The big idea
Modeling is demonstrating a behavior so that someone else will do it. The demonstration is an antecedent that evokes imitation: behavior that copies the form of the model, follows it closely in time and is controlled by it (see B.24). A good model can teach in seconds what instructions take minutes to describe.
Modeling works only when the learner attends to the model, can imitate, and is reinforced for imitating. Designing a modeling procedure means arranging all three.
Breaking it down
Prerequisites
Before relying on modeling, check that the learner:
- attends to the model, looking long enough to see the important parts;
- has generalized imitation: they copy new models that have never been trained;
- can perform the components of the skill, or close approximations.
If imitation is missing, teach it first. Start with simple motor actions, present the model, physically prompt the copy, reinforce it, and fade the prompt. Train many different actions until new models are imitated without training.
Types of modeling
- Live (in vivo) modeling: a person demonstrates in real time. Flexible and immediate, but it varies a little each time.
- Video modeling: the learner watches a recording of someone performing the skill, then gets a chance to perform it. Every viewing is identical, it can be replayed, and it can be edited to show only correct performance.
- Video self-modeling: the learner watches edited video of themselves performing the skill successfully.
- Point-of-view video modeling: the video shows the task from the learner’s own viewpoint, such as their hands making a sandwich.
- Peer modeling: a classmate or other same-age person demonstrates. Peers are relatable and are present in natural settings, which supports generalization (see G.15).
Video modeling has strong research support across many learners and skills, and some studies have found faster learning and better generalization than with live modeling. It does take time and equipment to record and edit.
Characteristics of effective models
Learners are more likely to imitate a model who:
- is similar to them in age, background or interests;
- is competent and has status the learner values;
- is seen receiving reinforcement for the behavior;
- shows the behavior clearly, in short steps, with the key features easy to see;
- is one of several models across settings, which helps the skill generalize.
For skills tied to fear or anxiety, such as a first dental visit, a coping model who struggles at first and then succeeds can work better than a flawless mastery model.
Designing the procedure
- Define the target behavior, and break complex skills into steps.
- Choose the type of model that fits the learner, the setting and the resources.
- Get the learner’s attention, then present the model.
- Give an immediate chance to perform.
- Prompt as needed, reinforce correct imitation and give feedback on errors.
- Fade the model, for example by showing it only after an error.
Modeling within behavioral skills training
Behavioral skills training (BST) combines four components: instructions, modeling, rehearsal and feedback, repeated until the learner meets a performance criterion. It is the standard approach for training staff and caregivers (see I.5) and for teaching safety skills to children. Modeling shows what correct performance looks like; rehearsal with feedback is what makes it reliable.
Evaluating modeling
Measure performance, not exposure. “Watched the video three times” is not an outcome; the number of steps performed correctly is. If performance does not improve, check:
- attention: did the learner watch the model?
- prerequisites: can they imitate, and can they do the components?
- complexity: does one model contain too many steps?
- consequences: is imitation reinforced?
Then probe the skill in the natural setting, without the model, to see whether it generalizes.
Seeing it in practice
| Setting | Learner | Model | Design notes |
|---|---|---|---|
| Preschool | Aiden, 4, joining play | Peer modeling | Two classmates show how to ask “Can I play?” and are praised |
| Clinic | Layla, 9, brushing teeth | Point-of-view video | Each step filmed from her angle; she watches, then brushes |
| High school | Mateus, 16, ordering at a café | Video self-modeling | An edited clip of his own successful order, viewed before outings |
| Group home | Rosa, a new staff member | BST | Supervisor explains, models, has her rehearse and gives feedback until 100 percent correct |
Common mix-ups
- Modeling versus imitation. Modeling is what the teacher does; imitation is what the learner does.
- Watching is not learning. A learner without attending or imitation skills will not benefit from a video, however good it is.
- Observational learning is broader. Learning from watching others’ behavior and its consequences, without a direct prompt to copy, is observational learning (see B.24).
- BST is more than modeling. Without rehearsal and feedback, modeling alone often leaves performance short of criterion.
Exam tip: If a learner watched a model but did not perform the skill, look first at prerequisites: attending and generalized imitation. If staff saw a demonstration but still make errors, the answer is usually to add rehearsal and feedback (BST).
Quick recap
- Modeling is a demonstration that evokes imitation.
- The prerequisites are attending and generalized imitation.
- Effective models are similar, competent, reinforced, clear and varied.
- Live, video, self, point-of-view and peer modeling each fit different needs.
- BST combines instructions, modeling, rehearsal and feedback.
- Evaluate by performance in natural settings, not by viewing.
Further reading
- Cooper, J. O., Heron, T. E., & Heward, W. L. (2020). Applied behavior analysis (3rd ed.), chapter 21.
- Charlop-Christy, M. H., Le, L., & Freeman, K. A. (2000). A comparison of video modeling with in vivo modeling for teaching children with autism. Journal of Autism and Developmental Disorders, 30(6), 537-552.
- Bellini, S., & Akullian, J. (2007). A meta-analysis of video modeling and video self-modeling interventions for children and adolescents with autism spectrum disorders. Exceptional Children, 73(3), 264-287.
Key terms
- Modeling
- Demonstrating a behavior so that an observer will imitate it. The model is an antecedent that evokes imitation, and it can be presented live, on video or by a peer.
- Imitation
- Behavior that copies the form of a model, occurs soon after the model and is controlled by it. In generalized imitation, a learner imitates new models that have never been directly trained.
- Generalized imitation
- Imitation of new models that have never been reinforced, after a history in which imitation of many other models was reinforced.
- Video modeling
- A modeling procedure in which the learner watches a video of someone performing the target behavior and then has a chance to perform it. The same correct demonstration is shown every time and can be replayed as needed.
- Video self-modeling (VSM)
- A form of video modeling in which the learner watches edited video of themselves performing the target behavior successfully.
- Peer modeling
- A modeling procedure in which a peer, such as a classmate of similar age, demonstrates the target behavior. Similar models in natural settings tend to be imitated readily and can support generalization.
- Behavioral skills training (BST)
- A training package made up of instructions, modeling, rehearsal and feedback, repeated until the learner performs the skill to a mastery criterion. It is a well-supported way to teach new skills to staff, caregivers and clients.
Check yourself
4 questions on G.9. Choose an answer to see whether you're right and why.
Show the answer
Answer: C. A model can only evoke imitation if the learner attends to it, and Aiden is not watching. Showing the video more often does not help if he looks away each time. Similarity and visible reinforcement for the model matter, but only once he is attending.
Show the answer
Answer: A. Instructions and modeling alone often leave performance short of criterion. Behavioral skills training adds rehearsal and feedback, repeated until staff perform the procedure correctly. A more senior model, a written summary or a video repeats parts of training that have already been tried.
Show the answer
Answer: A. A similar peer who copes with nerves and is reinforced for presenting combines several features of effective models: similarity, a coping style that fits an anxiety-related skill, and visible reinforcement. A flawless professional is less similar to Lucía. A description is an instruction, not a model, and a model with no visible consequence is less likely to be imitated.
Show the answer
Answer: B. Modeling depends on an imitation repertoire, and Malik does not have one yet. Teaching imitation directly, by prompting the copied action after the model and fading the prompt across many exemplars, builds that prerequisite. Changing the type of model or repeating it more often does not supply the missing skill.
Adds a step to your study record. Saved in this browser only.