RBT Cheat Sheet: Key Facts
Every key takeaway and common trap from the RBT study guide on one page, grouped by chapter. Tap a lesson link to read the full explanation.
41 facts · 6 chaptersFacts checked 2026-11-15
Chapter 1
Data Collection & Graphing
Key takeaways
- Latency runs from an instruction or other antecedent to the start of a response, while inter-response time runs from the end of one response to the start of the next. → Continuous measurement: frequency, duration, latency and IRT
- Partial-interval recording tends to overestimate a behavior, whole-interval recording tends to underestimate it, and momentary time sampling looks only at the end of each interval. → Interval recording and momentary time sampling
- Rate is the count divided by the observation time, so a count is meaningless without the length of the session. → Continuous measurement: frequency, duration, latency and IRT
- A phase-change line marks a change in conditions, and whether behavior changed is read from level, trend and variability on each side. → Graphs, trends and trustworthy data
- Missing data stay blank with a note, never estimated, copied forward or entered as zero, and any change in how a procedure was run is reported to the supervisor. → Graphs, trends and trustworthy data
Common trap
- Latency and IRT are easy to swap. Latency always starts at an outside event such as an instruction or a timer. IRT starts when the previous response ends. If no instruction is mentioned and the question is about pacing, the answer is IRT. → Continuous measurement: frequency, duration, latency and IRT
- A distractor that equals the raw count, or the per-minute rate when the question asks per hour, is the most common wrong answer on rate items. Check the unit before you choose. → Calculating rate, mean duration and percentage
Quick reference
| Measure | What it times or counts | Supervisor's question |
|---|---|---|
| Frequency or count | How many times it happened | How often? |
| Duration | Start to end of one instance | How long does each episode last? |
| Latency | Antecedent to start of the response | How long before the client starts? |
| Inter-response time (IRT) | End of one response to start of the next | How much pause between responses? |
| Summary | Formula | Worked example |
|---|---|---|
| Rate | Count ÷ observation time | 18 requests in 30 minutes = 0.6 per minute = 36 per hour |
| Mean duration | Total duration ÷ number of episodes | 5 episodes totaling 20 minutes = 4 minutes each |
| Percentage | Part ÷ total × 100 | 14 correct out of 20 trials = 70 percent |
Chapter 2
Behavior Assessment
Key takeaways
- In a multiple-stimulus assessment without replacement, the chosen item is removed, so the order of choices ranks the whole set in one pass. → Preference assessment formats
- Free-operant observation removes nothing and measures how long the learner engages with each item, which suits learners who react badly to losing an item. → Preference assessment formats
- A preference assessment identifies likely reinforcers; only a change in behavior when the item is delivered contingently shows that it is a reinforcer. → What preference results mean and why they change
- Interviews and rating scales are indirect, ABC recording is descriptive, and only a functional analysis that arranges and compares conditions can demonstrate function. → Indirect, descriptive and experimental functional assessment
- The RBT runs procedures exactly as written, records what was observed, and sends questions about causes or changes to the supervisor. → ABC recording, common functions and skill assessments
Common trap
- Options that say a preference result "proves" an item is a reinforcer, or that more of the item will restore responding, are wrong. Preference suggests a reinforcer; behavior change proves it. → What preference results mean and why they change
Quick reference
| Format | How it works | What you get |
|---|---|---|
| Multiple stimulus without replacement | Array of items; chosen item is removed and the rest are rearranged | Full ranking, quickly |
| Multiple stimulus with replacement | Chosen item goes back into the array | One strong favorite, weak ranking of the rest |
| Paired stimulus | Two items at a time, every item paired with every other, sides alternated | Ranking, but many trials |
| Single stimulus | One item at a time; record approach or turning away | Approach yes or no, no ranking |
| Free-operant observation | Items freely available, nothing removed, no demands | Time engaged with each item |
Chapter 3
Behavior Acquisition
Key takeaways
- Reinforcement always increases behavior; positive adds a stimulus, negative removes one, and it works best when delivered immediately after the response. → Reinforcement, conditioned reinforcers and motivation
- Continuous reinforcement builds a new skill quickly, and the plan then thins it gradually to an intermittent schedule, with variable ratio producing the most persistent responding. → Schedules of reinforcement and token economies
- Naturalistic teaching follows the learner's lead and uses the natural reinforcer, while discrete trial teaching is adult-led with short, separate trials. → Discrete trials, naturalistic teaching, generalization and maintenance
- Prompts must be faded, for example with a time delay, or the learner becomes prompt dependent. → Prompting, prompt fading and discrimination training
- Chaining links several steps into a sequence, shaping refines one response through closer approximations, and the RBT runs whichever the plan specifies without changing it. → Chaining and shaping
Common trap
- "Negative" does not mean bad. Options that call removal of a task punishment are wrong when the behavior went up. Ask one question: did the behavior increase or decrease? → Reinforcement, conditioned reinforcers and motivation
- A learner who waits for help on skills he has done alone before is prompt dependent. Prompting more, or prompting twice, makes it worse. The plan's fading step, such as a time delay, is the answer. → Prompting, prompt fading and discrimination training
- If the options mix up chaining and shaping, remember that chaining links several different behaviors while shaping changes the form of one behavior. Every chaining method, including backward chaining, still needs a written task analysis. → Chaining and shaping
Quick reference
| Schedule | Rule | Typical effect |
|---|---|---|
| Fixed ratio | Set number of responses | High rate with a pause after reinforcement |
| Variable ratio | Number of responses varies around an average | High, steady rate; most resistant to extinction |
| Fixed interval | First response after a set time | Pausing after reinforcement |
Chapter 4
Behavior Reduction
Key takeaways
- Function is identified by the consequence that follows the behavior: attention, escape, access to a tangible, or automatic reinforcement. → Functions of behavior and the written plan
- Noncontingent reinforcement delivers the maintaining reinforcer on a timer regardless of behavior, which lowers the motivation for the problem behavior. → Antecedent interventions
- DRA reinforces a specific replacement, DRI reinforces a response that cannot occur at the same time, DRO reinforces the absence of the behavior, and DRL reinforces a lower rate. → Differential reinforcement and functional communication training
- Extinction must withhold the reinforcer that actually maintains the behavior, every time, and an early extinction burst is expected. → Extinction and its side effects
- In a crisis, safety comes first through the trained emergency protocol, and every plan change is the supervisor's decision. → Time-out, crisis procedures and staying within the plan
Common trap
- DRA and DRI are easy to confuse. Ask whether both behaviors could happen at once. If a learner could raise a hand and shout at the same time, it is DRA, not DRI. → Differential reinforcement and functional communication training
Quick reference
| Function | What follows the behavior |
|---|---|
| Attention | Someone looks, talks, comforts or even scolds |
| Escape | A demand or task is removed or delayed |
| Tangible | A preferred item or activity is given or returned |
| Automatic | Nothing social; the behavior produces its own stimulation and persists when alone |
| Procedure | What earns reinforcement | Example |
|---|---|---|
| DRA (alternative) | A specific appropriate replacement | Tapping a shoulder gets attention; grabbing does not |
| DRI (incompatible) | A response that physically cannot happen at the same time | Hands on the steering toy instead of hitting |
| DRO (other) | No occurrence for a whole interval; timer resets if it occurs | Token after 5 minutes with no spitting |
| DRL (low rate) | Behavior stays at or below a set limit | Bonus if no more than 4 trips to the pencil sharpener |
Chapter 5
Documentation & Reporting
Key takeaways
- An objective note describes what an observer could see, hear and count, and leaves out feelings, motives and character labels. → Writing objective session notes
- Record the time actually worked and the data actually collected, and never copy notes, estimate data or fill gaps. → Accurate records, corrections and missing data
- Correct paper records with a single line, the right value, your initials and the date, so the original stays readable. → Accurate records, corrections and missing data
- Health, safety and serious incidents go to the supervisor right away, while routine logistics can wait for the next supervision contact. → What to report to the supervisor, and how fast
- Client information stays in approved, secured channels and storage, and is never posted on social media, even with permission. → Confidentiality, storage and approved channels
Common trap
- A note with accurate counts can still fail if one phrase is a label, such as "clearly defiant." Read the whole sentence; one interpretive word makes it the wrong answer. → Writing objective session notes
- Answers that add a condition, such as "with the parent's permission" or "without using the name," look reasonable but are wrong for social media and lobby questions. No condition makes posting or casual discussion acceptable. → Confidentiality, storage and approved channels
Quick reference
| Not objective | Objective |
|---|---|
| He was bored and lazy during math. | He put his head on the desk for about 4 minutes and completed 2 of 8 problems. |
| She hit to get attention. | She hit the RBT's arm 3 times during the reading task. |
| Dad never follows the plan. | Dad reported that the visual schedule was not used at home this week. |
Chapter 6
Ethics
Key takeaways
- Interpreting results, choosing goals and changing plans belong to the supervisor, and the RBT routes those questions there. → The RBT's role and scope of practice
- An RBT runs a procedure only after the supervisor confirms competence, and restrictive procedures also must be in the documented plan. → The RBT's role and scope of practice
- Even confirming that someone is a client is a disclosure, and client photos or information never go on social media. → Confidentiality, dignity and social media
- Personal relationships, paid side work and gifts worth more than $10 with client families are declined, reported to the supervisor and documented. → Multiple relationships and gifts
- Avoiding harm comes before procedural fidelity, and the legal duty to report suspected abuse cannot be cancelled by anyone else. → Do no harm, client rights and reporting concerns
Common trap
- Options that make the family responsible, such as "tell the parents to ask the supervisor," are wrong. The RBT declines, informs the supervisor and documents, because raising a multiple relationship is the technician's job. → Multiple relationships and gifts
- "Continue exactly as written, since only the supervisor may change it" sounds like good scope behavior, but it is wrong when the step is causing harm. Protect the client first, then report. → Do no harm, client rights and reporting concerns