Teaching Graphic Symbol Combinations to Children with Limited Speech During Shared Story Reading (summary)

young children


The use of graphic symbols can be helpful for children who do not develop enough spoken language to meet their communication needs but it is not always easy to combine symbols into messages of more than one word, particularly for those children whose understanding of spoken language is also delayed. There could be a number of reasons for this.

The skills needed to develop the use of symbol combinations do not seem to develop without direct teaching. Some studies have shown that the use of adult modelling of word combinations using the symbol format (aided input) helps promote use of symbol combinations by children who use augmentative and alternative communication (AAC). Modelling combined with direct teaching strategies might be even more effective.

What did they do?

The authors aimed to extend research into interventions that might help develop expressive use of symbol combinations. They looked at three types of language structures that are typically used in early language development.

They used a matrix training technique which is fully explained in the paper. It aimed to support participants to generate new combinations of words that have not been specifically taught, but the elements of which have been taught in different combinations, e.g. the combinations 'the dog cries' and 'the boy falls' were taught with the aim of generalising these to 'the boy cries' or 'the dog falls'.

Instead of teaching the structures out of context the researchers built them in to a story reading situation; looking at whether targeting two-symbol combinations using a hierarchy of prompts during story reading could facilitate the production of symbol combinations in response other to picture stimuli.

Four children who fulfilled strict selection criteria were selected to participate. A baseline assessment of their ability to combine symbols was completed. Five symbol combinations for each of three different semantic relations, agent- action e.g. 'the boy runs', possessor-possession e.g. 'the bunny's shoe' and attribute-entity e.g. 'broken car', were selected as intervention targets. A further five for each set of relations were the generalisation targets. Each child was given a colour-coded symbol board with the 21 symbols needed to generate the target combinations.

A story was developed to teach expressive use of each of the target sets. The intervention involved the researcher reading the stories to the children individually and using a five point hierarchy of prompting strategies to support the participant to use the target combinations as they occurred in the story. Repetition of the original assessment took place at regular intervals throughout the intervention period and between one and five weeks after the intervention ended.

What did they find?

Two of the participants showed an improvement in their ability to combine symbols which was largely maintained after the end of the intervention. These were the children with higher levels of receptive language ability and English as their home language and responded well to the first level of the prompting hierarchy.

The other two participants showed less clear and consistent improvement in the assessment task, though they did show improved ability to combine symbols within the story telling activity. These children needed more supports from higher up the hierarchy of prompts.


A number of factors that might have accounted for the variation in performance between the participants are discussed. These included levels of receptive language skills, home language and the participant's initial ability to identify the symbols used. Previous exposure to expressive use of symbols might also be significant.

The authors note that the results might have been poorer than expected because the assessment task was a 'test' in which no feedback on the correctness of responses was given. This might not have been particularly motivating, there is some evidence that this might be the case as the less successful participants performed better in the story reading intervention phase than in the testing element.


There are many issues identified as possible problems with the project: the responses of the weaker participants in the intervention tasks are not reflected in the assessment results, thus the gain in skills might not be fully represented in the results, there was a wide range of receptive language levels across the participants and two did not hear English as their home language which could account for the variability in the results. A more similar group of participants would have been more successful in isolating the effects of the intervention.

The limited number of teaching sessions offered might have been too few, leading to a lack of maintenance of the newly acquired skills after the intervention. No long term follow-up was undertaken.

Things you may want to look into:

The effect of aided language stimulation on vocabulary acquisition in children with little or no functional speech

Shared storybook reading with a student who uses alternative and augmentative communication: A description of scaffolding practices

The effect of shared book reading on the acquisition of expressive vocabulary of a 7 year old who uses AAC

The Effect of Aided AAC Modeling on the Expression of Multi-Symbol Messages by Preschoolers who use AAC

The effect of shared book reading on the acquisition of expressive vocabulary of a 7 year old who uses AAC

Added to site August 2015