RBT Exam Revision Course: Behavior Acquisition

Reinforcement procedures are used to increase the future likelihood of behaviour by arranging consequences that strengthen responding. As an RBT, your responsibility is to implement reinforcement exactly as written in the skill acquisition plan or behaviour support plan, including the timing, contingency, and schedule specified by the supervisor.

• Positive reinforcement: a stimulus is added after the behaviour and the behaviour increases in the future under similar conditions.
• Negative reinforcement: a stimulus is removed, delayed, or avoided after the behaviour and the behaviour increases in the future under similar conditions.

Both are defined by their effect on behaviour, not by staff intention or whether the consequence “feels positive”. Reinforcement should be contingent (only after the target behaviour occurs) and typically immediate unless the plan specifies otherwise. Inconsistent delivery, delays, or “freebies” weaken learning.

Schedules matter: Continuous schedules are common in early acquisition; intermittent schedules help build persistence and support maintenance. RBTs implement the schedule that’s written and report changes in effectiveness, but do not change the schedule independently.

Example: The learner independently completes a target response within the response window. Immediately after the response, the RBT delivers 30 seconds of access to a preferred video on a continuous reinforcement schedule, and records the response and consequence as required.

🔝 Back to Top

Conditioned reinforcers acquire reinforcing value through pairing with other reinforcers. Unlike unconditioned reinforcers, their effectiveness depends on learning history and continued pairing.

• RBTs may be instructed to establish conditioned reinforcers by pairing a neutral stimulus (e.g., tokens, praise, a visual symbol) with known effective reinforcers.
• Once established, conditioned reinforcers help teaching run smoothly and can bridge delays to primary reinforcement.
• Generalized conditioned reinforcers are especially useful because they are associated with multiple backup reinforcers and therefore remain effective across many motivating conditions.

RBTs implement pairing procedures exactly as written and report changes in learner responding. You do not decide whether something “is” a reinforcer; you observe whether behaviour strengthens under the programmed conditions and report to the supervisor.

Example: Across multiple learning opportunities, the RBT delivers a token at the same time as access to a preferred edible. Over time, token delivery begins to support responding even before the edible is delivered, indicating the token is functioning as a conditioned reinforcer under the arranged conditions.

🔝 Back to Top

Discrete-trial teaching is a structured instructional method that uses repeated, clearly defined learning opportunities. Each trial includes:

• Antecedent / instruction (often with a readiness response as specified)
• Learner response within the response window
• Scripted consequence (reinforcement for correct responding, or the programmed correction procedure)
• Intertrial interval

DTT is used when skills require precision, repetition, and strong stimulus control. RBTs implement DTT by presenting instructions exactly as written, delivering prompts and error correction as scripted, recording data trial-by-trial, and maintaining the planned pace of instruction.

Example: The RBT presents “Point to red” with a three-card array. The learner touches red within five seconds. The RBT delivers the programmed reinforcer immediately, records the response as independent and correct, waits the specified intertrial interval, and presents the next trial.

🔝 Back to Top

Naturalistic teaching embeds instruction within ongoing activities and naturally occurring interactions. These procedures emphasise motivation, learner choice, and functional use of skills in real-world contexts.

• Teaching opportunities are captured or arranged when the learner shows interest.
• Reinforcement is typically directly related to the response (e.g., the mand produces the item).
• Even though it’s flexible, it’s still systematic: prompting, response requirements, reinforcement, and data collection follow the written plan.

Example: During free play, the learner reaches for a preferred toy. The RBT briefly withholds access and prompts a mand as written. When the learner emits the target mand, the RBT provides access to the toy immediately and records the response according to the plan.

🔝 Back to Top

Task analysis breaks a complex skill into smaller, sequential steps that can be taught systematically. Chaining procedures teach those steps in an organised manner.

• Forward chaining: teach the first step first; add later steps as mastery is met.
• Backward chaining: teach the final step first so the learner contacts natural reinforcement early.
• Total-task chaining: teach all steps each opportunity, prompting as needed.

RBTs implement the chaining method selected by the supervisor and follow the prompting and reinforcement procedures for each step exactly as written.

Example: Using backward chaining for handwashing, the RBT prompts all steps except turning off the tap. The learner completes the final step independently and contacts reinforcement. Over sessions, earlier steps are transferred to the learner according to the plan.

🔝 Back to Top

Discrimination training teaches learners to respond differently to stimuli by arranging differential reinforcement across instructional contexts.

• A response is under stimulus control when it reliably occurs in the presence of an SD and occurs less, or not at all, under SΔ conditions.
• SD: signals reinforcement is available for a particular response (based on learning history).
• SΔ: signals reinforcement is not available for that response at that time.

At the RBT level, discrimination training means presenting the planned instruction and stimuli, delivering reinforcement only for the correct response under the current SD, withholding reinforcement for other responses, and recording performance accurately.

Example: With a cat/dog/bird array, the RBT says “Touch the bird.” Touching bird contacts reinforcement; touching cat or dog does not. On the next trial, “Touch the cat” changes which response contacts reinforcement.

🔝 Back to Top

Prompts increase the likelihood of correct responding during acquisition. Prompts can be:

• Stimulus prompts: changes to materials/environment that make the correct stimulus more salient.
• Response prompts: assistance that directly helps the learner emit the correct response (verbal, gestural, model, physical).

Prompting procedures must include a fading plan. Fading is the systematic removal or reduction of prompts so responding comes under the control of the natural antecedent rather than the prompt. Without fading, prompt dependence can develop.

Common procedures (as selected by the supervisor) include least-to-most, most-to-least, time delay, stimulus fading, and errorless learning arrangements. RBTs implement the specified procedure exactly and do not introduce extra prompts or fade independently.

Example: “Touch the circle” begins with a positional prompt (circle closer). Across trials, the positional advantage is removed according to plan so the learner responds to the instruction alone.

🔝 Back to Top

Generalization is when a learned behaviour occurs across different people, settings, materials, or stimuli without additional training. It often needs to be planned and programmed rather than assumed to “just happen.”

RBTs implement generalization procedures exactly as written. This may include teaching across multiple exemplars, varying materials, running sessions in different environments, or rotating instructors.

Example: After the learner masters picture-card requesting at the table, the RBT practises the same requests during snack, playtime, and with different instructors as specified, recording performance across contexts.

🔝 Back to Top

Acquisition procedures teach new skills that are not yet in the learner’s repertoire. They often involve dense reinforcement, more frequent prompting, and structured teaching arrangements.

Maintenance procedures ensure previously acquired skills continue over time. They often involve thinner reinforcement schedules, fewer prompts, and less frequent trials.

RBTs need to recognise that acquisition vs maintenance affects expectations, data collection, and what procedures are appropriate. Whether a skill is in acquisition or maintenance is determined by the supervisor.

Example: A new imitation skill is taught with prompting and continuous reinforcement (acquisition). After mastery, the skill is practised less frequently and reinforced intermittently (maintenance), as written.

🔝 Back to Top

Shaping is a behaviour change procedure in which successive approximations of a target behaviour are differentially reinforced. Each approximation must be closer to the final target response than the previous one.

Shaping is used when the target behaviour does not currently occur in the learner’s repertoire. Reinforcement is shifted from earlier approximations to later, more refined forms of the behaviour.

RBTs implement shaping exactly as written. Decisions about which approximations count and when to shift criteria are made by supervisors.

Example: To shape vocal requesting, early approximations (any vocal sound) are reinforced. Over time, reinforcement is delivered only for clearer approximations until the full word occurs reliably.

🔝 Back to Top

A token economy is a system in which tokens are delivered contingent on target behaviours and later exchanged for backup reinforcers. Tokens function as conditioned reinforcers and must be paired with backup reinforcement to maintain value.

• Tokens: conditioned reinforcers delivered immediately after target behaviour
• Backup reinforcers: what tokens buy (items, activities, privileges)
• Exchange ratio: how many tokens are required for each backup reinforcer
• Exchange timing: when exchange is allowed (often scheduled)

RBTs implement token economies exactly as designed and do not change token values, exchange rates, or the menu of backup reinforcers without supervisor direction.

Example: The learner earns one token for each completed task. After earning the specified number of tokens, the learner exchanges them for a preferred activity according to the programme rules.

🔝 Back to Top