Behavior Assessment
What this domain tests
Behavior assessment covers the methods used to identify the environmental variables that occasion and maintain behavior, and to link those findings to intervention. Questions ask you to choose the least intrusive assessment that answers the referral question, distinguish indirect, descriptive, and experimental methods, and interpret conditional probabilities rather than simple correlations.
Core rules
- Indirect methods (structured interviews, rating scales) are efficient but rely on memory and are not sufficient to identify function.
- Descriptive methods (ABC narrative or checklist, scatterplots, conditional probability) show temporal contiguity; they do not demonstrate causation.
- Experimental functional analysis (e.g., alone, attention, demand, play conditions) manipulates antecedents and consequences to isolate function.
- Preference assessments (paired stimulus, multiple stimulus without replacement, free operant) identify potential reinforcers; preference is not the same as reinforcing effectiveness.
- Always assess motivating operations and setting events; a behavior may occur only under specific EOs.
Worked example
A 7-year-old hits peers. An interview suggests attention, and ABC data show hitting is followed by teacher reprimands 80% of the time. That is still correlation. A brief functional analysis alternates attention, escape, alone, and play conditions. Hitting occurs only in the attention condition. The behavior analyst concludes attention-maintained hitting and writes a DRA: after hitting, attention is withheld (extinction) and appropriate requests for attention are reinforced. The reasoning is that the FA isolated attention as the controlling consequence; ABC alone could not rule out escape or automatic reinforcement.
Traps
Assuming topography identifies function; stopping at ABC data and calling it a functional analysis; overlooking EOs; using a default FA condition set without considering idiosyncratic reinforcers; omitting informed consent and assent.