logo

Video Modelling and Autism Ranking: Insufficient/Mixed evidence

Video Modelling

Video modelling is a method of teaching in which an individual learns a behaviour or a skill by watching a video recording of someone (the model) demonstrating that behaviour or skill.

The model can be someone else (such as a parent or teacher, a peer or sibling) or it can be the individual himself (when the process is called video self-modelling).

The video is edited to show the model demonstrating the correct way to behave in specific situations, such as how to engage in conversation with other people.

The supporters of video modelling claim that it has been used to teach a wide variety of social and functional skills, such as how to interact with other people or how to buy things. They also claim that video modelling can be used to teach an individual how to apply previously learnt behaviours and skills in new settings.

Video modelling is sometimes used alongside other behavioural techniques or as part of other interventions, such as social stories or visual schedules

Our Opinion

There is a small amount of high quality research evidence (one small randomised controlled trial and one small controlled trial) on the use of video modelling as an intervention for autistic children but the results of that research are mixed. 

There is a considerable amount of low quality evidence (more than 70 single-case design studies with three or more participants) which suggests that video modelling may be an effective way to teach some autistic children a variety of different behaviours and skills in a variety of contexts.

There is insufficient evidence to determine if video modelling is an effective way to teach autistic adults any behaviours or skills.

We believe that video modeling may be a useful intervention for some autistic people but more research is required. That research should use more scientifically robust, experimental methodologies with larger numbers of participants. It should also aim to identify which elements in video modelling, if any, are effective for teaching skills to different groups on the autism spectrum.

Disclaimer

Please read our Disclaimer on Autism Interventions

Aims and Claims

Aims

The main aim of video modelling is to teach an individual to change his or her behaviour and/or to learn new skills. For example, it may be used to teach an autistic child how to use the toilet by himself/herself or how to engage in conversation with other people.

According to Mason et al (2016), video modelling is based on social learning theory. The idea is that the person watching the video will learn by observing and imitating the person in the video.

Claims 

The supporters of video modelling claim that it has been used to teach a wide variety of social and functional skills, such as how to interact with other people or how to buy things. They also claim that video modelling can be used to teach an individual how to apply previously learnt behaviours and skills in new settings.

For example Nikopoulos claims that video modelling can be used to teach a wide range of behaviours and skills including generalised purchasing skills, daily living skills, conversational skills, social language skills, generative spelling, perspective taking, socially relevant behaviours, play skills, and toilet training skills. He also claims that it can be used to reduce disruptive transition behaviours.

Nikopoulos claims that, independently of whether video modelling has been examined alone or in combination with other procedures, it shows great promise as an efficient and effective instructional tool for autistic people. Videotapes can become individualised for any child, and because their use in treatment can encourage a structured teaching style, they may become an important means for parents and educators to enhance their children's functional skills that does not require extensive training.

Source: personal correspondence with Research Autism, 2009

Audience

Video modelling is designed to help a range of people learn new skills and behaviours. In practice, it is most often used as an intervention for autistic children and young people.

Some researchers have suggested that video modelling is particularly suitable for autistic people because of their preference for visual processing and learning approaches. (Rayner et al, 2009).

Some researchers have also suggested that, for video modelling to be successful, the participants need one or more prerequisite skills such as memory, paying attention and imitation. However Fragle (2014) has noted that there is currently insufficient research to prove if this is true or not.

Key Features

Description

Video modelling is a method of teaching in which an individual learns a behaviour or a skill by watching a video recording of someone (the model) demonstrating that behaviour or skill.

The model can be someone else (such as a parent or teacher, a peer or sibling) or it can be the individual himself when the process is called video self-modelling (VSM).

The video may show the whole of the model’s body, it may show part of the body (such as the hands) or it may show the scene as viewed from the perspective of the person watching the video (point-of-view modelling).

The video may be presented by itself or as one element of a multi-media presentation that includes text, music, pictures and video.

The video is edited to show the model demonstrating the correct way to behave in specific situations, such as how to engage in conversation with other people.

The participant may be asked to watch the whole of the video or they may be asked to watch the specific segments of the video that show them how to do a specific task or activity.

The participant is asked to watch the video shortly before they are asked to undertake the activity being modelled. An instructor may encourage the participant to pay attention to particular parts of the video while it is playing, either in person or via a recorded narration or voice over.

Video modelling is different to live (in vivo) modelling because the person modelling the behaviour is not present (except in the case of video self-modelling).

Video modelling is sometimes used alongside other behavioural techniques or as part of other interventions, such as social stories or visual schedules

Guidelines

Nikopoulos suggests the following guidelines for video modelling.

1. Make sure that the child can watch video/TV for at least two consecutive minutes.

2. Prepare a short videotape (initially 30 to 40 seconds) showing a model whilst he/she engages in some simple behaviours.

3. The number of different behaviours to be shown in the video needs to be gauged for a particular child experimentally.

4. At the initial stages of the intervention, the setting viewed in the videotape should be the same as the setting in which the child will subsequently demonstrate the imitative behaviour.

5. The child should be allowed to watch each video clip at least once before he/she tries to demonstrate the modelled behaviour. If he/she fails, then the same video clip is presented again; this should be done at least three times.

6. Depending on the target behaviour, the child must be allowed to have between 1 and 3 minutes to demonstrate the modelled behaviour.

7. Finally, the imitative behaviour should be performed in the absence of any previous video-watching in other settings or in front of other people.

Source: personal correspondence with Research Autism, 2009.

Cost and Time

Cost

The costs of using video modelling will depend on a number of factors including the equipment used, the model used and the length and frequency of the intervention.

The equipment may include a camcorder, a tripod, a video player, a television and a PC or Mac, as well as software to transfer the video recordings from the camcorder to the computer.

There may also be a cost for the model, although the model will be used only once whilst the video recording will be used for as many times as it may be needed.

Time

The amount of time it takes to use video modelling will depend on a number of factors including the time it takes to record the video, the length of the video (usually between 45 seconds and five minutes), the frequency with which it is played and re-played (usually between two and five times) and the needs of the individual watching it. In some cases, video modelling will require many hours of work each day and be implemented over many years.

Risks and Safety

Hazards

There are no known hazards for video modelling.

Contraindications

There are no known contraindications (something which makes a particular treatment or procedure potentially inadvisable) for video modelling.

Suppliers and Availability

Suppliers 

In practice, video modelling may be implemented by a variety of people, including parents and carers, as well as professionals such as teachers or behaviour analysts.

Credentials

There are no nationally recognised qualifications for people providing video modelling. However, you may find it helpful to obtain supervision from a professional who has some experience of video modelling.

History

Live modelling has been in use as a training tool since at least the 1920s. Symbolic forms of modelling (such as audio-tapes, videotapes and film) followed thereafter.

The first reported evidence for the use of video modelling as a treatment for an autistic child was reported in 1982 by Steinborn and Knapp. They used a behavioural training programme and a classroom-based model of a traffic environment in order to teach an autistic child pedestrian skills.

Specifically, they used video recordings to familiarise the child with traffic at local intersections. Since then video modelling has been used by a variety of practitioners to teach a range of autistic people a variety of skills and behaviours.

Current Research

We have identified more than 70* studies of video modelling as an intervention for autistic people which were published in English-language, peer-reviewed journals and which included three or more participants. 

These studies included more than 350 individuals aged from pre-school children through to adults, although the majority of studies looked at younger children.

Some of the studies used an adult model (teacher or parent), some used a peer of the same age (sibling or classmate), while some of the studies used the participants as their own models.

Some of the studies looked at video modelling as a standalone intervention. Some of the studies looked at video modelling combined with other interventions (such as social stories or a combination of other behavioural techniques).

Some of the studies compared different forms of video modelling (such as video self-modelling) with each other. Some of the studies compared video modelling with other forms of modelling (such as live modelling) or with other types of intervention (such as pivotal response training or reciprocal imitation training).

The studies were conducted in a variety of locations including schools, clinics and family homes.

The vast majority of these studies (more than 70) reported positive benefits from video modelling while a minority of studies (more than 10) reported no benefits or very limited benefits. For example

  • Some of the studies (such as Charlop-Christy and Milstein, 1989; Sherer et al, 2001; Tetreault et al, 2010) reported increased conversational skills in some autistic children.
  • Some of the studies (such as Nikopoulos and Keenan, 2004; Nikopoulos and Keenan, 2007; Plavnick et al, 2015) reported improved social interaction in some autistic children.
  • Some of the studies (such as Akmanoglu, 2015; Axe and Evans, 2012; Judah and Evans, 2012) reported improved facial recognition in some autistic children.
  • Some of the studies (such as Kim, 2016; Nikopoulos, 2007; and Ozen et al, 2012) reported increased play skills in some autistic children.
  • Some of the studies (such as Ayres et al, 2009; Meister et al, 2015; Shipley-Benamou et al, 2002) reported improvements in daily living skills in some autistic children.
  • Some of the studies (such as Cihak et al, 2008; Kellems and Morningstar, Epub; Van Laarhoven et al, 2012) reported improved vocational skills in some autistic adults.

* Please note: Because there are so many studies on this topic we have not included those studies with fewer than three participants in this section. However, you can find details of some of those studies, and other publications on video modelling, in our publications database.

Status Research

There are a number of limitations to all of the research studies published to date. For example

  • The overwhelming majority of studies consisted of single-case designs with small numbers of participants (mostly three, four or five participants).
  • Some of these single-case design studies used extremely weak methodologies (such as simple AB reversal designs) or were descriptive case studies only.
  • The two controlled studies had small numbers of participants (Kroeger et al, 2007 had 25 participants; Popple et al, 2016 had 18 participants) and the study by Kroeger et al was not randomised or blinded.
  • Some of the studies did not provide enough details about the participants, such as whether they had a formal diagnosis of autism, intellectual ability etc.
  • Most of the studies were limited to young autistic children, with very few looking at other groups (such as adults, females, individuals from ethnic minority groups).
  • Very few of the studies compared the different elements of video modelling (such as the type of model, the length of the video and the number of times it should be repeated) to determine which, if any, are the most effective elements for which groups of people.
  • Some of the studies used video modelling alongside other techniques or interventions, making it difficult to know if any effects were caused by the video modelling
  • Very few of the studies compared video modelling with other techniques or interventions which are designed to achieve similar results, such as social stories.
  • Very few of the studies allowed the participants to get used to the video technology before evaluating any effects it might have had. It is therefore not clear if any effects were due to the video modelling or the introduction of the technology.
  • Many of the studies did not state if the video modelling provided any beneficial effects which lasted in the medium to long term.
  • Many of the studies did not state if the video modelling provided any practical benefits in real world settings.
  • Most of the studies did not involve autistic people in the design, development and evaluation of the research.

For a comprehensive list of potential flaws in research studies, please see ‘Why some autism research studies are flawed’

Future Research

Summary of Existing Research

Individual studies

There is a small amount of high quality research evidence (one small randomised controlled trial and one small controlled trial) on the use of video modelling as an intervention for autistic children but the results of that research are mixed.

There is a considerable amount of low quality evidence (more than 70 single-case design studies with three or more participants) which suggests that video modelling may be an effective way to teach some autistic children a variety of different behaviours and skills in a variety of contexts.

There is insufficient evidence to determine if video modelling is an effective way to teach autistic adults any behaviours or skills.

Research Reviews

There have been a number of scientific reviews of video modelling as an intervention for autistic individuals. The majority of these concluded that video modelling is an effective intervention. For example, Acar and Diken (2012) reported,

“It was also noted that results of these studies showed that video-modeling was effective on teaching many behaviors or skills such as social skills, play skills, language and communication skills, functional skills, self-care skills, daily life skills. Based on the review process, it can be suggested that video-modeling can be used widely in practice on teaching various behaviors and skills to individuals with autistic disorders and other developmental disabilities.”

However, a minority of reviews have suggested that the evidence is less positive. For example, Sng et al (2014) reported,

“Based on the research reviewed, it would appear that video modelling was in the marginally effective range … in improving the conversational skills of children and adolescents with ASD, when applied to basic skills such as initiating and responding”.

Recommendations for Future Research

Future studies should

  • Use more scientifically robust, experimental methodologies (such as randomised controlled trials) with larger numbers of participants.
  • Provide more details about the participants, such as whether they have a formal diagnosis of autism, their level of intellectual ability etc.
  • Involve a wider range of participants (such as adults, females, individuals from ethnic minority groups).
  • Identify if there are any prerequisite skills (such as the ability to pay attention) that may make some individuals more likely to benefit from video modelling.
  • Examine which elements of video modelling (such as the type of model, the length of the video and the number of times it should be repeated) are most likely to benefit autistic people.
  • Compare video modelling with other interventions which are designed to achieve similar results, such as social stories
  • Identify if video modelling can be used to teach skills not previously examined, for example, study skills
  • Allow the participants to get used to the video technology before evaluating any effects it might have.
  • Identify if video modelling has any beneficial effects in the medium to long term
  • Identify if video modelling has any beneficial effects in real world settings
  • Involve autistic people in the design, development and evaluation of those studies

Studies and Trials

This section provides details of scientific studies into the effectiveness of video modelling for autistic people which have been published in English-language, peer-reviewed journals.

If you know of any other publications we should list on this page please email info@informationautism.org

Please note that we are unable to supply publications unless we are listed as the publisher. However, if you are a UK resident you may be able to obtain them from your local public library, your college library or direct from the publisher.

Related Studies and Trials


Other Reading

This section provides details of other publications on this topic.

You can find more publications on this topic in our publications database.

If you know of any other publications we should list on this page please email info@informationautism.org

Please note that we are unable to supply publications unless we are listed as the publisher. However, if you are a UK resident you may be able to obtain them from your local public library, your college library or direct from the publisher.

Related Other Reading


Updated
17 Jun 2022
Last Review
01 Sep 2017
Next Review
01 Dec 2023