shaping

name	shaping
description	Use when teaching behaviors that do not currently exist in the learner's repertoire and cannot be evoked through prompting or modeling, by reinforcing successive approximations toward a terminal behavior.

Shaping (Successive Approximation)

Shaping is the differential reinforcement of successive approximations toward a target (terminal) behavior. It is the fundamental procedure for establishing novel behaviors — behaviors that do not currently exist in the learner's repertoire in any form that can be prompted, modeled, or physically guided. Shaping is how entirely new response topographies, durations, intensities, and latencies come into being.

When Shaping Is Indicated

Use shaping when:

• The target behavior does not exist in the learner's current repertoire at all
• The behavior cannot be prompted or modeled (e.g., vocal productions the learner cannot imitate, emotional regulation states, fluency targets)
• The learner has a crude approximation that needs to be refined in topography, duration, frequency, intensity, or latency
• Prompting has failed to evoke the target response
• The target is a dimension of behavior (longer duration, softer voice, faster rate) rather than a new topography

Do not use shaping when the behavior can be efficiently taught through prompting, modeling, or chaining. Shaping is slower and less predictable than prompt-based instruction; use it only when other procedures are insufficient.

Core Mechanism

Shaping operates through differential reinforcement at each successive step:

• Reinforce responses that meet or exceed the current criterion
• Withhold reinforcement (extinction) for responses that fall below the current criterion
• When the current criterion is performed reliably, raise the criterion to the next approximation

This cycle of reinforcement → extinction → reinforcement at a higher standard drives the behavior toward the terminal target.

Steps for Implementing Shaping

1. Define the Terminal Behavior

Specify the target behavior precisely, including the relevant dimension (topography, frequency, duration, latency, magnitude). The terminal behavior is the end goal — what the behavior should look like when shaping is complete.

Example: "The learner will produce the word 'water' with intelligible /w/, /ɑ/, /t/, /ɚ/ phonemes within 3 seconds of the EO, audible at conversational volume."

2. Identify the Starting Behavior

Observe the learner's current repertoire and identify the closest existing approximation to the terminal behavior. This is the response you will reinforce first.

Example: The learner currently produces "wuh" when wanting water. This is the starting point.

3. Plan the Approximation Steps

Define intermediate criteria between the starting behavior and the terminal behavior. Steps should be:

• Small enough to prevent ratio strain (too large a jump causes the learner to stop responding because reinforcement is too infrequent)
• Large enough to produce observable progress within a reasonable number of sessions
• Sequentially ordered — each step is closer to the terminal behavior than the last

Example progression: "wuh" → "wah" → "wah-uh" → "wah-ter" → "water"

Not every step needs to be predetermined. Shaping is dynamic — the clinician adjusts steps based on the learner's behavior in real time. However, having a general plan prevents drifting or reinforcing tangential variations.

4. Differentially Reinforce Each Criterion

At each step:

• Reinforce every response that meets the current criterion (continuous reinforcement during acquisition)
• Do not reinforce responses that fall below the current criterion
• Use immediate reinforcement — delay reduces the precision of shaping
• Use potent reinforcers — shaping requires strong motivation to maintain responding through extinction phases

5. Raise the Criterion When Ready

Move to the next approximation step when the current step is being produced:

• At a stable rate
• On a high percentage of opportunities (typically 80%+ at the current criterion)
• Without excessive variability that would be extinguished at the next step

Raising criteria too quickly: The learner stops responding (ratio strain, extinction burst). Back up to the previous criterion, re-establish the behavior, then try a smaller step.

Raising criteria too slowly: The learner's behavior becomes fixed at the current level and resistant to change. Also risks reinforcing a "good enough" approximation that never reaches the terminal target.

6. Continue Until Terminal Behavior Is Achieved

Once the terminal behavior is occurring reliably, transition from continuous to intermittent reinforcement to maintain the behavior. Program for generalization across settings, people, and conditions.

Shaping Dimensions

Shaping can target any measurable dimension of behavior:

Topography: The physical form of the response (e.g., shaping speech sounds, shaping handwriting form, shaping a dance move).

Frequency/Rate: How often the behavior occurs per unit of time (e.g., shaping increased rate of social initiations from 1 to 5 per recess period).

Duration: How long the behavior persists (e.g., shaping eye contact from 1 second to 5 seconds; shaping on-task behavior from 2 minutes to 20 minutes).

Latency: The time between the S^D and the response onset (e.g., shaping compliance latency from 30 seconds to 5 seconds).

Magnitude/Intensity: The force, volume, or amplitude of the response (e.g., shaping voice volume from whisper to conversational level; shaping grip strength).

Shaping vs. Fading

These two procedures are frequently confused:

• Shaping changes the response — the learner's behavior changes across successive steps
• Fading changes the stimulus — the prompt or antecedent stimulus changes while the response stays the same

They operate on different components of the three-term contingency and solve different problems. They can be used together (e.g., shaping vocal approximations while fading echoic prompts).

Risks and Pitfalls

Accidental Shaping of Undesired Behavior

Shaping works on any behavior that is differentially reinforced — including problem behavior. A child who whines and sometimes gets what they want may have their whining shaped into screaming, then tantruming, through inadvertent differential reinforcement of increasingly intense forms.

Prevention: Be aware of what you are reinforcing. Extinction bursts produce response variability — if you reinforce a more intense variant during an extinction burst, you have shaped the behavior in the wrong direction.

Ratio Strain

If the criterion jumps too far, the learner enters extinction for too long, and responding decreases or stops entirely. The clinician must recognize ratio strain quickly and back up to a previously reinforced criterion.

Reinforcing Variability vs. Reinforcing Approximation

During shaping, the learner's behavior naturally varies. Not all variations are closer to the target. The clinician must selectively reinforce only those variations that move toward the terminal behavior, not simply any change from the previous form.

Getting Stuck at an Intermediate Step

If a particular approximation is heavily reinforced, it may become resistant to change. Strategies:

• Thin the reinforcement schedule at the current step before raising the criterion
• Use a DRO (brief pause) between current-step responses to disrupt the pattern
• Introduce a slight environmental change to evoke behavioral variability

Clinical Examples

Shaping Vocal Responses

A nonvocal learner who produces "mmm" sounds. Terminal behavior: saying "more" to request.

• Step 1: Reinforce any vocalization → Step 2: Reinforce only "mmm" or "muh" → Step 3: Reinforce "moh" → Step 4: Reinforce "mohr" → Step 5: Reinforce "more"

Shaping Duration of Task Engagement

A learner who currently attends to a work task for 30 seconds before leaving the table. Terminal behavior: 10 minutes of independent work.

• Reinforce at 30s → 1 min → 2 min → 3 min → 5 min → 7 min → 10 min
• Use a timer; deliver reinforcement at the end of each interval

Shaping Social Proximity

A learner who avoids peers and stays more than 10 feet away. Terminal behavior: playing within 3 feet of peers.

• Reinforce being in the same room → within 8 feet → within 6 feet → within 4 feet → within 3 feet
• Pair proximity with preferred activities at each step

Shaping Handwriting Legibility

A learner who produces illegible letter forms. Terminal behavior: letters formed within guidelines, recognizable.

• Reinforce any mark within the line boundary → reinforce approximations of the letter shape → reinforce correctly oriented letters → reinforce letters within size guidelines

Combining Shaping with Other Procedures

• Shaping + prompting: Use prompts to evoke approximations that are then shaped further (e.g., physical prompt to start a vocal approximation, then shape the vocalization)
• Shaping + chaining: Shape individual steps of a chain before linking them
• Shaping + differential reinforcement: DRA is the mechanism within shaping — reinforce the target approximation, extinguish lesser forms
• Shaping + visual supports: Visual models of the terminal behavior can guide the learner's approximations

Key References

• Cooper, J. O., Heron, T. E., & Heward, W. L. (2020). Applied Behavior Analysis (3rd ed.). Pearson.
• Skinner, B. F. (1953). Science and Human Behavior. Macmillan.
• Galbicka, G. (1994). Shaping in the 21st century: Moving percentile schedules into applied settings. Journal of Applied Behavior Analysis, 27(4), 739–760.
• Ricciardi, J. N., Luiselli, J. K., & Camare, M. (2006). Shaping approach responses as intervention for specific phobia in a child with autism. Journal of Applied Behavior Analysis, 39(4), 445–448.
• Hagopian, L. P., & Thompson, R. H. (1999). Reinforcement of compliance with respiratory treatment in a child with cystic fibrosis. Journal of Applied Behavior Analysis, 32(2), 233–236.