Implications for interventions to support speech and language development

Building on strengths - increase daily communication experience

Children with Down syndrome are keen to be social and their interactive communication skills (ability to understand and participate in conversations) are good. This strength should be recognised and every effort made to enable them to communicate in all the settings that they are in at home and at school. They should be encouraged to continue to use gesture and sign to enable them to be effective and proactive communicators until their speech can be better understood. Everyone around them needs to be sensitive, to take time to listen and to include them. Language is learned and practised because children want to communicate.

Increasing the quantity and the quality of everyday communication experience for children and adults with Down syndrome is probably the single most important intervention and everyone can help to do this without special training, at home and at school. One implication of this is the importance of full inclusion in education and in community activities. Children need communication partners and to be immersed in normal conversational worlds. Special classes which put together children with severe language delays are not an appropriate daily environment for any child, and adults cannot compensate for the lack of communication between the children. Indeed their own communication to each other and to the children will be distorted by the children’s difficulties.

Targeting weaknesses - specific additional interventions

Comprehension and production of vocabulary needs to be taught through games and specific activities to help children acquire a 300 word production vocabulary before 5 years of age.

Articulation and phonology should be targets for intervention from the first year of life right through to adult life. Children only try a word if they can at least attempt some of the sounds in the word, therefore early work on phonology should speed up vocabulary learning. Since the emergence of grammar is linked with vocabulary size this should lead to earlier acquisition of grammar.

Grammar needs to be taught and reading may be the most effective way to support this, capitalising on the children’s strengths as visual learners and compensating for the small verbal memory span.

Is intervention for speech and language effective?

While the reader may agree that the proposals for intervention set out in the last section follow logically from the available evidence from research, the profile of language development of children with Down syndrome and the probable causes, is there actually any evidence that speech and language therapy or any other interventions will make a difference?

Unfortunately, the number of studies that have actually evaluated the effectiveness of interventions are few, despite the many papers describing the language difficulties of children with Down syndrome . A number of studies have reported on the progress of small numbers of children and usually report that language games and activities do result in faster progress but few of these studies have sufficient numbers of children or comparison groups to provide convincing evidence of effectiveness (see Spiker and Hopmann^[1] for a recent review to 1997).

There have been some reported intervention studies, not covered in the 1997 review, in which the effects of the intervention have been adequately measured. These are discussed below in some detail as they illustrate many important principles for interventions. They also address issues often queried by speech and language therapists, such as the benefits of naturalistic, language interaction intervention compared to direct teaching approaches, and the benefits of direct work on phonology.

Communication

Communication is usually a strength for most children and young people with Down syndrome. In their early years they usually want to communicate, making eye contact, smiling, learning to take turns and developing joint attention only a little later than other children. Some children do have early communication difficulties if they are more disabled or on the autistic spectrum and we discuss these issues further in the communication chapter. While they usually show strengths in early communication and interaction, young people with Down syndrome may need some more support in school, teenage and young adult years to develop a wider range of social communication abilities and we will also discuss these in the communication chapter.

Early years: The communication intervention studies have focussed on early years as, even though non-verbal communication skills are developing in line with nonverbal abilities they will be delayed for age and even though they are developing, children may not use them as often as other children. These studies usually involve training parents, aiming to draw parent’s attention to the importance of this stage and encourage parents and carers to increase their responsiveness to their children’s communicative signals.

We should remind ourselves that while most of the research has focussed on young children, nonverbal communication skills, including making eye contact, engaging in joint attention, pointing, showing, turn taking, continue to be important through life.

There are a number of well-developed programmes for this stage of development, including the Hanen programmes, Prelinguistic Milieu Teaching and Responsive Teaching. They have all been used with children with Down syndrome. They all work to increase early non-verbal communication and increase adult responsiveness in naturalistic play sessions with children. This should encourage children to want to communicate as they get responses and it should also increase their opportunities to learn words.

The Hanen programmes developed in Canada are now widely used by speech and language therapists in their work with parents and children. Hanen also provide some excellent materials and information which parents can access directly. (refs) We will see that the Hanen materials have been used for parent training in some of research studies we are about to review.

The research groups of Ann Kaiser, Marc Fey, Paul Yoder, Steve Warren and colleagues have been developing and evaluating Prelinguistic Milieu Teaching (PMT) over many years with children with intellectual disability. (refs) Some of the studies have included children with Down syndrome but they do not always report the outcomes for them separately. We will describe PMT below in the first vocabulary intervention as the intervention started with PMT.

The Responsive Teaching approach (RT), developed by John McDonald and Gerald Mahoney, like PMT, encourages intervention in naturalistic play situations but with an important difference. Rather than targeting specific outcomes such teaching children to build a tower of bricks, post shapes, learn a new word or complete a puzzle, RT aims to develop children’s ‘pivotal behaviours’ which will equip them to be more successful and independent learners. These are behaviours include encouraging curiosity and exploration, sharing joint attention, imitating, initiating communication, maintaining attention, persistence and cooperation. There is also a focus on training parents in responsive rather than directive strategies when playing with and teaching their children.

Examples of key responsive parent strategies are: contingency - respond immediately to little behaviours, reciprocity - take one turn and wait, affect - interact for fun, match - do what my child can do, nondirective - following the child’s lead. Several articles have reported RT studies for children with autism and developmental delays which have included some children with Down syndrome and suggest benefits. (refs DSRP articles) A 2013 study in Turkey reported a significant benefit on a general measure of language development and overall developmental quotient for children with Down syndrome aged 2 to 6 years after a six month intervention with mothers based on Responsive Teaching.

We should also note that the more recent JASPER approach developed and evaluated for early intervention with autistic children by Connie Kasari and her team at UCLA in California is very similar. The initials stand for joint attention, symbolic play, engagement and regulation. We discuss autism and interventions for autism in the chapter on communication but note here that there is evidence that this type of intervention, including a study in the UK called PACT does improve outcomes for children on the autistic spectrum. This suggests that we can take a preventative view and there are benefits of teaching all parents how to understand early play and communication as it may reduce difficulties which only emerge later.

A review of studies that have provided training in early communication for parents of children from 1-5 years with developmental delay has been published (2017) by a research group led by Danielle Te Kaat-van den Os in the Netherlands. They reviewed 7 studies and all included children with Down syndrome as well as other children with delays but do not report their findings for the children with Down syndrome separately. They conclude that there is evidence that parents do become more effective communicators with their children and that the children increase their communicative interactions but that there is no clear evidence that these approaches lead to faster expressive vocabulary development. They suggest that a more directly linguistic approach may be needed. (explain) We may be able to provide some suggestions for this after reviewing the vocabulary intervention studies with children in this age range in the next section.

We will describe the practical approaches and activities used in the Hanen, PMT, RT and JASPER approaches in the practical chapter on communication.

Many of the early intervention studies which teach parents to be more responsive and aim to increase communicative exchanges have involved children who are already saying their first words so include a focus on vocabulary teaching as well.

Teaching word production

Teaching vocabulary

Teaching vocabulary should be an objective from infancy through to adult life but most attention has been focused on early vocabulary and getting children started on talking. Here we will summarise research studies toddler, preschool and primary years. We may draw some principles for teaching vocabulary to older teens and young adults.

Early and preschool years; As children are learning their first words there is an overlap with the prelinguistic, non-verbal communication strategies as every time a child looks at something, points or shows an object the adult is likely to talk about what they are looking at and use the words for the objects and actions. Milieu Teaching/enhanced Milieu Teaching (MT) follows on from Prelinguistic Milieu Teaching and focusses on teaching words and then combining words – again in naturalistic play sessions but also using behavioural approaches and prompts. Milieu Communication Teaching (MCT) includes both PMT and MT and has been developed and evaluated by Yoder, Warren, Kaiser, Fey and colleagues over many years.

A milestone study in early vocabulary intervention

In 2014 Yoder, Woynaroski, Fey and Warren published what we consider to be a milestone study because of what we can learn from it. They not only compared the effectiveness of delivering intervention sessions for 1 hour 5 times each week with delivering 1 hour sessions once a week for an hour, they also analysed the processes which were leading to progress.

They had 64 children with intellectual disabilities in the study of whom 35 had Down syndrome. At the start the children were between 18 and 27 months and had expressive vocabularies of no more than 20 words or signs. To be included they also had cognitive quotients between 55 and 75 (an infant ‘IQ’), normal hearing in at least on ear, normal or corrected vision and no symptoms of autism.

The children in the study were divided into two groups and one group received the intervention once a week, the other 5 times a week for a period of 9 months. Sixteen of the children with Down syndrome were in the ‘once a week’ group and 19 in the ‘5 times a week’ group. There were no significant differences between the development of the children in the two groups at the start of the study.

The intervention was delivered by trained paraprofessionals supervised by a speech language pathologist and the children’s parents also received some responsivity training during the first 3 months of the study which we discuss later.

The study used the McArthur-Bates Communicative Development Inventories (CDI) to measure their expressive vocabulary. This is a widely used measure based on research which recorded the typical range of first words that children say in English and has been used in many research studies since it was first published in 1988. The first studies were in the USA, but the CDI is widely used and has been translated into a number of languages. You will see in this chapter and the vocabulary chapter x that it has been used in a number of studies with children with Down syndrome in different countries and languages to both measure progress and to provide norms for vocabulary development.

Effect on developing spoken vocabulary

We will focus on the findings for the children with Down syndrome as they have published these separately. The children who received 5 sessions a week had significantly more spoken words in their vocabulary at the end of the 9 months (mean 17.3 words at end of intervention compared to 5.4 words). This may not seem like a big difference but, as the authors of the study point out, it is 3 times as many words and meaningful gain if judged by effect size (g=0.5). We also learn that developing spoken vocabulary is a considerable challenge for toddlers with Down syndrome. The other children in the study with intellectual disabilities but not Down syndrome made faster progress and had larger vocabularies at the end of the intervention. The team suggest this difference may be due to the verbal short-term memory delays and the speech production challenges in learning and saying speech sounds that are experienced by children with Down syndrome.

Individual differences

Children’s progress was related to their cognitive development and the team suggest that their cognitive quotients ‘may indicate learning capacity’ p28.

They also showed that children who engaged in more functional play with more objects at the start of the study produced more spoken words at end. This highlights a number of issues. First, that play matters and we need to consider developing children’s play which may be delayed by their motor delays. Secondly, when children play with objects, adults will talk about their play, the objects and what they are doing with them. Their play may increase opportunities for ‘linguistic mapping’. Thirdly, when playing children show their growing understanding of the world around them. They progress from using toy objects like the real versions to being able to pretend using an object to represent something else like bricks making a train. A child’s functional play may also indicate that they have reached a level of ‘symbolic’ understanding that will support language learning as words are symbols – they represent something.

The intervention

The intervention sessions were carefully planned and tailored to each child. Children who produced fewer than 5 spoken words on the CDI or during an initial observation session received Prelinguistic Milieu Teaching (PMT) designed to increase frequency, clarity and complexity of child’s non-verbal communication acts including joint attention, gestures, and vocalisations.

When using 5 or more words the children moved into MT and language became the targeted communicative behaviour. The goal of MT is to establish and/or increase the frequency and/or complexity of child’s verbal communicative acts. Now the focus is on teaching them to talk.

If child spontaneously used fewer than 40 words or signs, 5 to 10 words were selected by parent and clinician as vocabulary targets to teach. If less supportive prompts were not effective, clinician could use a directive prompt such as ‘say eat’. When child spontaneously used more than 40 words or signs, joining words together became the goal. Each MT teaching episode ended with a ‘phonemic or grammatical recast of child’s utterance’ – this means the therapist repeated a word said by the child to provide a clear model of how to say it or repeated a 2 or 3 keyword utterance filling in any missing grammar.

We will explain all these intervention strategies in more detail in the practical chapters.

The important point to note is that the intervention was delivered in the child’s home in a play setting and while the interventionist set up particular play activities to teach specific words, they took a responsive approach and followed the child’s lead particularly if the child had less than 5 spoken words. However once words were the target they prompted words and asked for words in a more directive manner. The teaching was quite intensive with the target of providing one teaching episode per minute.

Training interventionists and parents

The paraprofessionals who delivered the intervention sessions all had expertise in special needs and then completed minimum of 20 hours in–project training. They were supervised twice monthly by SLPs. The children’s parents received responsivity education (RE). They were given the Hanen Parent Program book ‘It takes two to talk’ plus one hour individual RE sessions for 9 weeks in their homes within first 3 months of study. You will see that a number of the studies we review in this chapter involved parent training and we will discuss this further when we summarise what we can learn from research.

The processes of change – the mechanisms

In a further paper the research team provide information on the processes which led to new word learning for the children with Down syndrome. They showed that after 3 months of intervention the children receiving 5 sessions each week were using significantly more vocal communications which they defined as canonical syllable communication (CSC). Canonical syllables are vocalisations with at least one consonant and vowel combination produced with adult-like speech timing. The amount of communicative vocalisations that a child produced after 3 months of intervention was related to the number of spoken words they had learned at the end of the 9 month intervention. The more they vocalised, the more words they mastered.

The children receiving 5 sessions each week also showed a greater growth in receptive vocabulary – the number of different words that they understood. The average increase in words understood after 6 months of intervention was 86 words for the children receiving 5 sessions each week and 46 words for the children receiving one session each week.

Both the increase in communicative vocalising and in receptive grammar mediate the increase in spoken words at the end of 9 months for the children receiving the more intensive therapy. This means that encouraging children to use their voices to communicate was an important step towards saying new words. It also means that teaching children to understand words, even though they cannot yet sign or say them, is an important goal.

In addition, the gains in receptive vocabulary correlated with the amount of vocalising which means that when one increased so did the other. In another analysis of the data for the whole group of 64 children including those with Down syndrome, the team (Woynaroski et al 2014) show that an increase in communicative vocalising at 3 months into the intervention produced an increase in parent ‘linguistic mapping’ at the 6 month point. In other words, when children use their voices to communicate, parents tend to respond and talk about what they are looking at or doing. Further, the amount of ‘linguistic mapping’ parents provided at the 6 month point predicted spoken words at 9 months. This means that the more ‘linguistic mapping’ a parent provided, the more spoken words their child had at the end of the 9 month intervention.

We can see the dynamic and interactive nature of development in these results. Parents are influenced by their child’s behaviour and children’s progress is influenced by their parent’s behaviour.

These findings indicate the importance of parent training – the results were not just produced by the interventionists and the team conclude that it is important to target both child communication and parent responsivity.

The authors note that the children with Down syndrome did not learn as many new words as the children with intellectual and developmental delays but not Down syndrome. The other children did not show the same benefit of 5 sessions a week compared to one session a week. xx

The authors point out that the progress of children with Down syndrome is being influenced by speech motor issues, described by Jon Miller in 1999 as difficulty with representing, planning or executing motor programs to produce the speech sounds that compose words. They suggest that ‘more MCT therapy may aid practice and feedback in a way that enables sufficient vocal control and accuracy to enhance spoken vocabulary despite possible motor dysfunction.’ 2015 p 312. The research team also suggest that children with Down syndrome need more models to learn new words because they may be slow to process phonemes. They discuss the part played by verbal short-term memory and suggest that more therapy means words offered with greater frequency and this may help children to learn them.

They also state that Our findings suggest that gains on measures of canonical syllabic communication and receptive vocabulary are appropriate intermediate goals when the overarching goal is for young children with DS to learn to communicate with spoken words. Children with DS are likely to exhibit significant progress on these measures before consistent gains in word production. P 312

In summary, what can we learn from this study

Points for practice:

Intensity of intervention; The first point is that intensity of interventions may make a difference. Daily practice in a session designed to encourage specific non-verbal communicative behaviours or to teach specific words will make a difference. Children with Down syndrome may need more repeated opportunities to learn to understand and say words than other children. In most countries, daily therapy sessions by trained therapists for 9 months are not going to be available. If this is what children need then we need to train parents and teachers – especially as an increase in parental linguistic mapping was shown to contribute to the gains made by the children.

Targets for intervention; A unique contribution that this research makes is to highlight the building blocks which led to increased spoken vocabulary.

The importance of encouraging speech sound production from the babble stage to combining sounds

The importance of developing receptive language – having targets for comprehension of words as well as for production

The importance of increasing ‘linguistic’ mapping – increasing the input/ vocabulary teaching children receive which enables them to learn the meanings of words and how to say them.

The importance of developing functional object play

Methods: this team used both naturalistic and behavioural approaches/strategies

The most recent research study that has evaluated the effectiveness of teaching specific target words to toddlers with Down syndrome is from the Canadian research team who developed the Hanen programme.^[2] In 1998 they reported on a study that involved twelve children with Down syndrome and their mothers. The children ranged from 29 to 46 months of age and they communicated using at least 10 signs and words but did not use any combinations. Six children were assigned to the experimental group and six to a comparison group initially. The comparison group continued to receive regular speech and language therapy but received the targeted word programme later. The therapy programme consisted of the Hanen programme provided over a 13 week period, with 3 modifications. During the second parent training session, parents were given a list of 20 target words and they chose 10 that they thought their child would be motivated to learn. If the child used a word three times spontaneously in three different contexts (as determined by parent diaries) the word was replaced by another word from those remaining. Parents were taught to set up new routines that permitted modelling the target words many times. They were also trained to use signs as they spoke the words. Parents kept diary records of their children’s imitations or spontaneous productions of words.

Parents learned to use the target words in daily situations and to repeat the word at least 3 times during each interaction. For example, if the target word was ‘baby’, the parent was instructed to join in the child’s ongoing doll play and model the word by using labels or short comments (e.g. ‘baby’, ‘baby is sleeping’). The child was not required to imitate the word.

The 20 words were selected on the basis of the child’s MacArthur Communicative Development record, using 4 criteria. They were words that the child already understood but were not producing expressively, they were all words that typically developing children say by 24 months, all the target words started with a phoneme that was already in the child’s speech-sound repertoire and all words represented objects that could be used functionally or words that could be demonstrated using gestures (e.g. more, all gone).

The children in the experimental group learned significantly more of the targeted words than the children in the control group. The authors comment that larger studies are needed but the conclusion is that targeted vocabulary work is needed in addition to training parents to be effective and sensitive communicators.

Vocabulary - Girolametto, Weitzman et al 1998

An early study by the Hanen team looked at teaching vocabulary to 12 children with Down syndrome aged from 29-46 months so slightly older than the children at the start of the last study. The children selected for the study used at least 10 signs and words but were not yet combining words. Six of the children were assigned to a treatment group and 6 to a comparison group. The comparison group continued with their speech and language therapy and received the intervention later.

The parents of the children in the intervention group received the Hanen Parent Program over 13 weeks (9 evening sessions, 4 home visits – role plays, videos, group discussions, given home assignments) with some modifications. In the second session parents chose 10 words they thought their child would be motivated to learn from 20 target word list. If the child used word 3 times spontaneously in different contexts – it was replaced with another word from list. The 20 words selected on basis of child’s CDI using 4 criteria; all were words understood but not yet spoken, words that typically developing children say by 24 months, started with phoneme already in the child’s speech-sound repertoire, words that represented objects that could be used functionally or words that could be demonstrated using gestures (e.g. more, all gone)

Parents were taught to set up routines that permitted modelling the target words many times and they were trained to use signs as they spoke words. Parents learned to use the word 3 times in daily play (for example in doll play, ‘baby’, ‘ put baby to bed’, ‘baby is sleeping’). The child was not required to imitate word. Parents kept diary records of children’s imitations and spontaneous words.

The children in the intervention group learned significantly more of the target words and were able to begin to generalise them as they used them in a new interactions with the therapist but the overall number of CDI words spoken at end of intervention was the same for both groups.

We include this study for two reasons; it is an example of parents being trained to deliver the intervention and, with the permission of the authors, we use the same criteria to choose the words to get into spoken production in the guidance we provide on the DSE Vocabulary Checklists.

Practical points: the intervention was parent delivered and provided more focussed opportunities to learn.

Provides guidance for choosing vocabulary to teach: Chose words already understood – receptive vocabulary and start with sounds child can produce. This fits with what the evidence that the sounds that child can produce influence first words (Stoel Gammon) and again highlights the importance of developing use of voice and speech sound production.

In their discussion these authors stress that many opportunities may be needed (100+?)

How do parents teach new words?

Another recent study which may inform interventions for teaching vocabulary was recently been published by Adamson and colleagues in 2015. This is an experimental study in which parents given novel words to teach in a play session. It is part of a series of studies in which this team explored the progress of 26 children with Down syndrome, 23 children with autism and 23 children developing typically. The children with Down syndrome or autism were 30 months of age and the typical group were 18 months of age but at the same language development level as the children with disabilities. They also divided the groups into children with less than 50 or more than 50 spoken words on the CDI. Nineteen of the children with Down syndrome had less than 50 words (mean 14 words) at 30 months and 9 had more than 50 (mean 98 words.)

Parents were given novel words to teach. One was a label for novel object and one a sound associated with the object. Parents given cue card and suggestions (e.g. This is a zuped. It goes loppy-lah). They were asked to introduce the novel words in play scenes.

The team coded the strategies that parents used to teach their child a new words and their child’s responses as follows:

1. Parent actively attempts to gain child’s attention before using word (point, shake object or speaking about it)
2. When producing word did parent try to draw attention to it by emphasising, stressing, elongating the word
3. Prompt child to say word (e.g. what is this? this is a ‘zuped’, say ’zuped’)
4. Did parent use novel word before i.e. is this a repeat?
5. Was there a child expressive response? – vocal or verbal response to parent use of word – could include producing word
6. Did child produce novel word?

They found that parents in all groups used different strategies depending on their child’s vocabulary size. If their child had less than 50 spoken words their parents were more likely to gain attention before saying the word and prompted less. If their child had more than 50 spoken words, they were much more likely to prompt the child to say the word in all the groups.

Parents of children with Down syndrome were more likely than those with children with ASD to repeat the novel word perhaps because they knew more repetitions would be needed for the child to store the word accurately and be able to imitate it. Importantly, repeating the word was a very important strategy as it increased the odds of their child making an expressive response by 84% for children with Down syndrome. The only effective strategy with positive effects for all children was prompting which increased the odds of a child producing an expressive response by 2 to 4 times depending on group, with the largest benefit for those with Down syndrome.

In their discussion of this study the authors report that parents were essentially good at introducing new words and gave child opportunity to respond on average 1.8 times a minute. (We may note here that this suggests that the target of 1 teaching episode a minute set in the Yoder et al 2014 study is not unrealistic). They continued to offer the word without expecting response. Parents were more likely to prompt when their child’s vocabulary is expanding and they know the child can produce words.

They also comment that more repetition by parents of children with Down syndrome fits with Mosse and Jarrold’s (date) recommendation for more repetition which we discussed in the last chapter, suggesting that parents may have observed repetition matters to establish phonological representation of word and ability to produce. They also suggest that repetition may support weak verbal short-term memory. This study supports the Girolametto (ref)study above in suggesting that the directive strategies of prompting and repeating are the most effective way to teach new words.

Practical points: we see again that directive approaches were the most effective - repetition of the word by the parent and prompts to elicit production. Parents were good at weaving in new words into play sessions – so both naturalistic and directive approach.

The authors draw attention to the need for support for delayed VSTM/PSTM and support for developing phonology.

Summary – vocabulary studies

Vocabulary needs to be taught. These studies identify why simply encouraging parents to be responsive and develop children’s non-verbal communication skills is not enough to get them talking and saying words. Repeating words and prompting production leads to gains.

Speech matters: The Yoder at al study makes clear that children’s communicative speech production matters and influences parent input. In addition, words attempted are influenced by the sounds child can produce – highlighting the importance of speech work from early.

Intensity: Teaching words and daily practice made a difference.

Use both naturalistic and behavioural approaches/strategies

Teaching vocabulary teaching in school years

In one of our own research studies, we evaluated a vocabulary teaching programme with 57 children with Down syndrome in primary/elementary age range of 5 to 10 years. The children were in 50 different mainstream primary schools in the UK and fully included in age appropriate classes. The children varied widely in ability and in receptive language (1 year 4 months to 5 years 8 months) at the start of the study as we had no exclusion criteria. We had children with additional needs and hearing issues as we wanted the study to be representative of all children with Down syndrome in our schools.

We developed and evaluated a reading and language intervention, so it has a language strand and a reading strand. The language strand was designed to teach new vocabulary, develop understanding of new words in different contexts and their use in spoken and written sentences. Eight new words, including nouns, verbs, adjectives and prepositions, were taught in themes over a two-week period with work being recorded each day in a topic book. This book was a record of learning and could be shared with peers and family. The topic books were very popular with the children.

We drew on the best practice guidance of Beck, McKeown & Kucan (2002) Bringing Words to Life in designing the activities for teaching vocabulary (ref). Most importantly, we worked with parents to make links between new words and words the children already knew and we engaged them in activities ‘to bring words to life’ so that they experienced the words in use to help develop their understanding. For example, when the theme was food and one new word was soup, discussing what soups they liked, bringing soups from home, exploring hot soups and cold soups, comparing textures - thin soups and soups with chunks in, making soups and trying them.

We used the same 4 step structured approach in every lesson: 1. Introducing new words - with a child level ‘definition’ of their meaning supported with pictures. 2. Reinforcing the meaning of new words - reinforcing and developing their meaning with examples of their use in different contexts, linking them with other words in the same category. 3. Using words in connected speech - using them in spoken sentences. 4. Using words in written language – putting them in written sentences, illustrated with pictures, in the topic books.

At all times children were encouraged to generate the use of the word in a sentence themselves to ensure they were fully engaged in the learning process and thinking for themselves. Of course, we provided prompts as needed and help to record sentences in the topic book. Many of the children were at the early stages of learning to read but the books were equally important for readers or non-readers as they could be read together with a peer or adult.

We focussed on extending their understanding of how the meanings of words may change in different contexts so helping them to increase their depth of understanding of words as well as develop links with other words. We will discuss the importance of these aspects of vocabulary development and ways to build both vocabulary and knowledge networks further in the chapter on vocabulary.

The vocabulary we chose to teach was based on collecting information on the vocabulary already known by the children by asking their parents to complete the DSE Vocabulary Checklists (ref) which cover the first 800 words most children learn. We included words to teach that many children understood but did not yet say. We also taught them to understand and say new words.

This study was a randomised control trial (RCT) so the children were randomly assigned to 2 groups. One group started on the intervention and the other continued with teaching as usual for 20 weeks.

The language teaching took place for some 20-25 minutes daily in a one-to-one session delivered by trained teaching assistants. They received two days of training before starting the intervention and were supported by school visits and telephone support from the team. They had a further training day after 10 weeks and the children’s progress was assessed after 20 weeks of intervention.

The children receiving the language intervention made significant gains on a bespoke measure of the taught expressive vocabulary (d=0.47, p=0.01) and taught receptive vocabulary (d=0.33 p=0.06) even though some of the taught vocabulary was already understood. However, like other language teaching interventions (ref) we did not advance vocabulary enough to improve the children’s performance on standardised language tests. Given their language learning challenges we would not expect to do so in 20 weeks. We could argue it is more important that we know they did add new words to their vocabulary when given specific instruction.

This intervention was deliberately designed to be delivered by teaching assistants in school to ensure that if it was effective it could be implemented at little extra cost. Most children with Down syndrome in this age range in the UK are being educated in their local mainstream schools and they have teaching assistant support.

This was a research study, so we had a protocol which applied to the intervention for all children so that we could assess their progress at 20 week intervals. Many of the children would have benefitted from spending more than two weeks on learning new words and a new topic but had to move on to the next topic for the purpose of the study. However, the topic books could be used at home and school to consolidate their learning. It is likely that more repetition was needed by many of the children to consolidate their understanding of words, to enable them to say them and to use them in sentences.

Like the earlier vocabulary study with toddlers, we provided lessons 5 days each week. On days 1 to 4 new words were learned and day 5 was for consolidation. We did not evaluate the factors influencing progress on the language strand but we did demonstrate that intensity of instruction influenced progress in learning to read. The more teaching sessions the children received the more progress they made with word reading. We discuss the teaching of reading further in the chapter on teaching sentences and grammar.

Practical points:

Intensity - daily structured teaching did increase vocabulary learning

Vocabulary should be taught during school years and it is usually a curriculum target for all children

Most children with Down syndrome will only have small vocabularies when they start school

Not all basic 800 word core vocabulary mastered and important to add words for school and curriculum

Bring words to life – experience the meanings of words through activities

Ensure we extend understanding of words, generalise to use in different contexts

Teach comprehension and practice production, categories…

Naess et al DSL+

Teaching vocabulary to school age children

Another randomised control trial of a vocabulary intervention was carried out in a population cohort of 103 first graders (6-7 years) with Down syndrome in Norway. The children were randomised into an intervention group (N=50) and control group (N= 53) who continued with their usual teaching. The aim was to evaluate the effects of the intervention on trained receptive and expressive vocabulary and vocabulary breadth and depth plus on a range of standardised language measures including expressive vocabulary, receptive vocabulary, receptive grammar, listening comprehension and narrative skills.

The intervention was based around shared book reading using digital multimedia presentation using pictures, video, animation, sounds and voices. The design of the materials was also intended to reduce the cognitive load for the children. The intervention strategies involved shared book reading, explicit teaching of new words, repeated exposure to target word, multiple connections to child’s experiences, and increase depth of understanding of words.

The intervention was implemented by a special education teacher or a teaching assistant in the school. All the materials were provided and they received a 6 hour training and were supported by the research team by phone and e mail throughout the study. The intervention was delivered one-to-one first 3 days of week, in a small group with other mainstream children on day 4 and in full mainstream class on day 5.

In total the children received daily sessions in school for 15 minutes over 15 weeks (75 sessions). The training was organised in 4-week pattern For 3 weeks it included picture book sharing and linked tasks, introducing new words each week. In week 4 the tasks were designed to consolidate and increase depth of learning (see examples of materials in paper..)

Results showed that the children made progress in both understanding and expression of the vocabulary taught. However, there was no transfer to standardised measures of vocabulary or grammar. Like RLI expressive language outcomes. – no effects attention NVMA – stresses no difference and expressive benefitted as much… so all children regardless of ability should receive vocabulary intervention.

Another method to increase the dosage of the DSL+ intervention would be to apply collaborative implementation across the school and home contexts; parent involvement can increase the intervention dosage and provide the child with diverse opportunities to use the target words in new contexts and with other communication partners. A short, simplified version of the DSL+ intervention has been tested in the USA for home implementation; the results show that the simplified DSL+ approach can be successfully implemented by parents (LeJeune et al., 2021).

The DSL+ intervention is cost-effective, suitable for digital intervention in schools, easy to distribute digitally both throughout Norway and internationally (c.f. also LeJeune et al., 2021), easy to administer, systematic, and predictable for both the child and the implementer.

Teaching two-word production

In a second study, members of the same team report on a study that compared interactive (modelling) versus didactic (direct instruction) therapy for teaching two-word utterances to two boys with Down syndrome over a six-week period.^[3] Each child received experience of both training approaches using different target utterances in each approach. The boys were 4 year olds and they had a core spoken vocabulary of at least 50 words and showed evidence of readiness for learning two-word semantic (meaning) relationships by demonstrating evidence of at least one at the start of the intervention. The therapy consisted of hour long sessions provided by a therapist in the boys’ homes twice weekly over six weeks. This gave the boys 140 minutes of each treatment approach. Parents did not observe the sessions and were not informed of the targets for training during the research study so that they did not influence the outcomes.

In both therapy approaches, the therapist followed the child’s lead in play. Only toys representing the therapy targets were available to the child. In the interactive approach, the therapist modelled the target two-word utterance ten times per session as the child engaged in appropriate nonverbal play. In the direct instruction condition, the child was asked to imitate the researcher’s two-word utterance in response to a prompt e.g. say ‘in box’. If the child did not respond to a prompt, then a more specific request for imitation was made. A total of ten elicitations were attempted at each session.

In both conditions, if the child spontaneously produced a target two-word phrase the therapist responded to the communicative function of the production appropriately and then expanded on the utterance with a semantically related two-word target item.

Both boys showed significantly better learning from the direct instruction than the interactive therapy approaches - more efficient productivity and generalisation. They did learn from both approaches but the elicited imitation procedure was more effective than simply modelling. However, it is important to note that both approaches were conversationally embedded into theme-based activities that encourage joint attention. Teaching during joint action and attention episodes has been shown to maximise the child’s attention, facilitate comprehension of adult speech and associated learning.^[4]

The boys learned in both therapy approaches and we suggest that the preferential effect of the didactic treatment may be related to the opportunity for them to practice target combinations and receive feedback on their attempts. They point out that as children with Down syndrome tend to experience difficulties with word finding and oral-motor speech skills, the didactic treatment embedded within naturalistic play may be an effective way to address the intrinsic language needs of these children.

Teaching grammar

Interventions for sentences and grammar

Two word stage

We start by considering another early study from the Hanen team. It is only a small pilot involving two boys with Down syndrome and published in Clements-Baartman, Girolametto 1995. The study compared the effect of two strategies, one more naturalistic modelling and one more behavioural, to encourage the boys to use two words together.

The boys were aged 4;1 and 4;9 at the start and both had core vocabulary of at least 50 spoken words and were considered ready to start joining words based on showing an understanding or use of at least one two-word semantic relation during the baseline assessment.

Both the boys received 15 visits from a therapist which included 2 to assess their language use before the intervention, 12 treatment sessions provided twice weekly and 4 post-test sessions. The boys received both interventions (20 mins each) for different two-word targets (140 mins therapy each condition over 6 weeks)

In the interactive, naturalistic approach the therapist verbally coded child’s nonverbal behaviour with a two-word target phrase. Each phrase was modelled 10 times in session.

In the didactic, more behavioural approach, the child was asked to imitate the two-word phrase with prompts for example (e.g. ‘say dog gone’, ‘your turn’) and again the therapist attempted to get 10 repetitions of the phrase into a session.

In both conditions, if child spontaneously produced a target two-word utterance, the therapist responded appropriately to the communicative function and then expanded with a related two-word target item. Both the boys learned more two-word targets in the prompted sessions. One used 5 of 5 target phrases from the didactic sessions and none of interactive or control items in probe sessions but used both in parent observations. The second child similarly learned 3 of 5 target phrases from didactic and 1 from interactive teaching. Both showed greater productivity and generalisation from didactic teaching condition.

The authors suggest that the children need the practice and feedback provided in the didactic teaching method which prompted imitation and spoken production. They note that in this study there were 120 models of target provided in interactive condition and aimed for in the direct teaching condition and suggest that other authors say over 200 repetitions may be needed to learn a new word. We should note that, while the didactic training which prompted production was the most effective, it was embedded in a naturalistic play situation.

Practice point:

Direct teaching and prompting production was more effective in moving the boys to 2 word production than just a naturalistic approach.

Indicates that practice and feedback matters - many repetitions are needed to learn to say a new word or how to produce 2 key word utterances.

Grammar teaching Sepulveda et al 2013

In 2013 a Spanish team published a study which aimed to teach a range of grammatical markers to 20 children and teenagers with Down syndrome, aged from 6 to 14 years. They divided the children into two matched groups, an intervention and control group.

The intervention was provided by the children’s familiar speech and language therapists in 30 sessions over 3.5 months with each session lasting for 30 minutes twice a week. Following two introductory sessions, the children received two teaching sessions for each of the following in Spanish; inflection for number in nouns, inflection for gender in nouns, inflection of articles for both gender and number, correct usage of pronouns, use of interrogative particles (who, what, where), use of adverbs and use of morphosyntactically (grammatically) correct adjectives. Four sessions were used to work on verb tenses; two sessions for the present, one for the past and one for the future. A total of five sessions were dedicated to work on sentence structures and another five sessions on dialogues.

The children in the control group continued to receive their usual speech and language therapy sessions for the same amount of time as provided in the intervention.

The children were all assessed on standardised measures of language and communication (morphology, syntax, semantics (word meanings) and pragmatics (communication).

The training activities for the intervention group were structured such that two sessions were set aside for each of the following aspects: inflection for the number in nouns, inflection for gender in nouns, inflection of articles for both gender and number, correct usage of pronouns, use of interrogative particles, use of adverbs, and use of morphosyntactically correct adjectives. Four sessions were used to work on verb tenses; two sessions for the present, one for the past and one for the future. A total of five sessions were dedicated to work on sentence structures and another five sessions on dialogues.

The children in the intervention group made significant gains on the standardised measures of morphology, syntax and semantics but not pragmatics at the end of the intervention when compared with the control group. The gain on semantics was unexpected as this was not targeted in the intervention.

The teaching approaches included

• direct teaching with prompts to elicit productions
• used a range of examples to encourage generalisation of structures
• provided at least two activities for each target, explained to child and modelled by therapist. If child made 3 consecutive errors therapist repeated the instructions.
• made full use of visual supports.
• Used sad and smiley faces – if showing a sad face did not lead to self-correction the teacher gave correct response but did not require repetition
• Teaching was made fun with use of a race game – at end of race child was given feedback and asked to repeat structure correctly
• (Examples of the teaching activities are given in the paper – put in grammar chapter?

In summary, this study demonstrates that the explicit teaching of grammar can be effective in improving its use in spoken production in children and young people with Down syndrome. Again, we see the use of explicit prompting of production which enables the young person to practice production.

Baxter et al 2021, Baxter 2023

A training study to teach the rule for using the regular past tense marker on verbs

The first part of this study was an RCT involving 52 children with Down syndrome in the UK aged 7-11 years and educated in mainstream, regular education. The children were randomly assigned to an intervention group or a wait group. The intervention was delivered one-to-one daily for 20 minutes over 10 weeks (50 sessions) in school by teaching assistants (TAs). The TAs were already working with the children in their classrooms. They were provided with a day of training, a manual and materials and fortnightly support visits from an SLT. The wait group continued with their usual education. At the end of the RCT, the wait group then received the intervention for 10 weeks.

The intervention involved teaching the children the regular past tense (- ed) compared with present progressive tense (-ing) on 4 verbs each week so that 40 verbs were used over the 10 weeks. Training included a range of activities including explicit teaching, multiple repetitions, modelling, recasts, acting out and recording sentences. The teaching made full use of visual supports including pictures, print, videos and photos. The printed materials made the ‘ed’ visible and explicit. The full details of the intervention are in Baxter 2021, 2023. The TAs delivery of the intervention was rated on a number of measures of effectiveness.

The RCT results showed a significant effect of the training, with the intervention children increasing their use of the regular past tense morpheme on the taught verbs and generalising to use it appropriately on untaught verbs. They also showed implicit learning of the regular past tense rule as, after the intervention, 21 of the 26 children overgeneralised the rule to irregular verbs (e.g. drinked, eated). The 5 children who did not show this overgeneralisation were the children who made the least progress on learning the regular past tense rule in the intervention. These 5 children could not have been predicted from scores at baseline as children with more delayed language or cognition made progress. However, they did have TAs with lowest scores for effectiveness of delivery.

Looking at individual differences in learning the grammar rule across the group, amount of progress was influence by cognitive ability as measured by non-verbal mental age (NVMA). After controlling for NVMA, those with better receptive language, able to provide more information on an expressive language test, or scoring more highly on reading and letter sound measure made more progress.

The waiting control group made similar progress when they received the intervention. The intervention group maintained their progress over the following 10 week period after their intervention ended supporting the conclusion that they had acquired implicit learning of the rule.

This is a landmark study showing that children with Down syndrome can master grammar rules if explicitly taught. The author notes that children with Down syndrome may need explicit teaching as they may not be able to extract the information for grammar rules from simply listen to conversations. Grammatical morphemes are not stressed and may be difficult to hear and remember, especially for children with limited verbal short-term memory. Their speech difficulties may mean they cannot say them and they find they are understood in context without them.

Baxter. R, (2023). The efficacy of a language intervention on the acquisition of past tense in children with Down syndrome. UCL (University College London). Baxter_10163931_thesis_redacted.pdf (ucl.ac.uk)

The only published intervention study that has attempted to improve comprehension and production of more advanced grammar and syntax is that of the author. [TODO: references 74],^[5] Twelve teenagers took part in a year long language teaching programme. Twice each week they had individual sessions in which they practised production of syntactically complete sentences. Each sentence was imitated twice - the student repeating it after the teacher. The student then repeated the sentence on their own twice more. Twelve different examples of the same sentence structure were worked through at every session. All the sentences were supported by picture illustrations to act as prompts. For half of the new sentence structures learned during the year, the written sentence was under the picture as an additional support during the training trials, to allow the effect of reading as well as listening to be evaluated.

Training began with sentences that the student understood but could not say in full - that is, the student would use the key content words and leave out the function words -‘’He sit chair’‘, rather than ’He is sitting on the chair’. During the year, new structures were taught in comprehension as well as in production. All the training was audiotaped, transcribed and analysed to evaluate progress. At the start and end of the year, conversations were recorded with each teenager to see if any training benefits had transferred to their everyday spoken language. The teenagers enjoyed all the language training and were eager to come to the sessions.

In addition to the carefully organised repetitive practice of the sentence structures, examples of the structures being taught in this way were looked for in the student’s everyday experience and language books were made. These books were illustrated with photographs taken as the students acted out illustrations of the sentences. For example, for sentences using ‘above’ and ‘below’ as new concepts, one student sat on a table and one underneath the table. They wrote in the book under the photo ‘Julie is above Grant’, ‘Grant is below Julie’. They then changed places and we photographed ‘Grant is above Julie’ and ‘Julie is below Grant’. In the same manner, we took photographs to illustrate comparatives ‘Grant is taller than Shaun’, ‘Shaun is shorter than Grant’, ’ Grant is shorter than Stephen’……and passives ‘Shaun is pushing Grant’, ‘Grant is being pushed by Shaun’.

To add to the language books and further support generalisation of the training to everyday use we made conversation diaries to go between home and school. In these, the student decided what they would like to tell their family about their activities at school and this was written in the diary in the first person, to support narrative language (story telling or event reporting). For example, the student might write ‘I cooked sausage and mash today with Julie. We went to the shops first to buy the food’. At home the student might write ‘I went to the cinema with Grandad last night’ or ‘I watched TV with my brother last night’. In school, the diaries were read out each morning during a newstime slot in class. At home and at school, parents and teachers were asked to read with the student as necessary, as some could not read independently. Some students could write and read their own sentences but most had to copy the sentences and have support to read them correctly. Parents were requested to read with them and to help them to practise the sentences that they took home.

All the students benefited significantly from the training. This was demonstrated both in gains in the formal assessments of their comprehension and production and in the structure of their language in everyday conversations. Unexpectedly, the ‘least able’ students benefited the most from the reading support. They were the students with the poorest verbal short term memory spans and they simply could not repeat a six word sentence even when they had just heard it and had a picture prompt in front of them. However, they soon learned the printed words by repetition during the training sessions and used them to support correct practice. The students who could already read to a 7 or 8 year level, had longer digit spans of 3 or 4 digits (probably as a consequence of reading instruction) and could repeat the sentences correctly with or without the help of the printed version.

Gains in production over the year were affected by speech difficulties, those with the worst phonological or fluency problems made the least progress in production of sentences (though not in comprehension). The teenagers with the smallest digit spans at the start of the study also made slower progress in extending their productive sentence lengths. This could be because working memory capacity influences information processing ability as well as sentence planning and production.

Developing grammar and language use in school years

In addition to considering the development of sentence structures and grammar for school age children, we should also consider how they are developing the ability to engage in longer conversations and talk about their experience. This is described as narrative development and it may be progressing quite slowly for young people with Down syndrome during primary and secondary school years. Anne van Bysterveldt and colleagues in New Zealand have begun to explore the issues of narrative development as have Kay Raining Bird and colleagues in Canada. Before considering a pilot intervention study we discuss a study which explores the role of conversation partners in narrative use by young people with Down syndrome.

In 2016 Anne van Bysterveldt and Westerveldt recorded 10 children and young teenagers with Down syndrome talking with their teacher aides. The children and teenagers ranged in age from 6;9 to 13;0 and were in primary schools in New Zealand. The children selected had to be able to speak in short ‘sentences’ of at least 3 words and to be largely intelligible to unfamiliar listeners as reported by their parents. They were asked to produce personal narratives with support of two sets of picture prompts, a standard set used in other narrative studies and their own personal photos.

The teacher aides had set instructions. For the standard set they asked ‘did anything like that happen to you?’ For the personal set they were asked to share what happened in their photos. The teacher aides were able to prompt for more information when needed but reminded to avoid leading questions and to use open-ended prompts e.g. ‘can you tell me more?’

The researchers analysed the first 50 intelligible utterances of the children’s narratives (one child only managed 38 utterances). The first point to note is that there were considerable individual differences in personal narrative performance as the children’s mean length of utterance (MLU) in words ranged from 1.7 to 5.9, the number of different words used ranged from 26 to 119; and the intelligibility of their utterances ranged from 69% to 98%.

There was an increase receptive vocabulary, MLU and number of different words used with age but not with intelligibility.

Most of the children were able to narrate at least 3 past tense events but only 1 child produced a classical narrative structure, that is, a personal narrative to containing at least two past events that are organized around a ‘high point’, the information that the speaker is most interested in sharing - ‘the point of the story’. Research suggests that children with typical language development suggests a that between 2 and 3 years children start producing simple two-event narratives without a high point, and that they gradually improve their narratives so that by 6 years they can tell a coherent story with a clear end point.

As we might expect, the children performed better with personal photos. When talking about events they had actually experienced, they produced longer utterances (higher MLU), they used a greater number of different words and they were more fluent shown by a lower percentage of mazing (such as repetitions and revisions) but these differences not statistically significant.

The researchers also analysed the way the teacher aides spoke to the children. They found that they generally adapted their grammatical complexity to that of their students and used short sentences. Importantly, there was a significant negative correlation between number of utterances produced by teacher aides and the children’s MLU in words and number of different words used. In other words, more TA talk led to shorter utterances from child and fewer different words (like the deaf study in chapter 1).

This is not an intervention study, but it highlights firstly that children with Down syndrome only slowly develop their ability to tell personal narratives to share what they have been doing. This will limit their social engagement and sharing with other children. Secondly it shows the way a communication partner talks influences the way children talk. When the partner follows the child’s lead, listens and asks open ended questions, children are able to develop longer conversations and narratives. The other important point to note is that these children are in primary school and during this time other children are making important progress in developing narrative abilities in conversations and in written work.

#### A narrative intervention with teenagers

Our young people may be more ready to develop these narrative abilities during their teenage and early adult years as suggested by a pilot study of a narrative intervention published by Finestack and colleagues in 2017. They recruited four girls with Down syndrome whose ages ranged from 10-15 years and provided them with 18 individual intervention sessions targeting personal narrative skills in their homes. The girls had to have an average MLU in conversation of greater than 3.0, to be largely (70% of utterances) intelligible, have no more than mild hearing loss and no addition diagnoses such as ASD.

A novel aspect of the intervention was that they provided the girls with an iPod Touch photo application so that they could take photos of their everyday activities which were then used to help them recall, create and retell personal narratives. (We might note that this similar to the approach we used in making conversation diaries in our research study and software to make stories using personal photos is readily available for phones and tablets).

The intervention visits were provided by speech language pathologists. They used story modelling, story retell, personal narrative generation and supported personal narrative retell activities to teach these young people how to develop a narrative. They also targeted grammatical goals to help the girls improve their sentence structures. We will describe the activities they used in the practical chapter on communication.

All four girls made small treatment gains on at least one measure. The greatest gains were made by the two older girls (12;1 and 15;4) the other two were both 10;1. It was not clear why these two made the most progress as one of them had the lowest language scores of the four girls and one younger child highest non-verbal mental ability score. The authors point out that age seems to be an important factor and that interventions should continue through teenage years.

While this was a pilot study, it does demonstrate that both grammar and narrative abilities may improve with intervention.

Practical point

Developing narrative abilities should be a target – and this can be combined with developing grammar in production

Phonology

There have been three studies, two in Australia by Dodd and colleagues^[6],^[7] and one in Ireland by Ni Cholmain,^[8] that aimed to improve the children’s phonology (speech sound production). All the studies involved the children’s parents as the main agents of change.

The first study initially involved 8 children aged 3 years 5 months to 5 years, one had no functional words and 3 made only delayed phonological errors. The other 4 were involved in the intervention programme as they had phonology that was inconsistent and typical of Down syndrome according to the authors. The parents met weekly for a comprehensive speech and language programme but with an emphasis on phonology. They chose to target whole word production, selecting 10 common and functionally important words for each child (family names, names of common foods and words like ‘more’, ‘no’, ‘again’). These 10 words were chosen in discussion with parents.

Some contrasting word pairs were given for voice/voiceless practice such as bee/pea, pat/bat, coat/goat, toe/dough or for omission of initial consonants egg/leg, egg/peg, eight/gate. Parents were told to accept an error if it was developmental (see side box) but to require consistent production. Words were to be elicited in games situations and parents made lotto games, played hiding games etc. High frequency words were chosen as far as possible in the hope that they would be used spontaneously and the child was then required to produce the word correctly. All the children made exceptional improvement in the percentage of consonants correct over 13 weeks (27% to 63%, 25% to 54%, 33% to 64%, 44% to 67%. Further the error patterns changed from deviant to normal (30% to 60%) and the gains were sustained 3 months later.

Targeting words, initially in isolation, then in sentence frames and spontaneous speech, allowed integrated teaching of phonology, vocabulary and syntax. Setting a small number of targets, especially as parents helped to choose them, provided feasible concrete goals. The choice of high frequency words - that is words that are in daily use - taught the child the power of having functional language. The intervention programme also involved the children’s preschools.

In the second of the Australian studies by Dodd and colleagues,^[7] parents took part in the same 13 week training programme. The programme informed parents about all aspects of communication, language and speech and trained them in observation and listening skills as well as in responsive strategies to enable them to elicit word production during play with their children. This involved viewing video samples of each child and discussing their progress. There was much group support and discussion, and also individual counselling sessions. The second study again reported dramatic improvement in the children’s phonology at the end of just twelve weeks. Measures of the children’s spontaneous speech showed a mean increase of 30% in the percentage of consonants correct and a change of error type from inconsistent errors to developmental errors.

Dodd emphasises that phonology is inconsistent in the spontaneous speech of children with Down syndrome, not just delayed. As their production in imitation is better, this suggests a planning and motor control issue. She suggests that children with Down syndrome have a memory deficit for sequences of fine motor movement and that phonology programmes should therefore take account of the need to train motor memory. Prior experience of words seemed to make little difference as nonsense words were produced as well as real words.

Dodd and colleagues also stress the benefits of groups for intervention work with parents and children. They point out that groups provide a mutually supportive atmosphere in which parents are more responsive, and more likely to change attitudes and learn new procedures. They provide a wider range of learning and management problems for discussion, allowing advice to be given for handling future difficulties and sharing experiences. Groups can break down parent-professional barriers and professionals can become less rigid. Parents can become a source of ideas for management and develop their confidence and self-esteem. Groups are also cost effective.

In the second Australian study, the parent’s communication skills were rated before the study, 8 weeks into the programme and within 4 months of the end of the programme by videoing each of them with their child engaged in the same play activities on each occasion. The children’s progress was related to the parents’ skill in communicating. The nine children in this study were aged from two to six years of age and varied in their communication skills at the start. The three most verbal children (aged 4 years) showed improvement in phonology as reported in the first study. The three youngest children (two and three year olds) used few words at the start and half of these showed rapid gains in word use during the programme. Three other non-verbal children, aged 4 to 6 years, made little progress. Two of these children were described as having behaviour problems and parents reported that they were rarely able to gain their children’s co-operation in any activity. The third child had chronic illness that reduced the amount and scope of parental intervention.

This study highlights the importance of recognising the variations in the development of children with Down syndrome. Two thirds of the children showed benefits in speech production skills and/or new word learning. Two children had poor attention skills and behaviour and the implication is that these needed to be improved before specific communication intervention would be effective. The third child was sick and therefore this was not a good time to ask parents to target communication work.

In the Irish study by Ni Cholmain,^[8] six children aged from 4:1 to 5:6, were involved in the study. Their expressive language ages at the start of the study ranged from 1:2 to 2:9 years and the percentage of consonants that they could pronounce correctly ranged from 3 to 38%. All the children had reported hearing loss of more than 40dB at times in the past. They were in therapy in the speech and language service before the study, so baseline measures of rates of progress for the 3 months before the intervention were available. The programme focused specifically on phonology and each child’s listening and production activities were individually planned.

Listening

Parents were provided with portable sound amplifiers. These were used to provide the child with low-level amplification while listening to word lists containing the targeted phonemes for the child. The numbers of words on a list varied from 10 to 20 and 4 to 6 lists were worked on during any given period. The lists consisted of one list of words with a phoneme already in use contrastively and unaffected by the process in the child’s phonological system and 3 to 5 lists with the phonemes absent or affected by the process targeted. These lists were read to the child by a parent for 6 to 10 minutes every day. Each list was read twice and the list order followed. Parents were also encouraged to read a short story to the children at night as an additional listening routine. These were selected to include repetitions of words and phrase structures, but without reference to the current process or phoneme focus.

Production

Words for production practice were selected from those best approximated by the child in imitation activities using amplification during a clinic session. Five to 10 words were given containing one or more examples of the target process. These words were practised using amplification following the listening session. Parents were also encouraged to focus on the words in activities such as shop games and picture lottos. They were also encouraged to use clarification requests, indicating uncertainty, regarding objects named or requested during games and everyday routines.

Sound practise

Sound cards or books were provided for the children. These contained picture symbols both for phonemes being targeted and those already in the child’s system. They were used to encourage the children to think about speech sounds as sounds, and provided opportunities for listening and production practice in play.

The programme was applied on a cyclical basis with input periods of 4 to 6 weeks followed by a break of 4 to 10 weeks. During the input period the children were seen by their therapist on a weekly or fortnightly basis for identification of processes to be targeted, provision of word lists for listening, and production practise and reassessment. Parents practised work for 20 minutes daily when possible.

All the children in the study showed change in their phonological systems within the first two weeks, which Ni Cholmain describes as reflecting the beginning of a reorganisation of the child’s system. Not all new phonemes or process changes were fully achieved in the time period of the study but they did indicate movement towards mastery. The percentage of consonants correct changed from 3 to 38% before intervention to 19-88% correct. The author describes that the progress of the phonological changes seen was as would be predicted from normal developmental progress. The processes eliminated first were those dropped earliest in typical development i.e. initial and final consonant deletion, nasalisation and stopping.

Ni Cholmain reports that the programme appeared to have encouraged the children to recognise that a phonological system existed and that their own systems required reorganisation. The increased intelligibility that resulted appeared to liberate the children’s syntax development as reflected by a move into early grammar.

Grammar and speech – school years Camarata et al. 2006

Another research group interested in speech development, Stephen Camerata and colleagues in the US, published a study in which they combined targeting both grammar and speech development using a Broad Target Recast approach.

The study involved 6 children with Down syndrome aged from 4 to 7 years. They had MLUs (in morphemes) above 1.00 so they were combining words and could produce at least 20 partially comprehensible utterances in a 20 minute conversational sample. The children received sessions twice a week over a 6 month period.

The speech measure used was the percentage of communication units understood (46.5% (SD 12.83) before intervention). This is an intelligibility measure which will link with clarity of production but they do not measure % consonants correct. This team suggest that improved comprehensibility is the most relevant measure of improvement. It is not necessary to achieve fully accurate production of words to be understood.

A speech recast was provided by repeating a poorly articulated word produced spontaneously by the child. For example, if the child said “this is a ‘wion’”, the therapist said “yes, ‘lion’”, providing a clear adult form of the word but not adding information. A grammar recast adds grammatical information. For example, the child says ‘She seep’ and the therapist says ‘yes she sleeps’). The aim was to provide 4 recasts a minute.

Four of the 6 children showed improved comprehensibility and 5 showed growth in MLU with 2x 2 children showing generalisability of each gain. This is an interesting pilot but no control group so we cannot be sure their progress over 6 months was faster than if they had no intervention. However, a valuable aspect of this study is the demonstration that it is possible to use the recast approach to support both speech clarity and grammatical development.

Fiani et al speech 2021

Effects of mother’s imitation on speech sounds in infants with Down syndrome

This small study reports on the effects of teaching mothers to reinforce their child’s vocalisations and shows that babies with Down syndrome will learn to increase their vocalisations from as young as 4 months of age. The authors recognise the importance of early development of babble for later speech and language. They suggest that babies with Down syndrome may spend less time practicing speech sounds than other babies and this may influence their ongoing speech development.

For the study, they enrolled 3 babies with Down syndrome at 4 to 4.5 months of age. The mothers were taught to imitate the child’s speech sounds (Contingent Vocal Imitation) – practice training sessions given. The mothers conducted 2-4 sessions per day, depending on state of infant, 2 to 3 days a week. Each session lasted 3 mins with a 3-5 minute break between on a given day. If infants were distressed session ended. The babies’ responses were compared to those from sessions when their mothers reinforced randomly but at same rate as in CVI but not contingent on speech sounds.

All 3 infants increased their production of speech sounds in response to CVI as has been shown with typically developing infants. Their mothers enjoyed the intervention, found it easy and indicated that they would continue. The researchers suggest that there may be a developmental window and at this very young age. Hearing their caregiver imitate a vocalisation may be particularly reinforcing to infants. It may also provide additional auditory input so may compensate for hearing loss. The authors also point out that in intervention includes positive affect, eye gaze and joint attention so may increase dyadic interactions between mother and baby. There was no long term follow up in this study to see if increasing early vocalisations had an effect on ongoing babble and word learning. However, it does suggest that the babies were capable of learning and increasing speech sound practice as early as 4 months of age. With what we know about the significance of early development of speech perception and production in these early months on later speech, language and verbal memory development. it is an important indicator of the benefit of starting intervention early.

Speech and phonological awareness intervention Van Bysterveldt et al 2009

We have discussed the importance of phonological awareness in relation to developing verbal-short term memory and Anne van Bysterveldt and colleagues have looked at the effect of explicitly combining phonological awareness activities with speech activities to improve speech sound development. This study involved 10 children with Down syndrome (5 boys, 5 girls) 4;4 to 5;5 years. They all attended an early intervention centre but did not receive any additional speech and language therapy during the study. Seven of the children had some hearing loss (slight to moderate 17.5dB to 39.37dB) and 6 wore glasses (one child had cataracts removed after study).

Four targets for speech intervention were chosen for each child and a control target based on individual speech profiles. Twelve cards were prepared for each sound with pictures and words containing target sound. Six of the cards were used in the intervention and 6 to assess generalisation of learning later.

The children learned the letter names and sounds, to say words with the targeted sounds in and to identify words starting and ending with sound. They received a 20 minute intervention session with a speech and language therapist weekly and another 20 minutes on working on the same targets on a computer at the early intervention centre. The therapist led intervention sessions typically began with a letter knowledge activity, followed by one or two phoneme matching tasks. Each session also typically contained at least one game where letter name, letter sound and phoneme matching were integrated into the one activity. The sessions integrated speech goals with phonological awareness and letter knowledge goals.

The intervention was delivered in two 6 week cycles with a 6 week break in between.

Parent component- trained to use print referencing technique to bring child’s attention to targeted letters and sounds during joint story book reading 4 times a week for 10 minutes over 18 weeks and monitored weekly. Parents were video-taped to check they could deliver the print referencing effectively.

All 10 children demonstrated significant improvement in production accuracy for trained and untrained words after 20 hours of intervention over 18 weeks and the authors note that their findings support those of the Dodd and Ni Cholmain studies we have reviewed above. There was considerable individual variation in response to the intervention and this could not be explained by differences in receptive language, age or hearing threshold.

All children learned some letter names and sounds that were their targets but this did not generalise outside the intervention sessions for all of them. The children who made greatest gains on letter knowledge and phoneme awareness had language ages of at least 3 years (receptive and expressive combined) and the authors suggest a certain level of language knowledge may be needed to support this learning.

In summary, this study again demonstrates that speech interventions can improve the accuracy of speech production in young children of 4-5 years and that an intervention approach that integrates speech, letter knowledge, and phonological awareness targets is effective in remediating speech error patterns at the single word level in young children with Down syndrome. In the UK children will be learning letter-sound correspondences in preschool from 4 years of age and playing phonological awareness games. In our experience, we find that we can teach children with Down syndrome letter-sounds from 4 years of age and when most will be in inclusive preschool settings at this age in the UK so benefit from learning them with their peers. It would therefore not be difficult to use letters to support speech work at this age. It is also possible that teaching letter- sounds and playing phonological awareness games such as listening for the first sounds in words will improve verbal short-term memory development.

Key points for speech development – young children

Targeted speech sound work can improve accuracy of single word production from as early as 4 years of age

Choose words important for the child

Include listening and production activities

Consider amplification

Teach letter-sound correspondences

Use phonological awareness activities

Speech comprehensibility Yoder, Camarata et al 2016

In a more recent study published in 2016, Paul Yoder, Stephen Camerata, the same team evaluated the effectiveness of a speech recast approach compared with a commonly used ‘contrast’ treatment (Easy does it - cycles) approach. They recruited 51 students with Down syndrome aged from 5-12 years and randomly assigned to them either a speech recasts (BTSR) or contrast treatment group.

In this study they again measured comprehensibility, which was the extent to which an unfamiliar listener can understand an utterance. They focussed on conversational samples, not single words. They rated utterances in two 20 minutes conversation samples to give a measure of the % utterances that were fully comprehensible.

Before the intervention they took a measure of the children’s ability to imitate words and sentences using the following verbal imitation measure (aah, night-night, ma, pa, hi, bye-bye, cookie, she wants more, his mother opens the box, Susan is brushing her teeth ) scored over up to 9 trials an item over 3 sessions.

The therapy sessions with an SLP took place for an hour twice a week for 6 months. In the Broad Target Recast sessions the therapist repeats the child’s utterance providing a model of correct production. For example, if the child says, “Ah wa du,” the speech recast might be “You want juice.” The “broad target” part of BTSR is that the therapist uses speech recasts for any word the child attempts to say that is inaccurately produced and provides a developmentally appropriate recast. It is proposed that these recasts help the child to improve their stored representation of words if repeated often enough. In the therapy the target for the therapists was to provide 4 recasts a minute – used verbal routines and topic continuing questions to encourage talking.

The comparison treatment was considered to represent the more common approach to speech work which includes working on specific sounds and phonological processes using cuing for articulatory placement and elicited imitation.

For the groups as a whole, there was no difference in the effectiveness of the two therapy approaches. However, the children who scored highly on the verbal imitation test did better on BTSR. This suggests that a level of imitation ability is necessary to fully benefit form recasts. This is described as a moderating effect.

The authors also reported an intensity effect in this group of high imitators, that is the more recasts that children heard during the intervention period, the more progress they made. This is a mediating effect.

They suggest that that the imitation task indicates when child has sufficient VSTM capacity and motor capacity to compare sound templates and alter their own productions when mismatches occur. This would be necessary to benefit from speech recasts. They also suggest that parents, classroom teachers, or paraprofessionals could be trained to implement BTSR effectively as it does not require any technical knowledge of phonetics or articulation.

Behavioural approaches Feeley, Jones et al. 2011, 2015

The last group of intervention studies we should consider is the work of Kathleen Feeley, Emily Jones and colleagues which are based on a tightly structured behaviour approach. Studies by this group working with children with Down syndrome from 6 months have used using an Applied behaviour Analysis (ABA) approach delivered by a trained and experienced interventionist. Therefore they use behavioural shaping and reinforcement strategies.

They have published several studies aimed at teaching non-verbal communication, requesting, using eye-gaze, vocalising and imitating of sounds. Sounds were chosen with parents based on child’s current sound production repertoire and toys used to elicit requests. They report progress in learning to request and to verbalised which generalised to new sounds and requesting new toys. However these are small case studies and there are individual differences in children’s responses.

Summary – what have we learned?

In this chapter we have reviewed x intervention studies and many have limitations. Some have involved only a small number of children or young people, others have larger groups of up to 30 or more participants. There have been few replications of interventions though some interventions such as PMT/EMT do have a considerable research base of studies with children with intellectual disabilities supporting them. Barbara Dodd and colleagues have developed and evaluated their Core Vocabulary approach for children with inconsistent speech disorders since their early work with children with Down syndrome. They have also conducted research which suggests that children with Down syndrome have this inconsistent speech disorder. Most of the interventions are based on sounds developmental principles.

Despite these limitations, we think we can draw some conclusions to inform effective interventions. We have summarised the specific practical implications for interventions for each domain, communication, vocabulary, sentences and grammar and speech, at the end of each section and we will take those forward to each practical chapter.

We can also identify some common findings which provide some principles for guidance across all interventions.

Principles for effective interventions

Interventions make a difference: Firstly, these studies, despite their limitations, do indicate that planned interventions do make a difference and help young people move forward in each domain, communication, vocabulary, speech and grammar, faster than without intervention.

The importance of prompting production: several studies indicate that children with Down syndrome need support to imitate, produce and practice spoken words in order to learn new vocabulary, develop sentences and improve speech clarity. Several researchers note that this is likely to be because of verbal short-term working memory delays and speech production issues.

Directive + naturalistic: However, while this more directive, behavioural approach seems to be important, most studies did embed this in naturalistic play or conversation sessions. This enables children and young people to learn in situations where they can ‘see and experience what language means’. However, just a naturalistic play approach is not enough to move language forward.

Importance of developing speech: the milestone study of Yoder and colleagues highlighted the processes at work and the pathway to vocabulary learning. They showed that when intervention increased canonical syllable production in the first 3 months, this led to saying more words at the end of 9 months. We saw in the last chapter on working memory that children’s speech sound production from 11 months of age influenced the development of verbal short-term memory.

Developing receptive vocabulary: the same study showed that developing receptive vocabulary predicted progress

Child-parent interactive influences: parent linguistic mapping

The way we talk to children: The narrative study by van Bysterveldt and colleagues showed the effect of teaching assistant style of talk on the children’s progress supporting the work we reviewed in Chapter 1 on adult talk to children.

Individual differences: not all studies have explored this but our work on teaching grammar suggested that progress was influenced by verbal short term memory and speech production abilities. Those with more limited VSTM or more speech difficulties made less progress. We see now across these first 3 chapters the repeated evidence that speech issues and working memory issues are holding back spoken language development.

Non verbal cognitive abilities: Studies also report that nonverbal abilities do influence progress.

Intensity matters: the Yoder study indicated that 5 sessions a week was more effective in teaching new vocabulary to toddlers that 1 session a week. Our research on language and reading development showed, for reading, the more lessons a child received the more progress they made. The implication of this are that we need to support parents and teachers to deliver daily interventions.

Working with parents and teachers: We hope that this book will enable parents and teachers to assess individual needs and develop interventions if they do not have a speech and language therapist to guide them. For therapists, parent and teacher training and support should be a priority.

A number of the studies we have discussed have included parent training as part of the programme and some have trained parents to deliver the training. There have been some studies of parent effectiveness comparing parent and therapist outcomes showing that parents can be as effective as therapists. In addition, parent training may lead to longer lasting gains (Kaiser & Roberts). However, in research studies parent and teacher training has often been over a number of weeks, involved individual coaching and ongoing support so needs to be planned and adequately resourced. Connie Kasrari’s research team have shown that parents and teachers can be effective in delivering their JASPER programme and they adapted training to reach families who needed training at weekends or evenings or in their own language. We will discuss the issues further and implications for practice in the final chapter.

Multi-sensory approaches to support learning from listening

Signing

The available evidence suggests that parents should be encouraged to learn to use sign to support the development of spoken language with their baby from 7/8 months of age. Research studies show that children with Down syndrome do not learn words easily from speech input on its own^[9], and that those who have been in sign supported therapy programmes have bigger spoken vocabularies at 5 years.^[10],^[11],^[12]

Practitioners have advocated the use of augmentative signing with babies with Down syndrome since the early 1980s^[13],^[14] and evidence for its effectiveness has slowly accumulated. It can help in a number of ways. If parents sign as they speak, they make sure the baby is looking, the sign holds the baby’s attention and it gives an added clue to the meaning of the words. Parents are also likely to stress the words they are signing. In other words, signing may help to structure more effective language learning situations.

For infants, signing can increase their productive vocabularies as they can usually sign words before being able to say them. Signing will reduce frustration and increase communication opportunities. They know what they want to say but cannot yet produce the words. At this stage, signing helps to maintain effective communication, and is likely therefore to increase the rate of language learning until speech emerges, as it is hard to keep talking to a non-talking child.

However, it is essential to keep up activities to encourage sound and speech production alongside the use of signing, if children are to move into using spoken words as early as possible. In the author’s experience, most children are able to drop the use of sign slowly from around four to five years of age, though they should not be discouraged from using sign at any age as a repair strategy when their speech is not understood.

The possible ways in which signing may assist speech and language skills in people with Down syndrome warrants more sophisticated analysis than we have available to date since one study illustrated that the speech clarity of adults with Down syndrome improved when they signed as they spoke.^[15] Individual case examples from parents and practitioners indicate that signing often helps the child with Down syndrome to find the word they want and to speak more clearly. Signs for sounds have helped production of initial and end sounds in words and signs for grammatical markers can help to teach grammar.

By school age, reading activities can increasingly take over as the support system for language learning and spoken language should be encouraged as the main means of everyday communication.

Teaching reading to teach grammar

Current evidence suggests that teaching children with Down syndrome a sight vocabulary should be a routine part of speech and language therapy and a priority in all early intervention and education programmes.^[16]

Learning to read will help vocabulary learning and it will particularly support grammar and sentence learning.^[17],^[5], [109-113] Three to four years of age is the optimal time to start for the majority of children, 2 years for some, later for others. Once a child has a comprehension vocabulary of about 40 to 50 words and can play picture lotto games, that is, match, select and name pictures in speech or sign, reading should be introduced.^[18] Production of multi-word utterances and sentences with correct grammar can be greatly assisted by reading sentences.

Reading also supports speech practice and can improve articulation. The sounds and word patterns become visual clues to how to say the sound or word. Once letter sounds are known, letters can be used to support practice. Teaching a sight vocabulary, choosing words with different initial sounds, sound patterns and number of syllables, will help to build speech practice. Children with Down syndrome speak more clearly, with more accurate phonology, when reading, as do other children. This is probably partially due to the fact they do not have to plan the sentences that they read, so freeing up more information processing capacity for controlling speech production.

In the author’s view, early reading activities may be one of the most effective interventions to develop the speech and language skills of children with Down syndrome. Children who have been taught to read from 2 to 3 years of age, as part of a language teaching programme are frequently among the most advanced children with Down syndrome in speech, language, and cognitive skills. It is possible that at this age the brain is most receptive to learning a first language visually, or simply that the reading is a very effective way of structuring the teaching of words and sentences, and supporting spoken practice. Children who engage in 20 minutes a day of reading practice, even if it is in imitation, are getting consistent practice of speaking in grammatically correct sentences long before they are able to generate such sentences spontaneously in daily communication.

Some research indicates that learning to read during the school years accelerates both language and memory development for children with Down syndrome in the way that would be predicted from research with typically developing children.^[19] Children with Down syndrome included in mainstream classes are likely to achieve higher levels of literacy than those in special schools - by more than 3 years progress - and will be ahead by more than 2 years in spoken language skills. The two gains are probably linked as the included children read and write everyday as they work in school, even if they cannot do so independently, as they have classroom assistants.

The importance of reading and writing activities, linked to the children’s level of language knowledge, the curriculum and social needs, throughout primary and secondary school, cannot be over emphasised. The benefits may be greatest for the children who become independent readers, but the benefits for the children who need considerable support for reading and writing are also likely to be significant.

Conclusion

The evidence from available research does suggest that: -

Improving the quality, quantity and sensitivity of the talk to children with Down syndrome does help - this would be the kind of change encouraged by a scheme such as the Hanen programme [TODO: references 128] - and suggested by the review of influences on the language development of all children at the start of this module.^[20–22]

This type of language experience approach is not sufficient on its own but needs to be accompanied by teaching specific vocabulary and sentences, using techniques which encourage the child to imitate production.^[2],^[3]

Targeted work does improve phonology from as early as 3 years for speech sound and word production and as early as the first year for listening, sound discrimination and sound copying practice.^[6–8]

Signing is an important aid to accelerating early language learning but the emphasis should always be on speaking and sound and word production work are important at the same time. Reading is also of fundamental importance to developing effective speech and language skills - both reading to children with Down syndrome and teaching them to read.

This module has shown that we know a great deal about how children learn their native language and become competent talkers. It also illustrates that we know a great deal about the language learning difficulties of children, adolescents and adults with Down syndrome and how to help them to overcome them.

The important principles, which should guide intervention and daily talk with children and adults, have been set out. The practical modules provide detailed plans to follow, including assessment lists so that the gestures, sounds, words and grammar a child is using can be assessed in order to start activities that are appropriate for each individual child. Ideally each child should have the support of a speech and language therapist but as this is not always available, the practical guidance should allow a parent, teacher or classroom assistant to effectively improve the spoken language skills of their child without other help if necessary.