These answers draw in part from “Generalization & maintenance” (ABA Courses), and extend it with peer-reviewed research from our library of 27,900+ ABA research articles. Clinical framing, BACB ethics code references, and cross-links below are synthesized by Behaviorist Book Club.
View the original presentation →Generalization refers to the occurrence of a trained behavior in contexts that were not part of the original training — different settings, stimuli, people, or response variations. Maintenance refers to the persistence of a trained behavior over time after the original training contingencies have been removed or reduced. A skill can generalize without maintaining (occurs in new settings but fades over time) or maintain without generalizing (persists in the training context but does not transfer). Both must be explicitly programmed and measured for a skill acquisition program to produce lasting, functional outcomes.
Behavioral principles explain why generalization does not happen automatically. Stimulus control is established over the specific stimuli present during training — the instructor, the room, the materials, the time of day. New stimuli that differ from the training context do not automatically acquire control over the behavior. Without systematic training across varied exemplars, the behavior remains under narrow stimulus control. Stokes and Baer's foundational work established that generalization must be programmed, not just hoped for — and the decades of applied research since then have confirmed this principle consistently.
Multiple-exemplar training — teaching across varied instructors, settings, and materials from the beginning of a program — is among the most effective generalization strategies. Training in natural environments, programming for common stimuli between training and generalization settings, using indiscriminable contingencies (varied reinforcement schedules that reduce the salience of 'this is training' signals), and teaching self-management strategies to extend behavior to novel contexts are all evidence-supported tactics. The strategy selected should match the type of generalization needed and the characteristics of the skill being taught.
Generalization probes should be conducted infrequently enough not to function as training trials — typically without prompts, in an unprompted test condition, with novel materials or in a novel location. RBTs should understand that probes measure whether the skill has transferred, not whether the client can perform with support. Data from probes should be recorded and communicated to the supervising BCBA, who uses it to make programming decisions. RBTs who understand what probes are for will implement them more accurately and interpret results more meaningfully than those who see them as routine data collection.
Maintenance probes should be conducted at regular intervals following removal of a skill from active training. Common schedules include probes immediately after removal (1-2 weeks), at 1 month, 3 months, and 6 months. For safety-critical skills, more frequent maintenance probes are appropriate. If a maintenance probe reveals performance below criterion, the decision rule in the program should specify next steps — typically brief retraining with mastery verification before the next scheduled probe. Documenting maintenance data is a clinical obligation; skills with unknown maintenance status should not be counted as truly mastered.
Natural contingencies are the most powerful driver of long-term maintenance. When a behavior produces reinforcement in the natural environment — requesting produces the desired item, a greeting produces a smile and conversation, a safety skill produces adult praise — it maintains without ongoing artificial programming. BCBAs who design programs should ask, for every skill: will this behavior contact reinforcement naturally in the client's daily environment? If not, the program should explicitly plan for how natural contingencies will be established, or the skill may fade when direct training is removed.
Motivating operations (MOs) affect how strongly a behavior is evoked across contexts. A skill trained under a specific MO state may not generalize to contexts where that MO is absent or weaker. For example, requesting trained under high deprivation may not generalize to contexts where the individual is less motivated for the item. BCBAs should consider MO state during generalization probes — if probes consistently occur under low motivation, the absence of generalized behavior may reflect MO sensitivity rather than true generalization failure. Programming across varied MO states is a sophisticated but important generalization tactic for communication and other motivationally sensitive repertoires.
Common errors include: closing programs at clinic-level mastery without generalization probes; failing to specify generalization criteria in the program document; conducting generalization probes too frequently (making them function as additional training trials); forgetting to schedule maintenance probes; designing programs in a single setting with a single instructor without planned variation; and assuming caregiver implementation will handle generalization without explicit training. Many of these errors stem from treating generalization and maintenance as procedural extras rather than core program components from program inception.
Supervisors should explicitly teach the rationale for generalization procedures — not just what to do but why. When RBTs understand that varying materials and settings is not arbitrary but is the mechanism through which skills become functional, they implement variations with more precision and less resistance. Supervision meetings should review generalization probe data, not just acquisition data. BST can be used to train RBTs to conduct generalization probes correctly. Including generalization programming concepts in initial staff training curricula and RBT competency assessments embeds these practices as standard rather than advanced.
The first step is identifying the specific barrier to generalization: Is it a different instructor? A novel setting? Different materials? Absence of the training prompt? Once the discriminative stimuli controlling the behavior in the training context are identified, the program can be modified to bridge the gap. Common approaches include conducting training trials in the generalization setting using the same discriminative stimuli present there, introducing common stimuli from the generalization setting into training, adding varied instructors to training, and gradually fading the distinctive training stimuli while maintaining correct responding.
The ABA Clubhouse has 60+ on-demand CEUs including ethics, supervision, and clinical topics like this one. Plus a new live CEU every Wednesday.
Ready to go deeper? This course covers this topic with structured learning objectives and CEU credit.
Generalization & maintenance — ABA Courses · 1 BACB General CEUs · $0
Take This Course →We extended these answers with research from our library — dig into the peer-reviewed studies behind the topic, in plain-English summaries written for BCBAs.
256 research articles with practitioner takeaways
252 research articles with practitioner takeaways
195 research articles with practitioner takeaways
1 BACB General CEUs · $0 · ABA Courses
Research-backed educational guide with practice recommendations
Side-by-side comparison with clinical decision framework
You earn CEUs from a dozen different places. Upload any certificate — from here, your employer, conferences, wherever — and always know exactly where you stand. Learning, Ethics, Supervision, all handled.
No credit card required. Cancel anytime.
All behavior-analytic intervention is individualized. The information on this page is for educational purposes and does not constitute clinical advice. Treatment decisions should be informed by the best available published research, individualized assessment, and obtained with the informed consent of the client or their legal guardian. Behavior analysts are responsible for practicing within the boundaries of their competence and adhering to the BACB Ethics Code for Behavior Analysts.