BLEPP common mistakes
BLEPP · Understand the traps and how to avoid them.
Choosing the answer that describes the most clinically sophisticated or comprehensive intervention rather than the one that matches the least restrictive, least intrusive effective option described in the scenario.
Candidates train in intensive behavioral programming and instinctively equate 'more thorough' with 'better,' so they pick restraint, punishment, or full-package plans when the vignette calls for reinforcement-based, socially valid alternatives. Ethics items reward the least restrictive effective procedure.
What to do instead
Before answering, ask 'what is the least intrusive procedure that would likely work here, and did the behavior analyst exhaust less restrictive options first?' Rank the four options from least to most restrictive and select the least restrictive one that is still effective. Practice by writing, for 20 ethics vignettes, a one-line justification naming the principle (least restrictive, right to effective treatment, dignity) that drove your choice.
Confusing negative reinforcement with punishment whenever an aversive stimulus appears in the item stem.
The word 'negative' colloquially means 'bad,' and many candidates memorize 'negative = take away' without anchoring the contingency to the future frequency of the target behavior. Items deliberately include escape/avoidance contingencies where a stimulus is removed and behavior increases.
What to do instead
For every contingency item, draw the three-term contingency in the margin: antecedent, behavior, consequence, then label the consequence by its effect on future behavior (increase = reinforcement, decrease = punishment) and by operation (add = positive, remove = negative). Do 30 rapid-fire flashcards where you must state the process name and the effect direction before checking the answer.
Reporting interobserver agreement as the sole evidence of measurement quality and ignoring whether the measure itself matches the behavior's dimensional qualities (e.g., using frequency for duration-based behavior, or partial-interval to estimate rate).
IOA is heavily emphasized in training, so candidates default to 'check IOA' for any measurement-quality question. Exam items test whether the selected measurement system captures the relevant dimension and is sensitive to change, which IOA cannot fix.
What to do instead
Build a decision table: behavior dimension (repeatability, temporal extent, temporal locus) mapped to the appropriate measure (count/rate, duration, latency, IRT), then a second column mapping observation method (event, interval, time sampling) to its known bias (partial interval overestimates, whole interval underestimates, momentary time sampling estimates duration). Answer measurement items by first naming the dimension, then the measure.
Selecting a reversal or multiple-baseline design without checking whether the design's preconditions hold in the vignette (irreversibility, ethical concerns about withdrawing treatment, number of baselines, staggered introduction feasibility).
Candidates memorize design names and their generic strengths but do not run the design-selection logic against scenario constraints, so they choose ABAB for a behavior that cannot ethically or practically be reversed, or multiple baseline when only one tier exists.
What to do instead
For each design, write the precondition checklist: reversal requires reversibility and ethical acceptability of withdrawal; multiple baseline requires at least three independent but similar tiers and untreated tiers remaining stable; alternating treatments requires rapid and discriminable alternation and no carryover; changing criterion requires stepwise gradual change. When reading a design item, first list the constraints stated in the stem, then eliminate designs whose preconditions are violated.
Writing or evaluating behavioral definitions that are circular or that use internal-state language ('aggressive,' 'anxious,' 'off-task because he is distracted') instead of observable, measurable movement.
Candidates use everyday shorthand labels and assume shared meaning, so they miss items where two observers would disagree because the definition lacks boundaries, examples, and non-examples.
What to do instead
Apply a three-part test to every definition: Can two observers score it independently and agree? Does it describe what the person does (observable movement) rather than an inference? Does it specify onset, offset, and what is excluded (non-examples)? Rewrite 15 vague labels into operational definitions with a testable example and non-example, and practice spotting the option that adds a boundary condition.
Misreading graph-based items by judging trend by eye without segmenting data into condition phases, or confusing level, trend, and variability when the item asks which feature supports experimental control.
Candidates look at the overall slope across the whole graph rather than within-phase and between-phase comparisons, and they conflate a change in level with a change in trend. Under time pressure they pick the option mentioning 'increasing trend' regardless of phase boundaries.
What to do instead
Adopt a fixed graph-reading routine: identify phase boundaries and condition labels, compute or estimate within-phase level (mean/median), trend (slope), and variability (range), then compare adjacent phases for immediacy, magnitude, and consistency of change. Practice on 20 graphs, writing the three features for each phase before looking at the answer options.
Answering supervision items by choosing the most supportive or least confrontational response instead of the response that fulfills the supervisor's gatekeeping and documentation obligations.
Candidates want to preserve the supervisory relationship and avoid appearing punitive, so they pick 'discuss informally and monitor' when the correct action is to document, retrain, and restrict independent practice until competency is demonstrated.
What to do instead
Memorize the supervisor's obligation sequence: assess competency directly (observation, not self-report), provide written feedback and a remediation plan with measurable criteria, document, and escalate or restrict duties if competency is not met. For each supervision item, ask 'what would a licensing board expect to see in the file?' and choose the option that creates a verifiable record.
Missing the transfer-of-stimulus-control and generalization-planning distinction, selecting 'train in the natural environment' for items that actually ask about programming common stimuli, teaching loosely, or indiscriminable contingencies.
Both concepts involve real-world settings, so candidates treat them as interchangeable and pick the first option mentioning natural environment regardless of the specific tactic the item targets.
What to do instead
Create a taxonomy card with each generalization tactic and a one-line operational description: teach sufficient exemplars, train loosely, program common stimuli, mediate generalization, train to generalize, indiscriminable contingencies, natural maintaining contingencies. For each item, identify the tactic named or implied in the stem before reading options, then match.
Treating consultation and systems items as individual clinical decisions, ignoring organizational contingencies, stakeholder assent, and the layers (client, caregiver, organization, community) at which the problem is maintained.
Training focuses on one-to-one clinical work, so candidates default to designing a client-level intervention when the stem describes a staff-turnover, resource, or caregiver-implementation problem that requires a systems-level analysis.
What to do instead
For every consultation item, draw a nested diagram (client within family within classroom within organization) and ask at which level the controlling contingencies operate. Then check the option for stakeholder collaboration, contextual fit, and measurement of implementation fidelity, not just client outcomes. Practice by sorting 20 vignettes by intervention level before choosing.
Selecting a response to an ethical dilemma based on personal values or loyalty to the family/employer rather than on the code-mandated sequence (identify the conflict, consult the code, seek consultation, attempt resolution within the organization, then escalate to the board).
Candidates face realistic pressures (parent requests for a specific diagnosis, employer demands for unbillable services) and pick the option that preserves the relationship, forgetting that the code specifies escalation and documentation when informal resolution fails.
What to do instead
Memorize the escalation ladder and apply it mechanically: identify the relevant code standard, consult the code and a knowledgeable colleague, attempt resolution directly and in writing, follow organizational channels, and report to the licensing body only when required or when resolution fails. For each ethics item, name the specific standard implicated and the step in the ladder before reading options.