Shaping is one of the core teaching procedures in applied behavior analysis (ABA), used to build entirely new behaviors by reinforcing small steps toward a target skill. Rather than waiting for a person to perform a complete, finished behavior and then rewarding it, shaping reinforces successive approximations, gradually raising the bar as the learner gets closer to the goal. The technique is used across a surprisingly wide range of settings, from teaching a child with autism to speak in full sentences to helping stroke survivors regain use of a paralyzed arm.
How Shaping Works in Practice
The logic behind shaping is straightforward: you start where the learner already is. If a child cannot yet say the word “water” but can produce the sound “wuh,” you reinforce “wuh.” Once “wuh” is coming reliably, you hold off on reinforcement until the child produces something closer to the target, like “wah-ter.” Each time the learner masters one approximation, the criterion shifts forward. The process continues, step by step, until the full target behavior is in place.
This makes shaping particularly useful for behaviors a person has never performed before. You cannot reinforce something that does not yet exist in a learner’s repertoire. Shaping gets around this by treating the path to the goal as a series of achievable waypoints. A therapist working with a nonverbal child is not expecting speech on day one. They might start by reinforcing any vocalization, then vowel sounds, then consonant-vowel combinations, and so on.
The key decisions in any shaping program are how big each step should be and when to move from one step to the next. Steps that are too large risk frustrating the learner, because they cannot bridge the gap between what they can do now and what is being asked. Steps that are too small can waste time or lead to the learner getting “stuck” at an intermediate level because reinforcement is too easy to earn. Finding the right pace is one of the skills that separates a well-run shaping program from a poorly run one.
How Shaping Differs from Fading and Chaining
People new to ABA sometimes confuse shaping with other gradual-change procedures like fading and chaining. A useful framework categorizes these by what is being gradually changed: shaping adjusts the response requirement (what the learner has to do), fading adjusts the stimuli or prompts (the cues the learner receives), and chaining links individual behaviors together into a sequence.1PubMed Central. Gradual Change Procedures in Behavior Analysis
Think of it this way: if you are teaching a child to tie shoes, you might use chaining to connect the individual steps (cross the laces, pull one under, make a loop, etc.) into a smooth sequence. If you are using physical hand-over-hand guidance and gradually reducing how much you help, that is fading the prompt. But if you are reinforcing the child for pulling the laces tighter and tighter until they produce a knot that actually holds, you are shaping the quality of a single step. In practice, these procedures often work together in the same program.
One study illustrated this overlap by using a stimulus fading program alongside differential reinforcement to expand the communication skills of two individuals with autism. Problem behavior dropped substantially once an alternative communication response was in place, and those gains held as the behavioral requirements were gradually increased during the fading program.2PubMed Central. Stimulus Fading and Response Elaboration in Differential Reinforcement for Alternative Behavior The takeaway for practitioners is that shaping rarely operates in isolation. It is usually one layer of a broader teaching strategy.
Building Communication Skills
Shaping has a long track record in communication training, particularly for children with autism and other developmental disabilities. One of the most researched applications is functional communication training (FCT), where a person is taught to use words, signs, or devices to request what they need instead of engaging in challenging behavior. A landmark study taught students assistance-seeking and attention-getting phrases to replace behaviors like aggression and self-injury. Not only did challenging behavior drop substantially, but the improvements transferred across new tasks, settings, and teachers and held for 18 to 24 months.3PubMed Central. Functional communication training to reduce challenging behavior: maintenance and application in new settings
A common challenge with FCT is that the initial communication response taught to the learner is often very simple, such as handing over a card or pressing a single button. That simplicity is intentional at the start because the goal is quick acquisition. But real-world communication demands more complexity. Shaping plays a critical role in building up from that simple starting point. Research has demonstrated that a structured shaping process, using a changing-criterion design, can develop complex functional communication responses from simple ones without causing a resurgence of the problem behavior the training was designed to replace.4PubMed. Shaping complex functional communication responses That last point matters. A poorly managed transition from simple to complex communication can backfire, with the learner reverting to problem behavior when the new, harder communication response feels out of reach.
Shaping also applies to the physical mechanics of speech. A study adapted vocal imitation training with prompting, chaining, and shaping to teach two children with autism and speech sound disorders to articulate words containing difficult consonant blends. Both participants improved their articulation of trained words, and correct pronunciation generalized to untrained words as well.5PubMed Central. Behavioral Interventions to Treat Speech Sound Disorders in Children With Autism The generalization finding is encouraging because it suggests that shaping does not just produce rote responses but can help a learner develop flexible skills they apply in new contexts.
Encouraging Variability with Lag Schedules
One of the subtler challenges with shaping is that reinforcement can inadvertently make behavior too rigid. If a child gets reinforced every time they say “cookie” in exactly the same way, they may never experiment with other words or sentence structures. Lag schedules address this. A lag schedule requires the learner to produce a response that differs from their most recent one (or from their last several responses) before reinforcement is delivered. The idea is to systematically reward novelty.
Research on children with autism found that a Lag 1 schedule (where the learner just had to differ from the immediately preceding response) increased the cumulative number of novel phonemes produced, boosted the percentage of trials with variability, and raised the number of different phonemes per session for two out of three participants.6PubMed Central. Using a lag reinforcement schedule to increase phonemic variability in children with autism spectrum disorders In a separate study focusing on conversational responses rather than individual sounds, lag schedules were used to strengthen the range of answers children with autism could give in back-and-forth exchanges. Some participants produced novel responses throughout the experiment, suggesting the approach can genuinely expand a learner’s repertoire rather than just shuffling among a fixed set of answers.7PubMed. Using lag schedules to strengthen the intraverbal repertoires of children with autism
Lag schedules pair well with shaping because shaping already involves pushing toward new behavior. A therapist might use shaping to bring a response closer to a target form while simultaneously using a lag contingency to keep the learner exploring rather than getting stuck in a rut. The combination can be especially useful for learners who have a tendency toward repetitive, stereotyped responses.
Common Pitfalls and Inadvertent Conditioning
Shaping is conceptually simple, but the timing and sequencing involved make it easy to accidentally teach the wrong thing. One well-documented risk involves prompting during functional communication training. If a therapist prompts a communication response immediately after problem behavior occurs, the prompting itself can inadvertently strengthen the problem behavior. The sequence the learner experiences is: I hit, and then someone helps me ask for what I want, and then I get it. From the learner’s perspective, hitting is the first step in getting reinforced.8PubMed. Understanding the effects of prompting immediately after problem behavior occurs during functional communication training This kind of accidental response chaining is one of the more common mistakes practitioners make, and it highlights why the timing of reinforcement and prompts during shaping is so important.
Another pitfall is the extinction burst, the temporary spike in behavior that occurs when reinforcement for an old response is withdrawn. When you raise the criterion during shaping (say, from accepting “wuh” to requiring “wah-ter”), you are placing the old response on extinction. The learner may respond with more intense or varied behavior. A recent study explored this phenomenon in detail and found that while prior exposure to extinction can make bursts more predictable, the bursts did not generalize to entirely new response forms.9PubMed Central. Shaping the extinction burst: Increasing its probability and preventing its emergence across topographies That is useful information for practitioners: if the learner has experienced criterion changes before, the brief uptick in old behavior is less likely to spread into something unexpected. Planning for extinction bursts and riding them out is part of competent shaping.
A third common error is moving the criterion too fast. When a therapist raises the bar before the learner has mastered the current step, the learner may stop responding altogether. Recovering from this usually means backing up to an easier step and rebuilding momentum. The costs are not just lost sessions; repeated failure can make the learner reluctant to engage with the task at all. Experienced practitioners tend to err on the side of spending one or two extra sessions at a given step rather than jumping ahead prematurely.
Measuring Whether a Shaping Program Is Working
Because shaping requires ongoing judgment calls about when to change the criterion, tracking data carefully is essential. But measuring shaping well is harder than it sounds. Standard ABA data collection often focuses on correct versus incorrect responses, which works fine for discrete-trial teaching but misses the nuance of shaping. A response that is “wrong” by the final-target standard may still represent real progress if it is closer to the goal than what the learner was doing yesterday.
Procedural fidelity, meaning whether the therapist is actually running the shaping program the way it was designed, is at least as important as measuring the learner’s progress. A guide for behavior analysts outlines a six-step process for creating fidelity measurement systems: breaking the treatment procedure into measurable components, assigning specific measures to each, planning the observation, collecting data, analyzing results, and taking corrective action when fidelity slips.10PubMed Central. A Practitioner’s Guide to Measuring Procedural Fidelity This matters for shaping specifically because the procedure depends so heavily on the therapist’s moment-to-moment decisions. Two therapists running the “same” shaping program can produce very different outcomes if one is reinforcing too loosely and the other too strictly.
Research on procedural integrity supports the intuitive conclusion that regular fidelity checks improve accuracy. When staff receive more frequent checks on how they are implementing behavior protocols, their implementation becomes more accurate.11DigitalCommons@CSP. Procedural Integrity: A Study of Implementing Integrity Utilizing Fidelity Checks and Performance Feedback For shaping programs in particular, this might mean a supervisor periodically reviewing session recordings to confirm that the therapist is shifting criteria at the right times and reinforcing the right approximations.
Applications Beyond Language and Behavior
Shaping is often associated with teaching children with autism, but its reach extends well beyond that population. One striking application is in physical rehabilitation. Research with stroke survivors found that a condition called learned nonuse, where a patient stops using an affected limb because early failed attempts go unreinforced, could be addressed through shaping principles. By restricting movement of the unaffected arm and then systematically shaping use of the affected one through verbal feedback and reinforcement, chronic stroke patients showed large improvements in both motor ability and functional independence.12PubMed Central. An operant approach to rehabilitation medicine: overcoming learned nonuse by shaping
This work laid the foundation for what became constraint-induced movement therapy (CI therapy), which is now one of the most evidence-based approaches in stroke rehabilitation. CI therapy’s protocol consists primarily of standard behavior-analytic methods, and it produces measurable changes in brain structure that correlate with the therapeutic gains.13PubMed Central. The behavior-analytic origins of constraint-induced movement therapy: an example of behavioral neurorehabilitation The fact that a procedure rooted in pigeon-lab operant conditioning ended up reshaping the brains of stroke patients is one of the more compelling cross-disciplinary stories in behavioral science.
Self-care skills represent another area where shaping shows up regularly. Teaching dressing, grooming, feeding, and toileting to children with developmental delays typically involves a combination of gradual-change procedures. A scoping review of interventions for self-care skills found that sessions as short as 30 minutes, delivered once or twice per week over 10 to 23 sessions, could produce measurable gains.14PubMed Central. A Scoping Review of Intervention Implementation for Self-Care Skills among Children with Developmental Delay Shaping is especially useful for skills like feeding, where a child might initially accept only a few foods and the goal is to gradually expand their tolerance.
Assent-Based Practice and Ethical Considerations
ABA has faced increasing scrutiny about whether its procedures respect the autonomy of the people they are applied to, particularly children and individuals who cannot easily advocate for themselves. Shaping intersects with this conversation directly, because the therapist, not the learner, decides the target behavior and controls when reinforcement happens. Modern practice has moved toward assent-based models, where the learner’s willingness to participate is treated as an ongoing requirement rather than a one-time consent event.
In an assent-based approach, the learner’s withdrawal from participation is honored, whether they say “no” verbally, turn away, cry, or express refusal through any other behavior.15PubMed Central. Assent in applied behaviour analysis and positive behaviour support: ethical considerations and practical recommendations This is a meaningful shift from older models where a child’s resistance might be treated as a behavior to work through rather than a signal to pause.
A concrete example of how this plays out with shaping comes from food selectivity treatment. Researchers designed a treatment package for young children with developmental disabilities and extreme food selectivity that incorporated differential reinforcement of successive approximations to eating nonpreferred foods while giving children the ability to opt in and out of sessions. The treatment was effective in increasing consumption, and it was extended successfully to caregivers.16PubMed Central. Prioritizing choice and assent in the assessment and treatment of food selectivity The study is a good illustration that shaping can work without forcing participation: the successive-approximation structure actually lends itself well to assent-based practice, because each step is small enough that the learner is unlikely to be overwhelmed.
Training practitioners to recognize and respond to assent withdrawal is its own area of development. A study evaluating assent withdrawal training for Board Certified Behavior Analysts found that all participants who received structured training acquired and maintained competencies in identifying when a client was withdrawing assent, modifying interventions accordingly, and embedding assent checks throughout sessions.17Behavioral Interventions. Evaluation of Assent Withdrawal Training for Board Certified Behavior Analysts: Impact on Competency and Clinical Practices
Shaping Across Species
Shaping did not originate in clinical work with humans. It comes from the animal operant conditioning laboratory, and it remains one of the primary training methods used with animals today. A scoping review of husbandry training in zoos found that shaping, often combined with clicker training, was the most frequently used technique. Over half of the behaviors trained were intended to make animal handling easier, while others helped with sample collection and medical treatment.18Applied Animal Behaviour Science. An examination of husbandry training in zoos: A scoping review The review noted a strong bias toward training mammals but pointed out that the few studies on non-mammal species (birds, reptiles, fish) demonstrated that reward-based training methods work broadly across animal groups.
The principles are identical whether you are working with a child or a parrot. Reinforce what is closest to the target, raise the criterion once the current approximation is reliable, and keep step sizes manageable. The main difference in animal training is the communication channel: since you cannot explain the goal to an animal, the precision of your timing and the clarity of your reinforcement signal (often a clicker or whistle that marks the exact moment of the correct behavior) carry even more weight.
Early automated approaches to shaping explored whether a computer could replicate the role of a skilled trainer. A study using a computer-controlled system tracked a pigeon’s head position continuously and reinforced successively closer approximations to a target location in three-dimensional space. The procedure worked: all three birds in the study acquired the target response rapidly.19PubMed Central. Shaping by automated tracking of an arbitrary operant response This line of research has implications beyond animal labs. As ABA incorporates more technology, including apps and devices for learners with disabilities, the idea of algorithmically managing the criterion-shift decisions in a shaping program continues to attract interest. Whether software can match the judgment of a skilled human therapist remains an open question, but the foundational proof of concept has been around for decades.