RBT Exam Cheat Sheet (2026)
Everything worth cramming the night before, on one page: the current domain weightings, the measurement procedures and their biases, IOA formulas, reinforcement schedules, the prompt hierarchy, the four differential reinforcement procedures, verbal operants and the four functions of behavior.
To save it: use your browser’s print function (Ctrl/Cmd + P) and choose “Save as PDF”. No email address required, and no download gate.
Aligned to the RBT Task List 3rd Edition (effective 1 January 2026) · Last content review: August 2026 · How we write our questions
Exam facts
- 85 questions delivered75 scored + 10 unscored pilot
- 90 minutes≈ 63 seconds per question
- 4 options per questionSingle best answer, no partial credit
- Scaled scoringNo published percentage pass mark
- Delivered at Pearson VUEPhoto ID required, personal items in a locker
- 43 task list items3rd Edition, effective 1 January 2026
| Domain (3rd Edition) | Formerly | Weight | Scored Qs |
|---|---|---|---|
| A. Data Collection and Graphing | Measurement | ~17% | 13 |
| B. Behavior Assessment | Assessment | ~11% | 8 |
| C. Behavior Acquisition | Skill Acquisition | ~25% | 19 |
| D. Behavior Reduction | — | ~19% | 14 |
| E. Documentation and Reporting | Documentation | ~13% | 10 |
| F. Ethics | Professional Conduct and Scope of Practice | ~15% | 11 |
| Total | 100% | 75 |
The real exam delivers 85 questions in 90 minutes: 75 scored plus 10 unscored pilot questions that do not count toward your result.
Measurement procedures
- Frequency
- Count of occurrences. Needs discrete behavior of consistent length.
- Rate
- Frequency ÷ observation time. Use when session lengths differ.
- Duration
- How long the behavior lasts. Can be totalled across a session.
- Latency
- Stimulus onset → response *onset*. Not to completion.
- IRT
- End of one response → start of the next.
- Partial interval
- Occurred at any point in the interval. Overestimates.
- Whole interval
- Occurred for the entire interval. Underestimates.
- Momentary time sampling
- Occurring at the moment the interval ends.
- Permanent product
- Measures the outcome left behind, not the behavior.
IOA formulas
- Total count IOA
- Smaller count ÷ larger count × 100. Never exceeds 100%.
- Interval-by-interval IOA
- Agreements ÷ (agreements + disagreements) × 100.
Reinforcement schedules
- CRF (FR1)
- Every response. Fastest acquisition, fastest extinction.
- FR
- Fixed number of responses. Post-reinforcement pause.
- VR
- Average number. Highest steady rate, most resistant to extinction.
- FI
- First response after fixed time. Scalloped pattern.
- VI
- First response after average time. Steady moderate rate.
Prompt hierarchy (least → most intrusive)
- 1. Gestural
- Pointing, glancing toward the correct response.
- 2. Verbal
- A spoken cue.
- 3. Model
- Demonstrating the response.
- 4. Partial physical
- Light touch or guidance at part of the movement.
- 5. Full physical
- Hand-over-hand. Most intrusive.
- Most-to-least
- Start intrusive, reduce. Best for brand-new skills.
- Least-to-most
- Start light, add help after an error. Best for partial skills.
- Constant time delay
- Fixed pause before prompting, every trial.
- Progressive time delay
- Pause lengthens across trials.
Differential reinforcement
- DRA
- Reinforce a specific alternative producing the same reinforcer.
- DRI
- Reinforce a response physically incompatible with the problem behavior.
- DRO
- Reinforce the absence of the behavior for an interval.
- DRL
- Reinforce the behavior at a lower rate.
Verbal operants
- Mand
- Request. Controlled by an MO, reinforced by the specific item.
- Tact
- Label something present. Reinforced socially.
- Echoic
- Repeat what was just said. Point-to-point correspondence.
- Intraverbal
- Answering / conversation. No point-to-point correspondence.
Functions of behavior
- Attention
- Behavior produces social attention, including reprimands.
- Escape
- Behavior removes or postpones a demand.
- Tangible
- Behavior produces access to an item or activity.
- Automatic
- Behavior produces its own consequence; occurs when alone.
Antecedent interventions
- NCR
- Reinforcer delivered on a time schedule regardless of behavior.
- High-p sequence
- Easy requests before a hard one (behavioral momentum).
- Choice
- Offering acceptable options before work begins.
- Priming
- Preparing the client for what is coming next.
The mistakes that cost marks
Collected from every study guide section — these are the errors that turn up most often.
A. Data Collection and Graphing
- Confusing latency (stimulus → start of response) with duration (start → end of response).
- Reporting a raw count when the question asks for rate, or vice versa.
- Forgetting that partial interval overestimates and whole interval underestimates.
- Inverting an IOA calculation and producing a figure above 100%.
- Recording an inference ("was frustrated") in a column that requires an observation.
B. Behavior Assessment
- Answering a parent’s diagnostic or prognosis question instead of redirecting to the supervisor.
- Treating a preference assessment result as proof that something is a reinforcer.
- Confusing an abolishing operation (satiation) with extinction.
- Prompting during an assessment, which measures prompted rather than independent performance.
- Miscounting paired-stimulus pairings by forgetting to divide by two.
C. Behavior Acquisition
- Confusing negative reinforcement (removal that increases behavior) with punishment.
- Calling a procedure "shaping" when the client can already perform the response.
- Mixing up response generalization with stimulus generalization.
- Forgetting that continuous reinforcement produces the *least* resistance to extinction.
- Missing an unintended prompt — position bias, size cues, the instructor’s gaze.
- Identifying a verbal operant from the words rather than from what controls the response.
D. Behavior Reduction
- Applying attention extinction to escape-maintained behavior.
- Stopping a procedure during an extinction burst.
- Delivering the reinforcer at the end of a DRO interval in which the behavior occurred.
- Choosing a replacement behavior that produces the wrong reinforcer.
- Lecturing a client whose behavior is attention-maintained.
- Using an unapproved physical management technique during a crisis.
E. Documentation and Reporting
- Writing an inference or a diagnostic claim into a session note.
- Storing or transporting client data on a personal device.
- Assuming that omitting a name makes a disclosure anonymous.
- Reconstructing data from memory to meet a deadline.
- Correcting a record by deleting rather than amending it.
- Answering a prognosis question instead of routing it to the supervisor.
F. Ethics
- Accepting a request because it came from a parent and seemed harmless.
- Assuming a supervisor can authorise an ethics-code violation.
- Continuing to practise during a supervision lapse or after certification expires.
- Treating "no name mentioned" as sufficient for confidentiality.
- Not reporting a colleague because the client is not yours.
- Asking a family to adapt to the clinic rather than adapting to the family.
Cramming is not a readiness check
This page helps you recall terminology. It cannot tell you whether you can apply it to a scenario in 63 seconds. Take a full-length timed exam and read the per-domain breakdown.
Take the 85-question exam →