BCBA Practice Questions (2026) — Free Sample Questions with Answers

Answer: D

Reflexivity: Reflexivity is identity matching (A-A, B-B, C-C); C-A involves different stimuli.

Symmetry: Symmetry is the reversal of a single trained relation (B-A or C-B); C-A spans two trained relations, not one.

Transitivity: Transitivity is A-C, in the trained direction across the shared stimulus; C-A also requires reversal, so it tests more than transitivity.

Equivalence: Correct. C-A reverses the direction of training across the shared stimulus (B), so it requires both symmetry and transitivity; this combined test is the test for equivalence.

After training A-B and B-C: reflexivity A-A; symmetry B-A and C-B; transitivity A-C; equivalence (combined) C-A.

Answer: D

Reflexive conditioned motivating operation: A CMO-R is a warning stimulus that makes its own removal reinforcing; the child is not trying to escape the container.

Discriminative stimulus for requesting help: An SD signals that a response is more likely to be reinforced than in its absence; adults will help at any time, so the container alters the value of help rather than its availability.

Surrogate conditioned motivating operation: A CMO-S gains its effect by being paired with another MO; the container's effect comes from its relation to the crackers, not from such pairing.

Transitive conditioned motivating operation: Correct. The closed container establishes adult help (opening it) as a reinforcer and evokes the request for help. Help is equally available at all times, so the container changes the value of help rather than signaling its availability: a CMO-T.

Ask whether the stimulus signals that reinforcement is available (SD) or makes some other stimulus valuable as a reinforcer (CMO-T).

Answer: C

Concurrent schedule: Concurrent schedules make two or more schedules available at the same time for different responses; here only one component is in effect at a time.

Mixed schedule: A mixed schedule also alternates components, but with no stimulus correlated with each; here each component is signaled by a colored card.

Multiple schedule: Correct. Two schedules (reinforcement and extinction) alternate, each correlated with a distinct stimulus (green card as SD, red card as S-delta), and breaks are available within the reinforced component.

Chained schedule: In a chained schedule, completing one component produces the stimulus for the next and reinforcement follows only the final component; here breaks are delivered during the green component itself.

Signaled alternating components = multiple; unsignaled = mixed; sequential with terminal reinforcement = chained (signaled) or tandem; simultaneous options = concurrent.

Answer: D

The undercount results from using total-count rather than interval IOA: The IOA formula cannot cause or fix a shared counting error. A different IOA method would still compare two observers who both missed the same mands.

The data are accurate, because agreement between observers is high: High IOA does not establish accuracy. Two observers can agree closely while both miss the same responses.

The data are invalid, because the observers measured the wrong behavior: Validity concerns whether the right behavior and dimension were measured. The RBTs measured mands, the intended behavior; they undercounted it, which is an accuracy problem.

The observers agree with each other, but the data are not accurate: Correct. IOA shows only that two observers recorded similar values. Accuracy is the match between recorded and true values, here checked against a video standard. Both observers made the same error, so they agree but are inaccurate.

IOA shows agreement between observers, not accuracy; accuracy requires comparing recorded data with a true value such as a scored video.

Answer: C

1. 1. Compare the two records interval by interval: they disagree on interval 2 (+ vs –) and interval 6 (– vs +).
2. 2. Agreements = 10 − 2 = 8 intervals.
3. 3. Interval-by-interval IOA = agreements ÷ (agreements + disagreements) × 100 = 8 ÷ 10 × 100 = 80%.

60%: This is unscored-interval IOA: of the 5 intervals that either observer left unscored (2, 3, 5, 6, 9), they agreed on 3.

71%: This is scored-interval IOA: of the 7 intervals that either observer scored (1, 2, 4, 6, 7, 8, 10), they agreed on 5.

80%: Correct. The observers agree on 8 of 10 intervals (they disagree only on intervals 2 and 6); 8 ÷ 10 × 100 = 80%.

100%: This compares totals: each observer scored 6 intervals. Matching totals hide the disagreements on intervals 2 and 6.

Interval-by-interval IOA = intervals agreed ÷ total intervals × 100; equal totals do not mean the observers agreed on the same intervals.

Answer: B

1. Tier 1 (Greeting): stable low baseline (0–1); immediate rise to 4–9 at session 5.
2. Tier 2 (Asking for help): stable low through session 4, then rises to 3–7 in sessions 5–8 while still in baseline, before its intervention at session 9.
3. Tier 3 (Turn-taking): stable low baseline through session 10; rises only after its intervention at session 11.
4. Conclusion: the change in Tier 2 during baseline (covariation) weakens experimental control; the effect is verified and replicated only by Tiers 1 and 3.

Control is strong, because each behavior rose after its intervention began: This overlooks Tier 2. Asking for help rose during sessions 5–8, before its own intervention started, so not every change followed the IV.

Control is weakened, because Tier 2 rose before its intervention began: Correct. When Tier 1 entered intervention, Tier 2 (still in baseline) also rose. The behaviors were not independent, so that tier cannot verify the effect; only Tiers 1 and 3 change when, and only when, the intervention starts.

The design is flawed, because the tiers have baselines of different lengths: Staggered baselines of increasing length are a required feature of a multiple-baseline design, not a flaw.

A functional relation is shown by Tier 1 alone, without the later tiers: One tier is an AB comparison. Without verification and replication in later tiers, it cannot show a functional relation.

In a multiple-baseline design, each untreated tier must stay stable while earlier tiers change; covariation in a baseline tier weakens the demonstration of control.

Answer: B

Explain the research on independent play to gain the family's support: Educating before listening treats the BCBA's view as the standard; it is a persuasion step, not humility.

Ask about daily routines and values, then revise goals together: Correct. Cultural humility is an ongoing, self-reflective stance in which the BCBA treats the family as the experts on their values and learns from them. Goals are then chosen together, for example cooperative play with cousins in place of solitary play (2.09, 1.07).

Remove all independent-play goals so the plan fits family practices: Over-accommodating without asking may drop goals the family actually wants.

Read about the family's cultural group and adjust the plan from that: General cultural information risks stereotyping; it cannot replace asking this family what they value.

Cultural humility means asking about and learning from the family's values and routines rather than assuming; goals are then set together.

Answer: A

Explain the testimonial standard and propose other website content: Correct. Code 5.07 prohibits soliciting testimonials from current clients or stakeholders for advertisements, because of undue influence and implicit coercion. The BCBA explains this to the director and offers compliant alternatives (e.g., outcome summaries without client quotes); keeping a record of the conversation is also good practice.

Collect the quotes and label each one on the website as solicited: Labelling solicited or unsolicited is a 5.08 requirement for former-client testimonials, not a fix for current clients.

Replace them with quotes from former clients, posted without any other steps: Former-client testimonials are allowed, but 5.08 requires disclosures, a description of use, and notice of the right to rescind.

Collect the quotes after each parent signs consent, and document it: Consent and documentation do not cure the problem; 5.07 bars soliciting current-client testimonials for advertising because of undue influence.

No soliciting current-client testimonials for advertising (5.07); former-client testimonials are allowed only with the 5.08 disclosures.

Answer: B

1. Play (control): level at 0 across all four sessions, stable.
2. Alone: 0–0.5 per minute, low and stable; no evidence of automatic reinforcement.
3. Attention: 2 in session 2, then 0.5, 0.5 and 0; a decreasing trend that overlaps control after the first session.
4. Demand: 3.5, 4, 4.5, 4; high, stable level with no overlap with any other condition.
5. Conclusion: differentiated responding in the demand condition only, so aggression is maintained by social negative reinforcement (escape).

Social positive reinforcement (attention): Misreads one early point: attention was 2 per minute in session 2 but fell to 0–0.5 and overlapped the control conditions.

Social negative reinforcement (escape): Correct. Demand is consistently elevated (3.5–4.5 per minute) and does not overlap any other condition, while play (control) stays at zero; this differentiation supports escape from demands.

Automatic reinforcement: Automatic reinforcement is suggested by responding in the alone condition or across all conditions; alone stayed near zero.

No function is identified: The demand condition is clearly differentiated from control, so the results are interpretable.

In a multielement FA, identify the condition whose level is consistently higher than the control condition with little overlap.

Answer: A

The attention may reflect its high overall rate, not a contingency: Correct. Attention followed yelling 90% of the time, which matches its 90% background rate, so the data show no contingency; an FA is needed to test function.

Attention is confirmed as the reinforcer maintaining yelling: Confuses correlation with function; descriptive data cannot confirm a maintaining reinforcer.

Escape is the likely function, because attention was ruled out: Draws a new conclusion without data: nothing in these records tests or rules out escape.

Yelling is automatically maintained, because it occurs during attention: Behavior occurring in the presence of social stimuli does not show automatic reinforcement; that needs, e.g., persistence when alone.

Compare the conditional probability of a consequence with its background probability; descriptive data show correlation only and cannot confirm function.

Answer: C

Reflexivity: Reflexivity is matching a stimulus to itself (A→A), such as matching the picture to an identical picture.

Symmetry: Symmetry is the reversal of a trained relation (B→A or C→B), such as selecting the picture and then naming it in reverse order of training.

Transitivity: Correct. A→B and B→C were trained, and the untrained A→C relation (spoken name → printed word) emerged, which is transitivity.

Combined symmetry and transitivity: The combined (equivalence) test is C→A, which reverses the transitive relation; here the probe was A→C in the trained direction.

After training A→B and B→C: B→A and C→B test symmetry, A→C tests transitivity, and C→A tests combined symmetry and transitivity (equivalence).

Answer: A

1. Baseline (sessions 1–5): problem behavior high and stable (about 4 per minute); mands near zero.
2. FCT FR 1 (6–12): immediate drop in problem behavior to near zero; mands rise to about 3 per minute. Clear treatment effect.
3. Delay 30 s (13–17): problem behavior stays near zero and mands stay stable; this thinning step was tolerated.
4. Delay 2 min (18–22): mands briefly rise then fall, while problem behavior shows an increasing trend (0.8 to 2.9).
5. Conclusion: the step from 30 s to 2 min was too large. Return to the last successful delay and thin in smaller steps.

Return to the 30-s delay, then thin in smaller steps: Correct. Problem behavior stayed low at the 30-s delay but rose steadily at 2 min. The jump was too large, so return to the last successful step and thin in smaller increments (e.g., 45 s, 60 s).

Keep the 2-min delay in place until responding stabilizes: Problem behavior is trending up at the 2-min delay; waiting lets it keep contacting reinforcement and may strengthen it.

Return to FR 1 reinforcement and restart thinning: The 30-s delay was successful, so there is no need to go all the way back to FR 1.

Add response cost for problem behavior during delays: A punishment procedure is more restrictive than needed when a reinforcement-based adjustment (smaller steps) is available (Code 2.14, 2.15).

During schedule thinning, if problem behavior rises at a new step, return to the last step at which responding was stable and use smaller increments.

Answer: D

Begin extinction during one short routine each day: Limiting extinction to one routine leaves tantrums reinforced at other times, prolonging extinction rather than reducing the burst.

Prepare caregivers to expect a burst and to keep recording: Preparing caregivers is important for integrity, but it does not reduce the size of the burst itself.

Apply extinction to about half of the tantrums at first: Withholding attention for only some tantrums is intermittent reinforcement, which can make tantrums more persistent.

Deliver attention on a dense fixed-time schedule: Correct. Combining extinction with NCR (or with reinforcement of an alternative response) provides the reinforcer through another route and reduces extinction bursts and extinction-induced aggression.

Extinction bursts and extinction-induced aggression are less likely when extinction is combined with NCR or differential reinforcement of an alternative.

Answer: B

Escape extinction with physical guidance through each writing task: Matches the function but adds no replacement behavior and is more intrusive than needed; physical guidance is hard to do in a full class.

Functional communication training for breaks, then thinning break access: Correct. FCT teaches a functional replacement, and planned schedule thinning (delays, more work before breaks) brings break frequency to a level the class can sustain.

Breaks delivered on a dense fixed-time schedule during writing: Noncontingent escape is function-based, but a dense schedule of breaks is exactly what the teacher says is not workable.

A high-probability request sequence before each writing task: An antecedent option with mixed evidence for escape behavior; used alone it teaches no replacement and may not hold up when demands are long.

Choose interventions that match the function, are supported by evidence, and can be carried out in the setting; FCT with schedule thinning fits escape behavior when breaks must be limited.

Answer: A

1. RBT 1: 55% → 92%, above the 90% criterion; 6 targets mastered.
2. RBT 2: 60% → 94%, above criterion; 5 targets mastered.
3. RBT 3: 58% → 62%, well below criterion; 1 target mastered.
4. Group mean with BST = (92 + 94 + 62) ÷ 3 ≈ 83%, which masks RBT 3's lack of change.
5. Conclusion: the supervision works for two RBTs. For RBT 3, assess why and modify the approach.

Continue for RBTs 1 and 2; assess RBT 3's barriers and adjust her plan: Correct. RBTs 1 and 2 rose from 55–60% to 92–94% integrity, above criterion, and their clients mastered 5–6 targets. RBT 3 changed little (58% to 62%), and her client mastered 1 target. Code 4.10 requires evaluating supervision effects with outcome data and adjusting promptly, so the BCBA assesses RBT 3's barriers (e.g., with a performance diagnostic) and modifies her approach.

Replace the training package for all three RBTs, since it failed for one: Ignores the clear improvement for RBT 1 and RBT 2; one non-responder does not show the package failed.

Continue as planned for all three, since mean integrity rose above 80%: The group mean (about 83%) hides RBT 3's lack of change; supervision is evaluated per supervisee.

Repeat the same training for RBT 3 until the 90% criterion is met: Repeating an approach that has not worked, without assessing why, is not a data-based change.

Evaluate supervision per supervisee, using integrity and client outcome data, and make individualized, data-based changes rather than all-or-none decisions.

Answer: B

Invite each RBT to request opportunities whenever they feel ready: Leaves access to whoever already feels confident to ask, which keeps the existing imbalance.

Plan each RBT's observations and opportunities from goals and data: Correct. Equity in supervision means that each supervisee gets the support and opportunities their goals require, tracked with data so that familiarity does not decide access (1.08, 4.07). A planned schedule makes the distribution visible and correctable.

Rotate every opportunity among the RBTs in a fixed order: Equal turns, but not individualized to each RBT's goals and needs; equality is not the same as equity.

Give leadership opportunities to the RBTs with the highest performance: Performance is shaped by the unequal feedback already given, so this rewards the inequity.

Equitable supervision allocates attention and opportunities by each supervisee's goals and data, not by similarity to the supervisor or by identical rotation.