An ABA prompt fading plan worksheet should show the natural cue, the initial response, the extra prompt, its timing, and the evidence for a plan change. A note such as “faded to gestural” leaves most of that record missing. It also leaves unclear whether ordinary communication or access supports were mistaken for prompts.

The working tables below turn one approved, case-specific prompt-fading plan into a record that another qualified reviewer can reconstruct. They do not rank most-to-least, least-to-most, time delay, stimulus fading, graduated guidance, or any other arrangement. Research comparisons have produced different results across learners and tasks, which is a reason to define and evaluate the selected procedure rather than treat a familiar hierarchy as a default.

Clinicians & ABA Professionals / Assessment and Treatment Planning.

Start with the transfer question

Write the question in plain language: Under which natural cue should this response occur, and what evidence would show that the extra prompt is no longer controlling it? That wording keeps prompt removal tied to a meaningful stimulus-control transfer question. It also prevents a smaller code number from masquerading as clinical progress.

Use a fresh copy for each target, setting, cue arrangement, or material change. A response that occurs with one person, one device layout, or one prompt history is not automatically established elsewhere.

Plan identityComplete before useClient-selected purpose and meaningful outcomeTarget response, including accepted formsNatural cue that should eventually occasion the responseOpportunity start and response-window endOrdinary supports that remain availablePrompt type or combination under reviewEvidence that the proposed prompt can occasion the responseCurrent plan version, author, qualified approver and effective dateSettings, materials, people and exclusionsConsent, assent or dissent process and stop conditions

Accepted response forms should include applicable speech, gesture, sign, AAC, writing, selection, movement, or other accessible forms. The ASHA AAC Practice Portal says people who use AAC should always have access to their communication tools or devices. Device access is therefore not a prompt to strip away for the sake of an “independent” score.

Specify exactly what will change

A fade map names what changes and what stays constant. That matters because “fading” can refer to a change in a prompt, a discriminative stimulus, a response requirement, or reinforcement. Kincaid's gradual-change taxonomy separates those functions; a reviewer should do the same in the plan.

StepNatural cue and materialsExtra promptDelay or prompt dimensionReinforcement arrangementAdvance evidence to reviewHold, return or stop evidenceBaseline or probeNone, unless safety/access plan says otherwiseCurrent approved stepProposed next stepPrior step available for return

Do not build the table by assuming that verbal, gestural, model, or physical prompts form equal rungs. Prompt topographies differ, and a reduction in one dimension may not be comparable with a change in another. The BACB Sixth Edition BCBA Test Content Outline distinguishes selecting prompts from designing prompt-fading procedures, but it is examination content rather than a client protocol.

One ledger for the initial response and prompt history

The event row should preserve what happened before any extra prompt. Otherwise, a response corrected after prompting may overwrite the evidence needed to judge natural-cue responding.

Date/timePlan versionNatural cue delivered as written?Initial response before promptInitial latencyPlanned prompt/delayDelivered prompt/delayResponse after promptConsequence deliveredAccess, assent/dissent or context noteValid for this review?Yes / No / UnclearIndependent correct / incorrect / no response / unavailableCorrect / incorrect / no response / not applicableYes / No, reason

“Unavailable,” “not observed,” and “invalid” need their own states. If the cue arrived late, the prompt preceded the cue, the device was not available, or the response window was interrupted, the row may reveal an implementation problem without telling us how the learner would have responded under the plan.

Three measures, three distinct questions

The fictional example schedules 20 opportunities. Two are invalid because cue and prompt timing do not match the current version, leaving 18 valid opportunities. Eight valid rows contain an independently correct initial response, seven become correct after the planned prompt, and three remain incorrect or have no response after the defined sequence.

The completed review below summarizes those event rows without replacing them. “Prompted correct” means the initial response was not independently correct and the response became correct only after the planned prompt. Each row belongs to the same plan version and target.

Review categoryScheduled opportunitiesValid opportunitiesIndependently correctPrompted correctIncorrect or no response after sequenceNatural cue and timing delivered as written1818873Cue or prompt timing did not match the version20Not applicableNot applicableNot applicableReconciled total2018873

  • Valid coverage is 18 / 20 × 100 = 90.0%.
  • Initial independence is 8 / 18 × 100 = 44.4%.
  • Final correct responding is (8 + 7) / 18 × 100 = 83.3%.

Those values answer different questions. Coverage describes usable observation opportunities. Initial independence describes responding before the extra prompt. Final correctness includes prompted success. Adding or averaging the percentages would erase their denominators. None proves that natural stimulus control transferred, that the prompt caused the response, or that the plan should advance.

Use the ABA prompt fading plan worksheet to retain those three views side by side, then return to the event rows whenever a reviewer needs to understand an invalid opportunity, a latency, or a prompted response.

Comparative studies reinforce that caution. One comparison of most-to-least and least-to-most prompting used specified one-step directions and a small autistic sample. Another prompting-strategy evaluation organized learner, task, and procedure variables for decision-making. Neither source supplies a universal hierarchy for this form.

Plan natural-cue probes explicitly

A planned probe can ask whether the target response occurs under the intended cue without the extra prompt. It should not silently appear because an implementer forgot a prompt, and it should not require removing ordinary access.

Probe dateProbe purposeNatural cue and contextSupports retainedResponse and latencySafety or dissent eventInterpretation allowedNext reviewerDescriptive only / other stated limit

Record whether probe materials, people, setting, motivation, and opportunity definition resemble the teaching context. A correct probe is one observation, not proof of generalization or maintenance. A failed probe does not by itself prove prompt dependence. Studies of positional prompts and transfer procedures across autistic learners illustrate why the intended controlling stimulus and the tested conditions must be explicit.

Predetermine who may schedule a probe, which extra prompt will be withheld, what ordinary supports stay in place, and what happens after an error or no response. That prevents a missed prompt from being relabeled as a probe after the fact and keeps safety and access decisions outside the data-cleaning process.

Check implementation, access, and unwanted effects

Before changing a fade step, review whether the step was actually implemented. A plan can look ineffective when the natural cue, wait interval, prompt, consequence, or materials drifted. It can also look successful while burdens, withdrawal, or inaccessible response demands go unrecorded.

Review domainEvidence inspectedWhat appears consistentWhat is missing or mismatchedAction owner/dateCue and response-window integrityPrompt topography, timing and dosageConsequence and reinforcement deliveryAAC and other ordinary accessAssent, dissent, withdrawal and stop responsesErrors, latency, avoidance, distress or other unwanted effectsImplementer, setting and material variation

The progressive prompt-delay comparison used defined instructional arrangements with four children; its timing and results belong to that study. They are evidence that timing can be measured, not a delay prescription. Likewise, a study that compared prompting tactics for intraverbals does not generalize its steps to every skill or communicator.

Exceptions and qualified decisions

DateTrigger or exceptionEvidence availableOptions consideredClient/caregiver inputDecision: continue, hold, revise, return, stop or referQualified owner and next review

An exception log is not a list of failures. It is where a missing device, health change, setting disruption, unexpected prompt response, withdrawal, or safety event remains visible. The decision owner should say which plan version applies next and why. Old rows retain their original version label.

The live BACB Ethics Codes hub and Ethics Code for Behavior Analysts inform competence, dignity, client involvement, records, data accuracy, risk, and evaluation boundaries for covered certificants. They do not determine a case outcome here. The BACB test-content hub confirms which outline is current, while the CASP ABA Practice Guidelines access page identifies Version 3.0 and its licensing terms; licensed guideline detail is not reproduced.

Use this worksheet beside, not instead of, the current assessment and treatment plan, consent and assent process, safety plan, authorization, applicable law, payer requirements, and interdisciplinary recommendations. It is a drafting aid for transparent review, not an automated fade engine.

Related resources

Sources