| name | prompt-fading |
| description | Use when selecting prompt types, designing prompt hierarchies, implementing systematic fading procedures, addressing prompt dependency, or choosing between most-to-least, least-to-most, time delay, stimulus fading, and stimulus shaping strategies. |
Prompt Fading
Prompts are supplementary antecedent stimuli that increase the probability of a correct response during skill acquisition. Prompt fading is the systematic removal of these supplementary stimuli so that the target behavior comes under the control of the natural discriminative stimulus. Effective prompt fading is the difference between a learner who can perform a skill independently and one who becomes prompt-dependent.
Prompt Types
Response Prompts (operate on the response)
Full physical prompt: Hand-over-hand or full body guidance through the entire response. Highest level of intrusiveness. Use when the learner has no approximation of the response and cannot imitate.
Partial physical prompt: Light touch, nudge, or guidance through part of the response. The clinician initiates the movement but the learner completes it.
Model prompt: The clinician demonstrates the target response. Requires the learner to have imitation skills in that modality (motor imitation for motor responses, echoic for vocal responses).
Gestural prompt: Pointing, motioning, head nods, eye gaze toward the correct response or materials. Less intrusive than modeling; relies on the learner attending to and interpreting the gesture.
Verbal prompt (vocal instruction): Telling the learner what to do. Ranges from full verbal ("Say 'I want cookie'") to partial verbal ("Say 'I want...'") to indirect verbal ("What do you say?"). Verbal prompts are among the hardest to fade because they are so natural for clinicians to deliver.
Stimulus Prompts (operate on the antecedent stimuli)
Positional prompt: Placing the correct item/response option closer to the learner or in a more prominent position. Easy to implement but easy for the learner to detect the pattern.
Visual prompt: Pictures, written text, video, visual schedules, or highlighted features that direct the learner to the correct response.
Within-stimulus prompt: Altering a feature of the target stimulus itself (e.g., making the correct answer larger, bolder, a different color). Faded by gradually reducing the exaggerated feature until the stimulus is in its natural form.
Extra-stimulus prompt: Adding an additional stimulus not part of the natural S^D (e.g., pointing to the correct card). Must be removed entirely during fading.
Within-stimulus prompts are generally preferred over extra-stimulus prompts because they direct attention to the relevant stimulus features, facilitating discrimination learning.
Prompt Fading Procedures
Most-to-Least (MTL) Prompting
Begin with the most intrusive prompt level that ensures a correct response, then systematically reduce prompt intrusiveness across trials or sessions.
Typical hierarchy: Full physical → partial physical → model → gestural → independent
Procedure:
- Begin at the full prompt level; reinforce prompted correct responses
- After the learner meets criterion at the current level (e.g., 3 consecutive correct), move to the next less intrusive level
- If the learner makes an error at a new level, return to the previous level briefly, then re-attempt fading
- Continue until the learner responds independently to the S^D alone
Advantages: Minimizes errors (errorless or near-errorless); reduces frustration; learner contacts reinforcement from the first trial.
Best for: New skill acquisition, learners with histories of failure or escape behavior, skills where errors are costly or dangerous (safety skills).
Least-to-Most (LTM) Prompting
Present the S^D and give the learner an opportunity to respond independently. If no correct response occurs within the response interval, provide the least intrusive prompt. If still no correct response, escalate to the next prompt level.
Typical hierarchy: Independent opportunity → gestural → model → partial physical → full physical
Procedure:
- Present the S^D and wait the predetermined response interval (e.g., 5 seconds)
- If correct, reinforce
- If incorrect or no response, provide the least intrusive prompt
- Wait again; if correct, reinforce (at potentially lower magnitude)
- If incorrect, escalate to the next prompt level
- Continue until a correct response is obtained
Advantages: Provides opportunity for independent responding on every trial; avoids unnecessary prompting; promotes learner initiative.
Best for: Learners who have some approximation of the target skill, maintenance targets, skills where independence and self-initiation are priorities.
Graduated Guidance
Continuous physical guidance that is adjusted moment-to-moment based on the learner's performance within a single trial. The clinician begins with full physical support and reduces pressure/contact as the learner begins to respond independently, then increases support if the learner falters.
Sometimes described as a "shadowing" technique — the clinician's hands hover near the learner's hands, ready to provide support only as needed.
Best for: Motor skills, self-care routines, skills taught through total task chaining. Particularly useful when the learner's performance varies within a single trial.
Constant Time Delay (CTD)
- Initial sessions (0-second delay): Present the S^D and simultaneously deliver the controlling prompt. The learner responds correctly with the prompt, contacts reinforcement.
- Delay sessions (X-second delay): Present the S^D and wait a fixed interval (typically 3–5 seconds) before delivering the prompt. If the learner responds correctly before the prompt (a "before" response), deliver higher-magnitude reinforcement. If the learner waits for the prompt and responds correctly (a "wait" response), deliver reinforcement at standard magnitude.
Advantages: Simple to implement; learner cannot make an error if they wait for the prompt; teaches the learner to either respond independently or wait for assistance.
Best for: Receptive and expressive identification, sight words, math facts, chained steps — any skill with a discrete correct response.
Progressive Time Delay (PTD)
Same as CTD but the delay interval increases incrementally across sessions: 0s → 1s → 2s → 3s → 4s → 5s.
Advantages: More gradual transfer of stimulus control; may reduce errors compared to CTD for some learners.
Considerations: More complex to implement; requires tracking the current delay interval.
Stimulus Fading
Gradually change the physical features of the S^D from an exaggerated/prompted form to the natural form. The prompt is embedded within the stimulus itself and faded by reducing the exaggeration.
Example: Teaching letter discrimination by initially presenting the target letter in red and larger, then gradually reducing the size and color difference until all letters are presented identically.
Advantages: Directs attention to relevant stimulus features; less susceptible to prompt dependency because the prompt is part of the stimulus.
Stimulus Shaping
Gradually transform the stimulus from a known/recognizable form to the target form. Unlike stimulus fading (which changes magnitude of a feature), stimulus shaping changes the topography of the stimulus.
Example: Superimposing a picture of a cat over the printed word "cat," then gradually fading the picture while the word remains, transferring control from the picture to the word.
Prompt Dependency
Prevention
- Always have a fading plan before introducing a prompt
- Use the least intrusive prompt that will be effective
- Fade prompts on a schedule — do not maintain the same prompt level indefinitely
- Collect first-trial data to detect whether the learner is responding to the S^D or waiting for the prompt
- Vary who delivers prompts; avoid associating one person with the prompt
Identification
- The learner consistently waits for a prompt before responding, even after extended training
- Independent responding does not increase across sessions
- The learner responds correctly only with specific therapists who prompt
- Latency to respond increases when prompts are withheld
Remediation
- Implement a structured time delay procedure (CTD or PTD)
- Differentially reinforce independent responses with higher-magnitude reinforcement
- Use transfer trials: prompt → distractor → re-present S^D without prompt
- Reduce the predictability of prompt delivery
- Consider switching prompt modality (e.g., from verbal to gestural)
- In severe cases, return to errorless learning with stimulus fading rather than response prompts
Selecting Prompt Type and Fading Method
Consider the following factors:
- Learner's existing repertoire: Can they imitate? Follow verbal directions? Attend to gestures?
- Target skill: Motor skill → physical prompts; vocal skill → echoic/verbal prompts; discrimination → stimulus prompts
- Error tolerance: Low tolerance (history of escape, aggression) → MTL or errorless; higher tolerance → LTM
- Setting: Naturalistic settings favor less intrusive prompts and time delay; structured settings can accommodate any procedure
- Data: If a procedure is not producing progress within a reasonable timeframe, change it
Key References
- Cooper, J. O., Heron, T. E., & Heward, W. L. (2020). Applied Behavior Analysis (3rd ed.). Pearson.
- 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. Foxx (Eds.), Making a Difference: Behavioral Intervention for Autism. PRO-ED.
- Touchette, P. E. (1971). Transfer of stimulus control: Measuring the moment of transfer. Journal of the Experimental Analysis of Behavior, 15(3), 347–354.
- Wolery, M., & Gast, D. L. (1984). Effective and efficient procedures for the transfer of stimulus control. Topics in Early Childhood Special Education, 4(3), 52–77.
- Halle, J. W., Marshall, A. M., & Spradlin, J. E. (1979). Time delay: A technique to increase language use and facilitate generalization in retarded children. Journal of Applied Behavior Analysis, 12(3), 431–439.