The big idea
A prompt is extra help, added to the natural SD, that makes the correct response more likely. Prompts let you reinforce correct responses that would otherwise rarely happen. They are temporary by design: the plan for removing them (see G.8) starts on the day you choose them.
Prompts come in two families. Response prompts act on the learner’s behavior: you tell, show or guide. Stimulus prompts change or add to the materials so the correct choice stands out.
Breaking it down
Response prompts
Listed roughly from least to most intrusive:
- Verbal: instructions, hints or cues (“Turn on the water”).
- Gestural: pointing, nodding or looking toward the correct item or action.
- Model: demonstrating the response for the learner to imitate (see G.9). This needs an imitation repertoire.
- Physical: partial guidance, such as a touch at the elbow, or full hand-over-hand guidance.
Physical prompts are the most reliable but also the most intrusive. Some learners dislike being touched, and families may have cultural or personal preferences about touch. Get consent and watch for signs of discomfort.
Sources classify a few prompts differently. Cooper, Heron and Heward list response prompts as verbal instructions, modeling and physical guidance, and treat pointing to or tapping the correct item as a movement-based stimulus prompt. Many other texts call pointing a gestural response prompt, and exam items usually use the four-part list above.
Stimulus prompts
- Within-stimulus prompts change a feature of the SD or S-delta itself: the correct word printed larger, the correct card placed closer, the S-delta faded to pale gray.
- Extra-stimulus prompts add a separate stimulus: an arrow, a picture beside a printed word, a highlighted box, a dot on the correct button.
Within-stimulus prompts often fade more smoothly because the learner is already attending to the relevant feature. Extra-stimulus prompts can pull attention away from it: a learner may read the picture beside the word and never look at the word.
Prompting systems
Most-to-least (MTL) prompting starts a new skill with the controlling prompt, the one that reliably produces a correct response (often physical), and steps down to less help as the learner meets criteria. If errors increase, it steps back up.
- Fits: new skills, early learners, safety skills, chains where an error would be practiced, and learners who react to errors with problem behavior.
- Risk: slower to show independence; without a planned fade, the learner learns to wait.
Least-to-most (LTM) prompting gives the learner a chance to respond alone, then adds more help step by step after an error or no response within a set wait (often 3 to 5 seconds).
- Fits: skills the learner partly knows, maintenance and generalization checks, and learners who respond well to light prompts.
- Risk: more errors, which get practiced and can frustrate.
Comparisons of the two (for example, Libby and colleagues, 2008) tend to find fewer errors with MTL, while LTM can be quicker for learners who need little help. Decide from the learner’s data, not habit.
Simultaneous prompting delivers the SD and the controlling prompt at the same moment on every teaching trial, so the learner responds correctly with help. Separate probe trials, run without any prompt before each session, test whether the learner can respond alone. When probes reach criterion, the skill is mastered. There are no fading steps to manage.
Errorless learning
Errorless learning is an approach, not a single procedure: arrange help so the learner makes few errors. MTL, simultaneous prompting, time delay that begins at 0 seconds (see G.8) and stimulus fading are all errorless methods. They reduce error practice and the frustration that comes with it, but they depend on fading the prompt.
Selecting and evaluating prompts
Choose the least intrusive prompt that reliably works for this learner and this skill, and favor prompts that are easy to fade. Record the prompt level on every trial, not just correct or incorrect. Change the plan when:
- errors stay high at a given level (the prompt is not controlling);
- the learner waits for help they do not need (prompt dependence, see G.8);
- progress stalls at one level for several sessions.
Seeing it in practice
| Learner and skill | Prompt or system | Why |
|---|---|---|
| Nia, 4, brushing teeth (new chain) | MTL: full physical, partial physical, model, gesture | New skill; errors upset her and get practiced |
| Diego, 10, facts he half knows | LTM: independent, verbal hint, model | Some independence already |
| Mei, 7, sight words | Within-stimulus: target word printed larger, then equalized | Keeps attention on the word |
| Hamza, 34, clocking in at work | Extra-stimulus: arrow sticker on the right key | Quick support; fade by trimming the sticker |
| Ruth, 5, naming animals | Simultaneous prompting with daily probes | Few errors and no fading steps |
Common mix-ups
- Stimulus or response prompt? Ask what the prompt acts on: the materials (stimulus) or the learner’s behavior (response).
- Within or extra? If the prompt is part of the stimulus being taught, it is within-stimulus. If it is added beside it, it is extra-stimulus.
- MTL and LTM differ at the start of each trial. MTL begins with help; LTM begins with an independent opportunity.
- More help is not always safer. Unneeded physical prompts slow independence and can be aversive.
Exam tip: If the stem stresses preventing errors, a brand-new skill or problem behavior after mistakes, choose most-to-least or another errorless method. If it stresses a learner who already knows part of the skill, or finding out what they can do alone, choose least-to-most.
Quick recap
- Response prompts: verbal, gestural, model, physical. Stimulus prompts: within-stimulus and extra-stimulus.
- MTL starts with the controlling prompt and steps down; LTM starts independent and steps up.
- Simultaneous prompting pairs SD and prompt on teaching trials and uses unprompted probes to test learning.
- Errorless methods reduce errors but must include a fading plan.
- Record the prompt level on every trial so the system can be evaluated.
Further reading
- Cooper, J. O., Heron, T. E., & Heward, W. L. (2020). Applied behavior analysis (3rd ed.), chapter 17.
- Libby, M. E., Weiss, J. S., Bancroft, S., & Ahearn, W. H. (2008). A comparison of most-to-least and least-to-most prompting on the acquisition of solitary play skills. Behavior Analysis in Practice, 1(1), 37-43.
- MacDuff, G. S., Krantz, P. J., & McClannahan, L. E. (2001). Prompts and prompt-fading strategies for people with autism. In C. Maurice, G. Green, & R. M. Foxx (Eds.), Making a difference: Behavioral intervention for autism. Pro-Ed.
Key terms
- Response prompt
- A prompt that acts on the learner's response: a verbal instruction or hint, a gesture, a model to imitate or physical guidance. It is added to the natural SD and later faded.
- Stimulus prompt
- A prompt that changes or adds to the antecedent stimuli so the correct response is more likely, such as making the correct item larger, brighter or closer, or adding a picture or arrow. Stimulus prompts are either within-stimulus or extra-stimulus.
- Within-stimulus prompt
- A stimulus prompt that changes a feature of the SD or S-delta itself, such as its size, color, position or intensity, so the correct choice stands out.
- Extra-stimulus prompt
- A stimulus prompt that adds a separate stimulus to the task, such as an arrow, a picture, a colored dot or a highlight, to help the learner choose correctly.
- Most-to-least prompting (MTL, Decreasing assistance)
- A prompting system that begins with the most intrusive prompt needed for a correct response, often physical guidance, and moves to less intrusive prompts as the learner meets criteria. It limits errors and suits new skills.
- Least-to-most prompting (LTM, System of least prompts, Increasing assistance)
- A prompting system that gives the learner a chance to respond independently, then provides increasingly intrusive prompts after an error or no response within a set time. It suits skills the learner partly knows.
- Simultaneous prompting
- A procedure in which the controlling prompt is delivered at the same moment as the SD on every teaching trial, while separate unprompted probe trials, usually run before each session, test whether the learner can respond correctly alone.
- Errorless learning (Errorless teaching)
- An approach to teaching that arranges prompts or stimulus changes so the learner makes few or no errors, then gradually removes that help. Examples include most-to-least prompting, time delay that starts at 0 seconds and introducing the S-delta faintly before bringing it to full strength.
Check yourself
4 questions on G.7. Choose an answer to see whether you're right and why.
Show the answer
Answer: B. The size of the word itself is exaggerated, so the prompt is a change to the stimulus being taught: a within-stimulus prompt. An extra-stimulus prompt would add something separate, such as an arrow or a picture beside the word. Response prompts act on the learner's behavior, not on the materials.
Show the answer
Answer: B. A new skill and errors that lead to problem behavior both point to most-to-least prompting, which begins with enough help to prevent errors and fades it as Nia succeeds. Least-to-most begins with an independent attempt, which would allow the errors that upset her. Repeated verbal prompts may not control a skill she has never performed.
Show the answer
Answer: D. The prompt is given at the same moment as the SD on every teaching trial, and separate unprompted probes test what the learner has learned: that is simultaneous prompting. Progressive time delay would gradually lengthen the wait before the prompt. Least-to-most would start each trial with an independent opportunity.
Show the answer
Answer: D. Diego already knows many facts, and constant models have taught him to wait. Least-to-most prompting gives him a chance to answer before any help, so independent answers can be reinforced. Answer cards and extra models add help he does not need, and more reinforcement for prompted answers would strengthen the waiting.
Adds a step to your study record. Saved in this browser only.