This technique facilitates positive engagement with children and adults, many of whom are on the autistic spectrum, and with whom communication is often difficult.

Phoebe Caldwell (UK)

This article introduces an approach that allows us to engage with children and adults with whom we find it hard to get in touch. Many are on the autistic spectrum. We find it hard to communicate with them and they with us. Many are very distressed.  In the context of autism, I use the term ‘distressed behaviour’ in preference to ‘challenging behaviour’, since outbursts in autism are triggered by sensory overload and the word ‘challenging’ sets up oppositional expectations.


We communicate in two different ways. Most of what we are consciously aware can be called Functional Communication, informing each other of our needs, on the level of ‘Do you want a cup of tea?’, or sharing more sophisticated information. In people who are non-verbal, functional communication may be assisted by sign systems such as Makaton (Grove and Walker 1990) or PECS (Picture Exchange Communication System). What we are all less aware of, is how we inform and monitor each other’s emotional states all the time.

This Emotional Engagement is expressed through our body language, not so much by what we say or do, as how we do or say it. In conversation, we watch our partner to check whether what they are saying matches how they are expressing it. Basically we both want to know if we can trust each other. And with people who are non-verbal, there is an evident difference in the affective state of someone who is flapping their hand gently and one who is thrashing the air with it: same gesture but in the first instance we know they are relatively calm, in the latter, that they are expressing severe distress.

In the 1980s, Geraint Ephraim,  Consultant Psychologist at Harperbury Hospital, introduced the idea of using body language to communicate with people whose ability to communicate was impaired by severe intellectual disabilities. This approach was so successful that it was taken up by Nind and Hewett who named it Intensive Interaction. (Nind and Hewett 1994). I worked for four years under the supervision of Ephraim during tenure of a Joseph Rowntree Foundation Fellowship and subsequently (although not exclusively), specialised in using it with children and adults on the severe end of the autistic spectrum. My experience as a Practitioner using Intensive Interaction extends over twenty years and I have worked with literally hundreds of adults and children, many of whose behaviour was extremely distressed.

The paradigm underlying Intensive Interaction is that of the Infant-Mother interaction, in which the infant initiates a sound or movement or rhythm and the mother responds in an imitative way. Once the baby’s initiative is sufficiently confirmed they are able to move on and try out something else. It is crucial to emphasise that in using Intensive Interaction we are not in any way infantilising our conversation partners, since for all of us, this non-verbal dialogue is a primary communication pathway, laid down in babyhood but remaining with us all our lives. We continue to look for underlying confirmation throughout adulthood.

Based on imitation, it has recently become clearer why Intensive Interaction is so successful in attracting the attention of a conversation partner. Much research is now devoted to the Mirror Neuron system, a network of nerve cells in the brain that recognise actions made by other people and fire off a sensory motor response. (Rizzolatti et al. 1995). Seeing another person yawning, triggers a tickling sensation in one’s own jaw, even if not an outright yawn. This mechanism can also apply to emotions: it is easy to feel dragged down in the company of someone who is depressed.

Intensive Interaction

People who are unable to communicate develop ways of interacting with themselves in such a way that brain and body engage in an internal conversation. The brain sends messages to the body - for example, ‘flap your hand’ - and listens to the feedback that such an activity generates in the form of physical sensation. Instead of drawing confirmation from the world outside, the brains of such people are self-confirming in a repetitive loop. In effect each individual develops their own personal language of stimuli that have significance for them and to which they listen. They are having a self-generating conversation. If we want to communicate with them, the first question we have to ask ourselves is, ‘how is this person talking to themselves?’

We start with ‘observation’ - but need to think of observation not so much as a period of  shadowing  but rather as the development of an ongoing picture of what our conversation partner is doing now, this minute. Particularly, we want to avoid the pitfall of drawing up a list of activities we ‘do’ with them, since it is absolutely essential that our responses are contingent, not only to their initiative but also as to how this initiative is made, since it is the ‘how’ that will allow us to tune into their affective state. I have to empty myself of any behavioural expectations and learn to ‘be with’ this person as they are at present, using their initiatives, gestures, rhythms and sounds to respond in a way that has meaning for them.

Because interactions are all personal, each practitioner tends to develop their own style. Personally, since I find that it is easy to block intuitive perception with information that is not relevant to our current interaction, I tend not to take elaborate histories or go through long preliminary observations. My only qualification is that I do wish to know if I am likely to be attacked and under what circumstances, so that I can avoid distressing my partner.

As to the quality of our observation, we need to think in terms of a ‘listening with all our senses’ to any minute movements, gestures or sounds and especially, to focus on what this person is doing at this particular moment. This will be our way in to our partner’s inner language, our aim being to draw their attention from their solitary inner world onto ourselves in the world outside, so that their sensory monologue becomes a dialogue, an interactional conversation that we can now share. At the same time we need to be aware that a person’s attention may be focused on as little as a small click of the tongue or even their own breathing rhythm, activities that we overlook because they do not have significance for us.

Debbie, who has cerebral palsy and severe learning disability, sits in the day centre with her head down, staring at the floor. She is unresponsive and apparently uninterested in any form of activity. I am told that the physiotherapist has been trying for two years to get her to sit in a more upright position since her current posture is likely to cause her spinal problems. I am told that she is not doing anything that might be used to claim her attention but when I listen carefully it is apparent that she is making a small but regular sound by sucking her saliva. Together with three support workers whom I am teaching, we begin to answer her minute sounds. Within a few minutes Debbie’s head has come up and she is looking from one of us to another for responses and smiling. Her supporters continue to interact with her as shown, and within three weeks I am informed she now sits with her head up on a regular basis, looking around her to see what is happening. Hitherto she has not encountered any inputs from the world outside her that are sufficiently meaningful to have drawn her attention out of her inner world onto her environment.

This is a very simple interaction which not only claims Debbie’s attention - she raises her head - but also encourages her to engage with us, in the sense that she refers back to us, deliberately making a sound and waiting for our response. No longer self-confirming, she has developed an expectancy from the world outside herself.

Although Intensive Interaction is based on the mother-infant paradigm, our responses may be loosely attached, in the sense that we do not just imitating or mimicking, even though this is where we may start. But in order to move from attention to engagement, we need to be aware of our partner’s entire body language and may well use one element to respond to another. For example, we may answer a sound with a relevant touch or vice versa – but always keeping within the parameters of what is part of their repertoire. And sometimes, just as verbal language can be used to express complicated ideas, so the way that people without speech express how they feel is complex.

Pranve is on the autistic spectrum and hypersensitive to sound. His behaviour is often extremely distressed and he has outbursts where he attacks his mother or others who are close by, to the point where it has become difficult to find people who will support him. I am warned that when I arrive at the house he will probably either attack me or run away.

He lives on the edge of an airport runway and is distressed by the high frequency engine whine of incoming planes. He responds to these by lifting his head and rolling his eyes up to the left towards the sound. When he is anxious, he touches the fringe of the lampshade beside his chair and runs his hand down the stand. His father says that when he is angry he will sit in the hall and bang the door with his fist.

Pranve constantly rubs his hands together - a very common way that people on the spectrum self-confirm. He carries a ball of strings underneath his armpit and spends time pulling these out and sorting them.

Pranve also makes sounds, a particular rhythm, ‘er-er-er’, which turn out to be a pre-verbal version of ‘Where’s Charlene?’, his sister who no longer lives with the family. This is the only thing he has ever been known to say.

Since Pranve is easily disturbed, when I arrive, I take care not to invade his personal space before making contact with him. So when his mother opens the door I listen – and from another room hear, ‘er-er-er’, so I respond, ‘er-er, er-er-er?’, with a lift at the end, rather in the way one might say,  ‘Hello, how are you?’ He comes straight out and takes my hand and leads me to the sitting room. I ask him if I may sit down and he responds by pointing to the chair, so I know he understands at least simple speech.

I sit beside him and respond to each of his small sounds, tuning into how they make me feel, but altering the rhythm or pitch occasionally. I am answering rather than copying. At first he is half turned away from me but he gives me his hand which I shake in time to the sounds we are exchanging. He becomes more interested and turns round to face me, laughing. He introduces new sounds and movements to which I respond. We are soon engaged in a complex non-verbal interactive conversation.

I draw the shape of his different sounds on his forearm and he leans forward and looks with interest, then tries a different sound to which I respond with a shape that reflects its rhythm and pitch.

At one stage I become over-confident and move in when he is not looking. Immediately Pranve thumps my arm, but quite gently. He is telling me that he cannot cope if something happens unexpectedly. Now he is anxious he goes into his lamp stand routine, a feature of his language that I do not fully understand until I revisit our interaction on video. He wants me to confirm this sequence of touching the fringe and running a hand down the stand and tries to guide my hand to meet his need. He is clearly disappointed when I miss this. However we are able to return to using his sounds to interact. Eventually our session comes to an end when I fail again to pick up on his anxiety routine. He pushes me away gently and we have a break. Then he goes into the hall and bangs the door. I respond by banging my feet on the floor. He laughs and throws his ball of string into the sitting room, a strategy which means he has to come back to us to retrieve it. He comes in, spots his mother and goes over and hugs her.

After lunch, he drags a spring-backed chair over in front of me and bounces on it, turning round to me inviting me to bounce him. I do this every time he makes a sound. He begins to hum four notes, as it turns out,  the first line of the nursery rhyme, ‘Baa Baa Black Sheep’. After trial and error he gets the words and then goes on to produce the tune and rhythm of the second line, his face and jaw working for some considerable time before he can make the necessary muscular movements. As he places his head at different angles one can see the meaning of the phrase, ‘getting one’s head round something’. His jaw wobbles with the effort but eventually he sings these two lines confidently.  His family and the speech therapist who is present are astonished: while he must have heard this song when he was younger, to their knowledge he has never before said anything before except the phrase, ‘where’s Charlene?’

During the three hours I am with Pranve, apart from the one time that I startle him, he shows no aggressive intent and is clearly delighted with our interactions. But also, after about twenty minutes, he is no longer reacting to the scream of the planes passing over the house, so close their wheels are down for landing. His interest in our conversation is overriding his hypersensitivity to the high frequency whines of the jet engines. (Caldwell 2006a)

I have presented this intervention in detail because it illustrates some of the interwoven subtleties of a body language interaction. One is not just working with repetitive behaviour but more with what might be thought of as the total ecosystem of a person’s life and how they interact with their surroundings. More than ever this applies to people on the autistic spectrum since they are so sensitive to their environment.

Rather than focusing on the physical sensations from their bodies, more able people on the autistic spectrum may have adopted stimuli from their environment on which to concentrate. They may watch a particular video: Thomas the Tank Engine is a particular favourite. Others will take refuge in particular phrases. Finding a way to respond to these is difficult since echoing back the phrase simply draws one into the loop - but it can help to respond with the rhythm of their phrase rather than using the same words again. The brain appears to recognise this without being thrust back into the echolelic loop, drawing attention on to the world outside. In my experience they can be so relieved to be released that they will hug their partner.

And when we are using Intensive Interaction with people with autism, practitioners find that once they start to engage, they relax, their whole demeanour and posture change. Eye contact increases, they start to look round, they can generalise and copy and are interested in their conversation partner in a way that is not generally accepted to be typical of people on the autistic spectrum. And contrary to the idea that people on the spectrum have a deficit in their mirror neuron system (Ramachandran 2011), which might account for the communication difficulties they have, they always appear to recognise and respond to gestures and sounds, provided they are already part of their significant repertoire and offered within the field of any physical limitations.

Intensive Interaction can be applied across a wide field of disability. For example, Hart writes about his extensive experience of using Intensive Interaction to work with people who are deaf-blind (Hart 2008). He highlights the need for care staff to learn the capacity to ‘feel’ the world from a tactile perspective, developing ‘communicative landscapes’ to capture the attention of their deaf-blind partner in order to negotiate joint activities.

Intensive Intervention is not mysterious; it is a technique that applies the well-known psychological theory of imitation. Its general principles can be learned. Perhaps the easiest way to learn is to watch it being done and to this end there are a number of films available. But each time, no matter how experienced we are, we are beginners, in that we have to learn the significant language for a specific individual from our conversation partner. Inevitably there will be some trial and error: our partner will latch on more quickly to our using certain aspects of their language than others. For example, someone who is biting pieces of jigsaw may be unmoved when we try replicating his activity, since the feedback he is giving himself is that of pressure in the area of the mouth rather than visual. In this case he responds very quickly to the application of vibration in this area. (Caldwell 2009)

In this brief introduction, Intensive Interaction may sound complicated but in fact one does not have to be an expert in order to be effective. What is necessary is that we lay aside our own agenda, respond to our partner’s body language and allow intuitive awareness to lead us into and through non-verbal exchange. It is particularly effective with people on the autistic spectrum who are struggling with a sensory environment that is behaving like a kaleidoscope, where the pattern never settles. This instability can appear as life- threatening: responding in terms that are meaningful to the brain confirms what the person is doing. It offers what Josh describes as a ‘Delicious Conversation’ when he is losing his grip on sensory reality (Caldwell, Personal Communication). Donna Williams speaks of the relief when in the midst of sensory chaos she recognises an element of her own language. She says that it is like having a life belt thrown to her in a stormy sea. (Williams 1993)

When I am engaging with people with behavioural distress, I need to address two questions. The first asks, ‘what do I do now when I am being attacked, or my partner is self-harming?’ And the second is, ‘why does he or she feel the need to do these things?’ So when I use Intensive Interaction, it is as part of a dual approach. On one hand, I shall be looking to reduce the triggers to sensory distress such as the sensory hyper- and hypo-sensitivities, emotional overload and the difficulties caused by failure to process speech – and on the other hand, I shall be looking to increase signals that the brain can process easily, such as use of body language combined with strong proprioceptive input.

Some supporters will raise the objection that they feel silly engaging in behaviour that they may view as childish, or as age inappropriate for their partner. If we feel silly it is because we are centred on ourselves and not in our partner: we need to shift our attention to building a relationship through signals that their brain can take on board and process, rather than trying to frogmarch them into a world that is at best confusing and may present as terrifying. (Jolliffe et al 1992).  We need to ask ourselves: ‘What is it that improves the quality of my partner’s life?

Using Intensive Interaction to construct an Autism-friendly Environment

People with autism live in a sensory maelstrom (Ramachandran 2011).  It is difficult for them to know what they are doing.  At a non-verbal level, Pranve is confirming himself by activities that are hard-wired-in elements of his body language such as hand movements and string sorting, that help him construct a landscape that has meaning for him. Barron tells us that when he repetitively switched the lights on and off it gave him a wonderful sense of security since it was exactly the same each time. In an unpredictable world he knew what was going to happen (Barron and Barron 1992). Grandin tells us that she used to ask the same question over and over again in order to hear the predictable answer (Grandin 1992)

One of the reasons Pranve attacks people is if something happens that he has not foreseen. Already on a knife-edge because of his hypersensitivity to high pitched sounds, he becomes sensorily overloaded and his autonomic nervous system tips him into what Ramachandran terms an ‘autonomic storm’ (Ramachandran and Oberman 2006), an experience that is both confusing and can be extremely painful. Touching the lamp-stand in two different ways helps him to orientate himself, giving himself signals that confirm what he is doing in an environment in which he is becoming sensorily overloaded. Here is something that has meaning for him. He is unhappy when he did not get confirmation of this routine from me.

As Pranve becomes more relaxed his brain finds it easier to organise his muscular responses and he says clearly, ‘Where’s Charlene? ‘  instead of muttering the rhythm and then goes on to astonish his family by introducing the nursery rhyme.

And as Pranve’s parents become more proficient in using his body language to communicate with him, his behaviour becomes less stressed and he is able to return on a part-time basis to the day centre from which he had been excluded. Two months later his mother tells me that while he has the odd off-day, on the whole they could now interact with him and manage his behaviour. And he is calmer.

Perhaps the effect of using Intensive Interaction is summed up by a teacher following a first meeting intervention with Olly, who is seventeen and deeply isolated and self injuring. He wanders around by himself twiddling and rubbing his hands together. Trying to engage him through these actions had failed, so I used his breathing rhythm to focus his attention. He kept on and on coming back to me. His teacher said, ‘I have never seen Olly want to be with people that much’ (Caldwell, 2009). Wanting to be with other people, desiring relationship is what I hope to achieve for my conversation partners.

How well does Intensive Interaction work?

Intensive Interaction is about learning and using the language that has meaning for an individual to build a relationship with them. It is not a ‘cure’, in the sense that we do a few sessions and the person responds and so we can stop using it. If we do this, they will regress, because what we have done is to learn their language, use it with them and then walk away, slamming the door to relationship in their faces. All their distressed behaviour will return. We have to use it as a continuing way of communication and use it to explore and build on the relationship that it fosters. When the brain is no longer under processing pressure it begins to work more effectively on its own account, within the limitations of its learning disability if this is present. This is especially true for people with autism who are so vulnerable to environmental stress.

So success is dependent on maintenance. But even here we have a problem in matching outcomes to our expectations. For example, there is the question of getting our partners to conform to what society considers to be ‘normal’, without taking in to account the sensory distortions stemming from processing difficulties, the overloading input into the autonomic nervous system and the anxiety this induces. A classic example is that of taking people with autism shopping in a supermarket, where the sensory overload of high pitched hums, the lighting, the ‘pings’ and moving patterns of people, are for some a sensory nightmare. (Williams, D 1995).  However, even here it is sometimes possible to guide a partner through this kaleidoscope by constantly supplying sounds or movements or gestures that are part of their repertoire. These act as landmarks that the brain can focus on and exclude the avalanche of stimuli that threaten to overwhelm them.

But the first question that parents almost always ask is, ‘when will my child speak?’ For them, the ability to talk is paramount and success is measured by the time when they are able to engage with their child by talking to them. And while speech is sometimes released in non-verbal people when one is engaging with them through their body language, this is only an indicator of reduced stress level that allows the brain to function more effectively. Perhaps even more important is the desire to engage, and what has been expressed by a number of families as, ‘I have a happy child now.’

While there is a massive number of anecdotal reports from practitioners all round the world as to the effectiveness of Intensive Interaction there is rather less empirical evidence. This arises partly from the difficulties of generalisation from interactions that are unique to each intervention and also to some extent, divergence among practitioners as to how exactly they approach their practice. For example, some see Intensive Interaction as a sessional therapy applied on occasion, while others, in the words of a team leader working with a man who was showing extreme distressed behaviour find it far more effective if, ‘it just becomes part of the way we interact’ Caldwell, P. (2002b). However, in spite of the difficulties presented by standardisation, Zeedyk, et al. (2009) have analysed filmed Intensive Interaction interventions frame by frame and shown that, although the time-line may vary, there is always a measurable increase in eye contact, in social responsiveness and an increase in the desire for proximity. In Chapter 1 of her interesting book, ‘Promoting Social Interaction for individuals with Communicative Impairments’, Zeedyk (2008), reviews relevant papers, including an account of Intensive Interaction with a schoolboy who is very disturbed but uses his own body language to make contact with his teacher. While they are both still wearing the necklaces they are building, he smiles radiantly and physically links their necklaces together (Caldwell 2008).

Nind and Kellett (2002) show a significant decrease in disturbed behaviour in adults with learning disabilities when their support staff engage with them through corresponding actions. A survey commissioned by Mencap and the Department of Health, on communication with people with PMLD, Goldbart, J. (2010), finds that Intensive Interaction is one of the approaches most widely used. Over 85% of Speech Therapists in the survey were using it.

Fundamentally, Intensive Interaction is straightforward. Laying aside our own agendas, we start by looking and listening to what our conversation partners are doing, the physical feedback they are giving themselves and how they are doing it – and tuning in to how they feel, use our own body language to respond and build up non-verbal conversations and emotional engagement.


Barron. S and Barron, J., (1992) ‘There’s a boy in Here’. Simon and Schuster.

Caldwell, P.(2006a) ‘Finding you Finding Me’. Jessica Kingsley Publishers

Caldwell, P. (2002b) ‘Learning the Language’ Brighton. Pavilion Publishers

Caldwell, P. (2008) Intensive Interaction: Getting in Touch with a Child with Severe Autism’ in Zeedyk, S. Ed. ‘Promoting Social Integration for Individuals with Communicative Impairments’ Jessica Kingsley Publishers

Escalona, A., Field, T., Nadel, J. and Lundy, B.(2002) ‘Brief Report: Imitation effects on children with autism’. Journal of Autism and Developmental Disorders. 32, 141-144

Field, T., Field, T., Sanders, C. and Nadel, J.(2001) ‘Children with autism display more social behaviours after repeated imitation sessions.’ Autism 5. 3. 317-323

Goldbart,  J. (2010) ‘Communication and people with the most complex needs: what works and why this is essential’. Mencap Publications

Grandin , T. (1992) ‘A is for Autism’ Fine Take Productions and Channel 4 Television Corporation.

Grove, N., & Walker, M. (1990). The Makaton Vocabulary: using manual signs and graphic symbols to develop interpersonal communication. Augmentative and Alternative Communication, 6 (1),15-28.

Hart, P. (2008) ‘Sharing Communicative Landscapes with Congenitally Deaf-Blind People’ in Zeedyk, S. Ed. ‘Promoting Social Integration for Individuals with Communicative Impairments’ Jessica Kingsley Publishers

Jolliffe, T, Lansdown ,R. and Robinson, C. (1992) ‘Autism: A Personal Account’. Communication 26. 3. 12-19.

Nind, M. and Hewett, D. (1994) ‘Access to Communication’. London. David Fulton

Nind, M. (1996) ‘Efficacy of Intensive Interaction: Developing sociability and communication in people with severe and complex learning difficulties using an approach based on the caregiver-infant interaction’. European Journal of Special educational Needs 11, 1, 48-66

Nind, M. (1999) ‘Intensive Interaction and autism: a useful approach?’ British Journal of Special Education. 26 2. 96-102

Nind, M. and Kellett, M. (2002) Responding to individuals with severe learning difficulties and stereotyped behaviour. Challenges for an inclusive era. European Journal of Special Needs Education 3. 265-282

PECS Pyramid Educational Consultants UK, Pavilion House, 8 Old Steine, Brighton, BN1 1EJ

Ramachandran and Oberman (2006) ‘Broken Mirrors: A Theory of Autism’ Scientific American 295 (November) 39-45),

Ramachandran, V.S. (2011) ‘The Tell-Tale Brain’ Harvard University Press

Rizzolatti, G., Camarda, R., Gallese, V., and Fogassi, L. (1995) ‘Premotor Cortex and recognition of motor actions’ Cognitive Brain Research 3,  131-141

Williams, D. (1995) ‘Jam-Jar’ Fresh Film in association with Channel 4, UK

Williams , D. (1993) ‘My experience with Autism, Emotion and Behaviour’ Documentary with Connie Chung, USA

Zeedyk, S. ed. (2008) ‘Promoting Social Interaction for Individuals with Communicative Impairments’. Jessica Kingsley Publishers

Zeedyk, S, Caldwell, P. and Davies, C (2009) ‘How Rapidly does Intensive Interaction promote social engagement for adults with profound learning disabilities?’ European Journal of Special Needs Education, Vol 24. 2009, p.119-137

Intensive Interaction Training Films

Caldwell, P.(2002b) ‘Learning the Language’ Pavilion Press

Follows a three day intervention using Intensive Interaction to get in touch with a young man with very severe autism, using his body language, followed by discussions with Care Staff.

Caldwell, P.  (2004)‘Creative Conversations’ Pavilion Press.Intensive Interaction being used with people with multiple disabilities, mainly severe Cerebral Palsy.

Caldwell P (2007) ‘Reaching Ricky’ Teachers TV Made by ‘Available Light’ . Working with a child with autism in school.

Caldwell, P.(2009) Autism and Intensive Interaction: using body language to get in touch with children on the autistic spectrum’. Jessica Kingsley Publishers. Includes a twenty minute uncut Intensive Interaction intervention with an eight year old child, whom staff cannot make contact with and whom Caldwell has never met before. The film moves from initial rejection to total attention. Other interventions with children aged 3-18.

This article was first published on this site in 2011.