BCBA Domain B: Concepts & Principles

Verbal operants are the most frequently tested slice of B, and they are classified by their antecedent and their reinforcer — never by what the words sound like. Four carry most items:

• Mand — evoked by a motivating operation and reinforced by its specific consequence. A child deprived of water who says “Water!” and gets water is manding. When a play piece goes missing and the client asks “Where is the final piece?”, that question is a mand too — an establishing operation for missing information, reinforced by the specific answer.
• Tact — evoked by a nonverbal stimulus and reinforced by generalized social reinforcement (praise, a token). Labeling a spoon as “spoon” for a “nice job” is a tact.
• Intraverbal — evoked by a verbal antecedent with no point-to-point correspondence (“What do you drink when thirsty?” → “Water”). If there is no verbal antecedent, it is not an intraverbal.
• Echoic — a verbal antecedent with point-to-point correspondence: the speaker repeats a vocal model.

Real language is rarely pure. A thirsty child who points at a water fountain and says “Water!” is under multiple control — the deprivation supplies mand strength, the fountain supplies tact strength, so the response is an impure, multiply-controlled operant, not a “pure mand” or “pure tact.” The exam rewards you for reading both sources of control rather than forcing the response into one box.

Reinforcement always increases future behavior; punishment always decreases it. Beyond that, B tests two independent axes you have to read separately:

• Positive vs negative — was a stimulus added (positive) or removed (negative)? Both can reinforce. A student who screams during math, has the worksheet taken away, and screams more next time is on negative reinforcement — the demand was removed.
• Socially mediated vs automatic — did another person deliver the consequence (socially mediated), or did the behavior produce it directly (automatic)? The teacher removing the worksheet makes that example socially mediated negative reinforcement.

The exam loves to bury the real contingency under a salient distractor. In the screaming example the teacher also hands over an iPad — but the functional analysis shows the behavior stays high whenever the demand is removed, iPad or not. The added iPad is a decoy; the maintaining variable is escape from the demand. Read the FA data, not the most visible event.

Stimulus control is about when a behavior pays off. A discriminative stimulus (S^D) signals that reinforcement is available and evokes the behavior; a stimulus delta (S^Δ) signals it is unavailable, so responding drops in its presence. In the operant-chamber classic, the green light under which lever-presses pay is the S^D; the red light under which they no longer pay — and pressing stops — is the S^Δ. It is neither a punisher nor an eliciting stimulus; it simply signals extinction of that response.

Do not confuse a signaling stimulus with a motivating operation (MO). An S^D changes the probability a behavior occurs by signaling availability; an MO changes the value of a consequence and evokes behavior that has produced it. An establishing operation (deprivation) raises value; an abolishing operation (satiation) lowers it. That value-versus-availability split is exactly what separates a mand (MO control) from a tact (S^D control) in the verbal-operant items above.

Schedule items give you a rate-and-pattern description and ask for the mechanism. Learn the signatures:

• Fixed ratio (FR) — a high, steady “ratio run” then a postreinforcement pause right after reinforcement is consumed. A token board paying off after a fixed number of responses (5 envelopes per token × 10 tokens = an FR 50) produces exactly this: fast work, then a pause, then fast work again.
• Ratio strain — when the requirement is stretched too thin too fast, responding breaks down during the run: pausing, avoidance, emotional behavior. Note the timing — strain shows up mid-run, a postreinforcement pause shows up after payoff.
• Variable schedules — steady responding without large predictable pauses.

The distractor here is mentalism: staff call the FR pause “task fatigue” or “satiation.” Both are explanatory fictions layered on a predictable, schedule-induced pattern. The behavioral answer names the schedule, not a feeling.

Two kinds of generalization are routinely swapped in the options:

• Stimulus generalization — the same response is evoked by novel stimuli that share physical features. A learner taught to tact flashcards as “vehicle” who later calls a never-seen delivery van “vehicle” is showing stimulus generalization: identical topography, new antecedent.
• Response generalization — a novel response topography serves the same function.

The tell is whether the stimulus changed or the response changed. Same word to a new object is stimulus generalization; a new way to accomplish the same outcome is response generalization. And beware over-claiming: if a probe adds prompts, changes reinforcement and moves rooms all at once, you cannot attribute the gain to stimulus control at all — the honest answer is inconclusive until the confounds are stripped out.

Most Domain B misses come from a handful of recurring confusions. These are the ones the real question explanations flag again and again: