Generalization and maintenance are tasks C.8 and C.9 of the Registered Behavior Technician (RBT) Test Content Outline: implementing generalization procedures and distinguishing maintenance from acquisition. These 10 questions cover stimulus generalization and overgeneralization, rotating technicians, faulty stimulus control, fading an RBT's presence, thinning reinforcement, and running maintenance probes.
Question 1
Marisol was taught to greet her parents with a hug at home. During a generalization probe at the grocery store, she hugs the cashier and several shoppers she has never met.
Which term BEST describes Marisol's hugging in this situation?
- A) Response maintenance
- B) Response generalization
- C) Overgeneralization✓
- D) Stimulus discrimination
💡 Key Takeaway
A learned behavior showing up where it does not belong is overgeneralization.
Show rationale
Overgeneralization is correct because Marisol is emitting a trained behavior, hugging, under stimulus conditions where it is inappropriate and was never intended, with strangers in a store. Generalization is usually a goal, but when a response spreads to people or places where it does not fit, the RBT records what happened objectively and tells the supervisor. Response generalization is incorrect because it describes a new, untrained response emerging, while Marisol is performing the same trained hug. Stimulus discrimination is incorrect because it means responding differently to different stimuli, which is the opposite of hugging everyone. Response maintenance is incorrect because it refers to a behavior continuing over time after teaching ends, not spreading to unintended people.
Question 2
Kiara's program lists three new acquisition targets to be introduced one at a time, alongside her maintenance targets. She is responding quickly on the first new target, and the RBT wants to keep her momentum going.
How should the RBT proceed?
- A) Introduce all three new targets today since she is doing well on the first one.
- B) Pause her maintenance targets so all session time goes to the three new targets.
- C) Teach all three new targets in one mixed array so she cannot memorize the order.
- D) Keep teaching the first target to mastery, then add the next one as written.✓
💡 Key Takeaway
Introduce unlearned targets one at a time, as the program specifies, so the learner is not overwhelmed.
Show rationale
Following the written sequence is correct. Acquisition targets are unlearned skills, so introducing them one at a time keeps the learner from facing several unfamiliar tasks at once, which can increase errors, prompt dependence, and frustration. Introducing all three targets today because the first is going well abandons the sequence and loads Kiara with multiple unlearned tasks at the same time. Teaching all three in one mixed array still presents three unlearned targets together and departs from the program as written. Pausing her maintenance targets is a program change the RBT cannot make alone and risks losing mastered skills. If Kiara seems ready to move faster, the RBT shares the data with the supervisor.
Question 3
To promote generalization, a supervisor schedules three different RBTs to run Aiden's requesting program across the week while the skill is still being acquired. Aiden's primary RBT asks why the rotation starts now.
Which explanation is BEST?
- A) Rotating RBTs keeps his requesting from depending on one technician.✓
- B) Rotating RBTs lets the supervisor compare which technician teaches fastest.
- C) Rotating RBTs gives the team a way to check interobserver agreement daily.
- D) Rotating RBTs mainly helps the clinic cover schedules and prevent burnout.
💡 Key Takeaway
Teach with several people early so the skill does not depend on one technician.
Show rationale
Rotating technicians during acquisition is correct because a skill taught by only one person can come under that person's stimulus control, so the learner may request only when that technician is present. Practicing with several RBTs from the start builds generalization across people into teaching. Comparing which technician teaches fastest is not the clinical purpose, and progress differences would not show whether Aiden's skill generalizes. Checking interobserver agreement requires two observers recording the same session at the same time, not different technicians on different days. Covering schedules and preventing burnout may be helpful side effects, but they are staffing reasons rather than the reason the supervisor programmed the rotation for Aiden's skill.
Question 4
Before Nora's new self-care program begins, an RBT probes several skills and records her performance so the supervisor can sort them into acquisition and maintenance targets.
Which probe result BEST indicates an acquisition target?
- A) She put on her socks independently on every probe this past month.
- B) She tied her shoes correctly only when given full physical help.✓
- C) She washed her hands independently at home and at school each time.
- D) She zipped her coat independently on 9 of 10 probes this week.
💡 Key Takeaway
Acquisition targets are skills the learner cannot yet do independently; prompted success does not count.
Show rationale
Shoe tying is the acquisition target because acquisition targets have a low probability of independent correct responding before systematic teaching, and Nora succeeded only with full physical help, meaning her independent responding was zero. Correct responses that depend on prompts do not show the skill is learned. Zipping her coat independently on 9 of 10 probes shows a high probability of correct responding, which fits maintenance rather than new teaching. Putting on socks independently on every probe for a month is already a mastered skill suited to maintenance. Washing hands independently at home and at school shows the skill is mastered and already generalized across settings, so it does not need acquisition teaching either.
Question 5
An RBT taught Priya to touch the apple picture using one worn flashcard with a bent corner. When the RBT tests generalization with a new apple picture, Priya touches whichever card has a bent corner, even a picture of a shoe.
What is the MOST likely cause of Priya's errors?
- A) Poor response maintenance
- B) Prompt dependency
- C) Faulty stimulus control✓
- D) Overgeneralization
💡 Key Takeaway
When an irrelevant feature controls responding, the skill will not generalize to new examples.
Show rationale
Faulty stimulus control is correct because Priya's touching came under the control of an irrelevant feature, the bent corner, rather than the apple itself, so she fails to generalize to a new apple picture and selects unrelated cards that share the bend. Varying materials as the plan directs helps prevent this. Prompt dependency is incorrect because no prompts are involved and Priya responds without waiting for help. Overgeneralization is incorrect because it means responding to similar but inappropriate stimuli, such as touching a red ball when asked for the apple, whereas a shoe shares nothing relevant with an apple. Poor response maintenance is incorrect because the skill is not fading over time; it was never controlled by the right stimulus.
Question 6
Mateo just met mastery on a coin-counting target, and his supervisor has moved it from acquisition into maintenance. The RBT will keep practicing it with him periodically.
What should practice on this target now emphasize?
- A) Accurate first responses with prompts as needed
- B) Responding only with the RBT who first taught it
- C) Continuous reinforcement after every correct trial
- D) Quick, fluent responding that holds up over time✓
💡 Key Takeaway
Acquisition builds accuracy; maintenance builds fluency and durability.
Show rationale
Quick, fluent responding that holds up over time is correct because acquisition focuses on establishing accurate responding, while maintenance focuses on speed, fluency, and durability so the skill stays functional in daily life. Now that Mateo counts coins accurately, practice shifts toward smooth, efficient performance that persists. Accurate first responses with prompts as needed describe the acquisition phase, when the goal is getting the response correct at all. Continuous reinforcement after every correct trial is typical of early acquisition; maintenance relies on thinner schedules and natural reinforcement. Responding only with the RBT who first taught the skill works against durability and generalization, because a maintained skill should occur with many people, not just one technician.
Question 7
Grace now greets classmates on the playground, but only when her RBT stands right beside her. Her supervisor's plan targets generalization of these greetings to peers throughout recess.
What should the RBT do next?
- A) Gradually increase her distance from Grace across sessions.✓
- B) Stand behind Grace and whisper each greeting for her to repeat.
- C) Stay beside Grace and prompt each greeting so none are missed.
- D) Stop attending recess so Grace has to greet her classmates alone.
💡 Key Takeaway
Fade your physical proximity gradually so the learner responds to others, not to you.
Show rationale
Gradually increasing her distance is correct because Grace's greetings are currently controlled by the RBT's nearby presence. Systematically moving farther away across sessions fades that cue while greetings continue to contact natural reinforcement from peers, promoting independent, generalized responding. Stopping recess attendance removes support abruptly, which can cause greetings to drop off before peers' responses maintain them, and leaves no one to collect data. Staying beside Grace and prompting every greeting keeps her dependent on the RBT and blocks independence. Whispering each greeting from behind is still a prompt; it simply hides the RBT, so Grace's greetings remain tied to the technician rather than to her classmates.
Question 8
Elijah earns a token for every correct response on a sight-word target that is close to mastery. His supervisor plans to move the target into maintenance soon and has written changes to the token schedule.
Which change is consistent with preparing this target for maintenance?
- A) Switch to giving tokens only on responses that the RBT had to prompt
- B) Gradually require more correct responses before each token is delivered✓
- C) Keep giving a token after every response until the move, then stop all tokens
- D) Give a token after every response and add an edible to keep the skill strong
💡 Key Takeaway
Thin the token schedule gradually before a target moves to maintenance, never all at once.
Show rationale
Gradually requiring more correct responses per token is correct because token schedules are thinned systematically before a target enters maintenance. Thinning moves Elijah from continuous reinforcement toward an intermittent schedule, which makes responding more resistant to extinction and closer to the reinforcement he will meet in daily life. Giving a token every time and then stopping all tokens at the move is an abrupt shift to extinction that can make responding drop sharply. Adding an edible makes reinforcement denser, the opposite of preparing for maintenance. Giving tokens only for prompted responses reinforces dependence on help and withholds reinforcement from the independent reading the program is trying to maintain.
Question 9
An RBT has been teaching Omar to buckle the seat belt in his family's car. At a generalization check-in, his mother reports that on his first flight he buckled the airplane seat belt independently, even though that buckle looks different.
Which outcome does this BEST demonstrate?
- A) Response generalization to a new skill
- B) Overgeneralization to unintended stimuli
- C) Stimulus generalization across settings✓
- D) Maintenance of the skill across time
💡 Key Takeaway
The same trained response in a new setting or with new materials is stimulus generalization.
Show rationale
Stimulus generalization across settings is correct because Omar performed the trained buckling response in a new setting, an airplane, with a different-looking buckle, without being taught there. The same response occurring under new stimulus conditions is stimulus generalization, a key goal of teaching. Response generalization to a new skill is incorrect because Omar did not emit a new, untrained behavior; he buckled a belt, the response he was taught. Maintenance of the skill across time describes a behavior persisting in the original conditions after teaching ends, while car teaching is still ongoing and the key feature here is a new setting. Overgeneralization to unintended stimuli is incorrect because buckling an airplane seat belt is appropriate and desired.
Question 10
Zoe's session plan includes maintenance probes on mastered animal labels and acquisition teaching on new labels that uses prompts. The RBT is deciding how to order the activities in today's session.
When should the RBT run the maintenance probes?
- A) Right after acquisition trials, while her prompted answers are still fresh
- B) At the end of the session, once all of the new labels have been taught
- C) During acquisition trials, by prompting each mastered label as she goes
- D) At the start of the session, before any acquisition teaching begins✓
💡 Key Takeaway
Probe maintenance first, before prompted teaching can carry over.
Show rationale
Running maintenance probes at the start of the session is correct because probes should capture Zoe's independent performance before recent prompting can influence it. Prompt carryover happens when prompts from acquisition trials affect later responses, making probe data look better or worse than her true skill level. Probing right after acquisition trials invites exactly that carryover, since prompted answers are still fresh. Waiting until the end of the session, after all the new labels have been taught, also follows prompted teaching and may reflect fatigue. Prompting each mastered label during acquisition trials turns the probe into a teaching trial, so the data no longer show whether the skill has been maintained independently.