carried out. We only wish to emphasize that the field is still young and that very few systematic studies have been carried out. Most of those studies tend to report positive effects although mostly of a modest strength. However, several reports have also failed to find support for any specific gains relating to computer aided instruction, and there is - to our knowledge- not a single scientific report study supporting any extensive change in the level of autism due to the effectiveness of the computer.
Computers do not cure autism and computers are not used outside a social context. People use computers and children with autism are people that might gain from using computers together with important others!
It is interesting to note that the earliest report we have found in the scientific literature is a paper published already in 1973 by Kenneth Colby. He used a program with the aim to stimulate language development in children with autism and the strategy employed was not at all very different from the one we have explored in our studies of more recent origin. Colby's program was constructed in such a way that it allowed the child to press a letter on the computer and simultaneously hear the computer say the letter. Or, the child could press the letter "H" whereas the computer displayed a horse that trotted across the screen together with sound from the horse's hoofs. The aim of this early multimedia attempt was to mimic normal spontaneous language acquisition and free exploration was encouraged. Colby's idea was to allow the child to explore the program with a "minimal interference by adults". Over all his results were positive: 13 out of 17 mute children with autism showed positive gains.
(SLIDE) The programs we have used - and in part developed - allows the child to explore language and to create sentences in a playful mode: The child writes sentences by combining nouns and verbs and receive fast one-to-one feedback through multiple
channels: text, spoken language, graphic animation, and sign language. In this example (SLIDE), a child writes the sentence "The buzzard chases the carrot". The child then sees the written sentence in text, hears the sentence being read, and sees it (SLIDE) as an animation. This gives the child a unique freedom to play around with language and at the same time develop a dialogue with the teacher A possibility not commonly provided for in educational programs for children with autism.
We have used the computer in order to facilitate language growth among children with autism that had already showed interest in other people and some motivation for communication. All children were between 5 and 15 years old and they always worked together with his or her teacher. IQ varied from mild mental retardation to well above average.
The children worked with our programs twice a week for two to four months. The role of the teacher was to be an active partner/supervisor and to promote a dialogue based on what the child created with the computer. The goals of instruction were to improve overall communication skills and motivation, not just to learn specific material embedded in the software presentations. Therefore, we instructed the teachers to use conversation strategies and timing of social interaction in ways that supported the child in enjoying and understanding the sentences and the graphics presented through the computer software.
We tried to measure each child's reading development both before, during, and after our computer program had been introduced to the child. We then transformed our measurements into a global reading score indicating each child's average gain in reading per month.
And what did we find?
Let's start with our first study: Here we found clear evidence
of an increase in reading skills. (SLIDE) This increase also seemed
to be specific to the intervention. The children increased their
reading with eleven per cent during the period in which our program
was used as part of the curriculum, but only with five per cent
when the intervention was over.
Recently, we confirmed these observations in our second study. A total of 53 children participated of which 13 belonged to group "children with autism". The rest of the group was made up of 16 children with learning disorders (e.g. dyslexia, hyperactivity and attention problems), 13 were deaf or had severe hearing difficulties, and eleven children were multiply handicapped. To date, reading data for 46 of the children has been analyzed (SLIDE). The result as shown in this slide clearly indicate that - overall - the children speeded up their language learning as a consequence of our intervention. More specifically, 37 out of 46 children displayed a pattern indicating a positive effect.
(SLIDE) The pattern is similar if we choose to describe only the 13 children with autism which is shown in this slide. They displayed no gain at all in reading during the initial baseline period but a clear increase during the intervention period. Stated individually, 10 out of the 13 children in the group displayed a positive effect.
Gains in reading was, however, not the only findings of interest (SLIDE). We also noted a positive effect on the social and verbal interaction between teacher and student. The children with autism talked and discussed more with the teacher as an effect of the intervention and they were also more prone to express positive feelings at the end of the intervention period than early on. Thus, they seemed to enjoy working with the program together with their teacher.
Finally, we also investigated different background factors. We wanted to know more about why some children did show strong progress while others didn't gain anything from our intervention. We found four factors that had a significant impact on how a child learnt language through our program (SLIDE). The first two did not surprise us. Both the child's language age and the child's mental age did contribute significantly. But they were not the most important factors that we found. Much stronger influence was exerted by the teacher's attitude and the child's motivation! From the analysis of our videotapes, we could conclude that the amount of enjoyment displayed by the teacher had a strong impact on the child's actual learning. Furthermore, we also found that the child's proness to participate in a simple language game was a very good predictor of the child's development.
To us, these findings indicate that traditional tests of language or IQ alone or together are insufficient in selecting children for training based on our strategy. Instead, we strongly recommend that information about the child's motivation as well as the teacher's attitude towards computers are included in the decision process. Otherwise we increase the likelihood for failures and negative effects! Our hopes must be realistic and the intervention well planned!
(SLIDE) In short, we believe that our findings strongly suggest that a highly motivating multimedia environment indeed enhances reading skills for many - but not all ! - children with autism. This might be especially true if the computer program is employed in a setting where the teacher promotes a warm and supporting atmosphere and uses recasts, questions and elaborations that ties in with the child's activity on the computer and thus promotes a learning environment that maximizes the likelihood for the child to learn new language structures.
BUT! (SLIDE)
Remember that computers are not equal to magicians!