Understanding Prompt Hierarchy and Prompt Levels in ABA

Key Takeaways

A prompt hierarchy is an ordered sequence of prompts - from full physical (FP) to independent (I) - that guides systematic fading during ABA skill acquisition. Prompt hierarchies ensure consistent support levels across team members and provide objective data for fading decisions. The standard hierarchy includes seven levels from most to least intrusive: Full Physical (FP), Partial Physical (PP), Model (M), Verbal (V), Gestural (G), Independent (I), and No Response (NR). Whether you use least-to-most or most-to-least prompting depends on the learner and skill. A good ABA data collection app makes tracking prompt levels during active sessions practical rather than overwhelming.

Every skill your client learns requires a journey from dependence to independence. Some learners need hand-over-hand guidance to get started. Others just need a quick verbal reminder. The difference between good and excellent teaching often comes down to using the right level of support at the right time - and knowing when to pull back.

That’s where prompt hierarchies come in. A prompt hierarchy provides a systematic framework for delivering and fading instructional support. It answers two critical questions: what type of help should I give right now, and how do I know when to give less?

This guide covers the types of prompts in a standard hierarchy, when to use least-to-most versus most-to-least approaches, how to record prompt data effectively, and how to use that data to make smart fading decisions.

What Is a Prompt Hierarchy?

A prompt hierarchy is an ordered sequence of prompts ranging from most assistance to least assistance (or vice versa). It creates a continuum of support levels that you can systematically move through as a learner develops competence.

The concept comes from the principle of stimulus control transfer - gradually shifting the control of behavior from prompts (supplementary stimuli) to natural stimuli in the environment. As Cooper, Heron & Heward explain in Applied Behavior Analysis, the goal is always to fade prompts as quickly as possible while maintaining accurate responding.

A defined hierarchy ensures:

• Consistency across team members - Everyone provides similar support levels
• Objective measurement - You can track exactly how much help was needed
• Clear fading criteria - Data shows when to move to less intrusive prompts
• Prevention of prompt dependency - Systematic fading reduces over-reliance on prompts

Types of Prompts in ABA

The standard prompt hierarchy includes six to seven prompt levels, ordered from most to least intrusive.

Full Physical Prompt (FP)

The therapist physically guides the learner through the entire response, hand-over-hand.

Example: Placing your hands over the learner’s hands and moving them through the motions of writing their name.

Partial Physical Prompt (PP)

The therapist provides partial physical guidance - a light touch, gentle nudge, or guidance through part of the movement.

Example: Lightly tapping the learner’s elbow to prompt them to reach for an object.

Model Prompt (M)

The therapist demonstrates the correct response for the learner to imitate.

Example: Showing how to fold a towel before asking the learner to fold one.

Verbal Prompt (V)

The therapist provides spoken instructions, hints, or reminders about what to do.

Example: “Remember to look both ways before crossing” or “What do we do with our hands during circle time?”

Gestural Prompt (G)

The therapist uses non-verbal cues like pointing, nodding, or looking toward the correct response.

Example: Pointing toward the correct choice during a matching task.

Independent (I)

The learner completes the response without any prompt. This is the goal for all skill acquisition programs.

No Response (NR)

The learner does not attempt the skill within the allowed response interval. Important to track because it indicates the current prompt level isn’t sufficient or the learner isn’t motivated.

Least-to-Most vs. Most-to-Least Prompting

Two primary strategies determine how you move through the hierarchy.

Least-to-Most Prompting (LTM)

Begin with the least intrusive prompt and increase support only if the learner doesn’t respond correctly.

How it works:

1. Present the natural cue and wait
2. If no correct response, provide a gestural prompt
3. If still no correct response, provide a verbal prompt
4. Continue increasing until the learner responds correctly

Best for: Learners who already have some skills, when assessing current ability, and when independence is the primary goal.

Drawbacks: Can result in more errors during acquisition; may be frustrating for learners who need significant support.

Most-to-Least Prompting (MTL)

Begin with the most intrusive prompt necessary to ensure correct responding, then systematically fade to less intrusive prompts.

How it works:

1. Begin with full physical or model prompt (whatever ensures success)
2. After criterion is met (e.g., 3 correct trials), fade to the next level
3. Continue fading until the learner responds independently

Best for: New skills with no prior learning history, learners who become frustrated with errors, and safety-critical skills.

Drawbacks: Can create prompt dependency if fading is too slow; may not reveal what the learner can already do independently.

Choosing Your Strategy

Many practitioners use a hybrid approach - starting with most-to-least for initial acquisition, then switching to least-to-most probes to assess independent performance.

Other Prompting Strategies

Time Delay

Time delay introduces a waiting period between the natural cue and the prompt. As the learner demonstrates competence, the delay increases.

Constant time delay: Same delay duration throughout (e.g., always 4 seconds).

Progressive time delay: Delay increases over sessions (e.g., 0 seconds, then 2, then 4, then 6).

Graduated Guidance

A variation where the intensity of physical guidance adjusts moment-to-moment based on learner resistance or cooperation. You provide only as much support as needed at any instant.

Recording Prompt Data Effectively

Prompt data is only useful if collected consistently.

Define Your Hierarchy Clearly

Before teaching begins, document exactly what each prompt level looks like for this specific skill.

Weak definition: “V = Verbal prompt”

Strong definition: “V = Any spoken instruction that directly names the target action, such as ‘Touch the red one’ or ‘Put the spoon in the bowl.’ Does NOT include general encouragement like ‘Try again.’”

Use Consistent Codes

Standard prompt hierarchy codes make data collection faster:

Record in Context

Task analysis: Record the prompt level for each step. See our task analysis data collection guide for detailed strategies.

Discrete trial training: Record the prompt level for each trial, or track percentage of independent responses.

Natural environment teaching: Record the prompt needed during each teaching opportunity.

Using Prompt Data to Make Fading Decisions

Establish Fading Criteria Before Teaching

Define what level of success warrants fading to the next prompt level.

Example criteria:

• “Fade from FP to PP after 3 consecutive sessions with 80%+ correct”
• “Move to next prompt level when learner responds correctly on first attempt for 5 consecutive opportunities”

Clear criteria prevent subjective decisions. If the data meets criteria, you fade. If it doesn’t, you maintain current support.

Look for Stability, Not Just Success

A single good session doesn’t mean it’s time to fade. Look for:

• Consistency across opportunities - Similar performance throughout the session
• Stability across sessions - Performance maintained over multiple days
• Consistency across people - Similar responding with different therapists

Recognize Prompt Dependency

Prompt dependency occurs when a learner waits for prompts even when capable of independent response. Warning signs:

• Looking at the therapist before responding
• Correct responding with prompts but low correct responding on independence probes
• Increasing latency when prompts are delayed

If you suspect prompt dependency, use least-to-most probes, increase wait time before prompting, and reinforce independent attempts heavily.

Common Prompt Patterns and What They Mean

Steady Decrease in Prompt Levels

What it means: Teaching is working. The learner is acquiring the skill as expected.

Action: Continue current approach, fade according to criteria.

Stuck at One Prompt Level

What it means: Current teaching approach may not be effective for this learner or skill.

Action: Check if the prompt is being delivered consistently, consider whether the step is too large, evaluate motivation and reinforcement.

Variable Performance

What it means: The skill may not be truly acquired, or environmental variables are affecting performance.

Action: Look for patterns in setting events (time of day, preceding activities), increase practice at a more consistent prompt level, extend mastery criteria.

Regression After Fading

What it means: Fading was premature, or the skill wasn’t maintained adequately.

Action: Return to the last successful prompt level, increase reinforcement density, practice to more stringent criteria before fading again.

How TallyFlex Handles Prompt Levels

Traditional prompt data collection means managing paper data sheets with prompt codes, manually calculating percentages of independent responding, and hoping everyone on the team records the same way.

TallyFlex includes a complete prompt hierarchy system designed for clinical workflows. Built with input from practicing BCBAs, the system supports the data collection patterns real teams use.

Configurable Prompt Hierarchies

The default hierarchy includes seven prompt levels: I (Independent), G (Gestural), V (Verbal), M (Model), PP (Partial Physical), FP (Full Physical), and NR (No Response). Each level is color-coded for quick visual identification during data collection.

Customize to match your program: Add, remove, or reorder prompt levels to match your organization’s codes or a specific learner’s needs. The weight-based system tracks intrusiveness - Independent has weight 0, and higher weights indicate more intrusive prompts.

Task Analysis with Per-Step Prompt Tracking

When running task analysis sessions, TallyFlex requires prompt level recording for each step. As learners work through chained skills like hand washing or tooth brushing, therapists tap the appropriate prompt level badge after each step. The app auto-advances to the next step, keeping pace with the session.

What you see during a session:

• Color-coded prompt badges for each completed step
• Real-time count of independent versus prompted steps
• Visual progress through the task chain

Percent Correct with Optional Prompt Tracking

For discrete trial training (DTT), you can enable prompt level tracking to measure percentage of independent responses - not just overall accuracy. This optional toggle transforms standard percent correct tracking into a richer data source.

When enabled:

• One-tap recording for independent trials (the most common response during mastery)
• Quick picker modal for prompted trials
• Real-time stats showing independent count, prompted count, and % independent
• Undo support for quick corrections

Prompt Distribution Visualization

TallyFlex generates prompt distribution charts showing how prompting patterns change over time. Stacked area charts reveal trends across sessions, while summary views show the overall distribution of prompt levels for a program.

These visualizations help answer questions like: Is the learner becoming more independent? Which steps still require the most support? Is a specific prompt level being over-used?

Mastery Criteria Options

For task analysis programs, TallyFlex supports three mastery criterion types:

• Percent Independent - Mastery when independent responses reach a threshold (e.g., 80% of steps independent)
• All Independent - Mastery requires 100% independent responding
• Maximum Prompt Level - Mastery when no prompts exceed a specified level in the hierarchy

Data Export

All prompt data exports to CSV or Excel with full detail - prompt level per step, session summaries, and distribution data. Task analysis exports display steps as rows and sessions as columns, matching familiar paper data sheet layouts.

Explore our recording methods documentation to see how prompt data collection works in TallyFlex.

What’s Next?

• Learn about task analysis: Task Analysis Data Collection Guide covers how to use prompt hierarchies in chained skill teaching
• Choose recording methods: How to Choose the Right ABA Recording Method helps match methods to behaviors
• See TallyFlex in action: Spend your time on analysis, not data entry - configurable prompt hierarchies and mastery tracking that sync across your team’s devices

Good prompt data shows you exactly where a learner is on the path to independence. That clarity drives better fading decisions - and faster skill acquisition.