What a Functional Analysis Is and Why It Matters
A functional analysis (FA) is an assessment in which environmental variables are systematically manipulated while the occurrence of problem behavior is measured. Unlike indirect assessments (interviews, rating scales) or descriptive assessments (ABC data collection), an FA is the only method that demonstrates a functional relation between behavior and its maintaining consequence. On the BCBA exam, you will be expected to know when an FA is appropriate, how its conditions are structured, and how to interpret the resulting graph.
Step 1: Complete Pre-Assessment Work
Before running any FA, gather information from indirect and descriptive assessment. Interview caregivers, review records, and observe the client in natural settings. This information helps you select the conditions to run, identify idiosyncratic variables, and decide whether an FA is safe.
Safety and Ethics Checklist
- Confirm that medical causes (pain, illness, sleep deprivation) have been ruled out or addressed.
- Obtain informed consent and, where required, approval from a review committee.
- Define criteria for terminating a session if injury risk becomes too high.
- Arrange appropriate padding, staffing, and protective equipment.
- Consider whether a less intrusive method, such as a precursor or latency-based FA, would be safer for dangerous behavior.
Step 2: Operationally Define the Target Behavior
Write a precise, observable, and measurable definition of each target behavior. Include examples and non-examples. Choose a measurement system, usually frequency or rate for discrete responses, or percentage of intervals for behavior that is hard to count. Train all observers and check interobserver agreement before starting.
Step 3: Set Up the Standard Conditions
The classic multielement design includes three test conditions and one control condition. Conditions are alternated rapidly, often in a randomized or semirandom order, with each session lasting about 10 minutes.
Attention Condition
The therapist diverts attention, for example by reading a book. Contingent on problem behavior, the therapist delivers brief attention such as a mild reprimand or expression of concern. This tests for social positive reinforcement in the form of attention.
Escape (Demand) Condition
The therapist presents instructions using a three-step prompting sequence. Contingent on problem behavior, the therapist removes the task for a short period, typically 20 to 30 seconds. This tests for social negative reinforcement.
Alone (Ignore) Condition
The client is alone in a room with no materials, or is observed with no social consequences programmed. Problem behavior produces no consequence. High, stable responding in this condition suggests automatic reinforcement. Safety considerations are especially important here, so some clinicians use a no-interaction condition with a therapist present.
Play (Control) Condition
The client has free access to preferred items, the therapist delivers frequent attention on a fixed-time schedule, and no demands are presented. Problem behavior is ignored. This condition functions as a control because the establishing operations for attention, escape, and sensory reinforcement are all abolished.
Optional Tangible Condition
If the interview suggests access to items maintains behavior, add a tangible condition. The client gets brief access to a preferred item, then it is removed. Contingent on problem behavior, the item is returned for a short time.
Step 4: Run the Sessions
- Establish a stable, brief pre-session period so the relevant motivating operation is in effect (for example, 5 minutes of attention deprivation before the attention condition).
- Alternate conditions in a predetermined order and use a distinct discriminative stimulus for each, such as a different colored shirt, room, or therapist.
- Implement each condition with high procedural integrity. Score the therapist behavior as well as the client behavior.
- Graph data after every session so you can make decisions in real time.
- Continue until the data show clear differentiation or until it is evident that no pattern is emerging.
Step 5: Interpret the Graph
Visual analysis of the multielement graph is the standard approach. Look for the following patterns:
- Attention function: Higher responding in the attention condition than in play.
- Escape function: Higher responding in the demand condition than in play.
- Tangible function: Higher responding in the tangible condition than in play.
- Automatic function: Responding is high and similar across conditions, including alone.
- Multiple functions: More than one test condition is elevated relative to control.
- Undifferentiated: Responding is low and variable across all conditions, or high across all including play.
When results are undifferentiated, review for procedural integrity errors, check whether an idiosyncratic variable was missed, extend the assessment, or modify conditions. A test-control comparison, sometimes called a pairwise design, can also help clarify unclear results.
Common Variations You Should Know
Brief Functional Analysis
Uses shorter sessions (about 5 minutes) and fewer repetitions. It is helpful in outpatient settings with limited time, but it may miss functions and produce false negatives.
Trial-Based Functional Analysis
Embeds short test and control trials into the natural routine. Each trial has a control segment followed by a test segment. It is useful in classrooms and when sessions would be impractical.
Latency-Based Functional Analysis
Terminates the session after the first instance of problem behavior and uses time to the first response as the dependent variable. This reduces the number of problem behavior occurrences, which is valuable for severe behavior.
Precursor and Synthesized Analyses
A precursor FA targets a response that reliably precedes a dangerous behavior. A synthesized contingency analysis combines multiple reinforcers (such as escape plus attention plus tangibles) in a single test condition. The tradeoff is that you may lose the ability to isolate which variable is responsible.
Using FA Results to Select Treatment
The function guides intervention. For attention-maintained behavior, consider functional communication training with attention as the reinforcer, noncontingent reinforcement on a dense schedule, and differential reinforcement of alternative behavior. For escape-maintained behavior, teach a break request, use demand fading, and keep extinction in place. For automatic reinforcement, provide competing stimulation and matched sensory items. A treatment should always match the function; an intervention that ignores the function can inadvertently strengthen the problem behavior.
How This Appears on the BCBA Exam
Expect scenario questions that ask you to identify which condition would be elevated given a described history, which assessment is least restrictive, or what to do when an FA is undifferentiated. You may be asked why the play condition is a control, what the programmed consequence is in each condition, or which FA variation fits a client who engages in dangerous behavior. Also expect ethics-linked items: you should select the least intrusive assessment that will yield useful information and take steps to protect the client during analysis.
A good study strategy is to draw the four conditions in a table with columns for antecedent, consequence, and function tested. Rehearse until you can fill it in from memory. Then practice interpreting sample graphs under timed conditions.
Key Takeaways
- An FA is an experimental analysis that identifies a functional relation.
- The standard conditions are attention, escape, alone, play, and optionally tangible.
- Safety planning and ethics come before implementation.
- Graph data each session and interpret using visual analysis.
- Variations such as latency-based and trial-based FAs reduce risk and time.
- Always match treatment to function.