The big idea
A functional analysis (FA) tests the function of a behavior directly. The analyst arranges conditions that each contain one possible maintaining contingency, then compares responding in each with a control condition. If behavior is reliably higher when, say, attention follows it, attention is shown to be functionally related to the behavior.
That is the step from prediction to control (A.1). It makes the FA the most precise part of a functional behavior assessment, and it brings the most responsibility: you are briefly arranging the conditions that evoke and reinforce problem behavior.
Breaking it down
The standard conditions
The model comes from Iwata et al. (1982/1994).
| Condition | Establishing operation | Consequence for the behavior | Tests for |
|---|---|---|---|
| Attention | Therapist’s attention is diverted, for example by reading | Brief attention, such as a statement of concern | Positive reinforcement by attention |
| Demand (escape) | Tasks are presented with prompts | Task removed for a short break | Negative reinforcement by escape |
| Tangible | Preferred item removed after brief access | Item returned briefly | Positive reinforcement by access to items |
| Alone or ignore | No interaction; few or no materials | None | Automatic reinforcement |
| Play (control) | Preferred items, frequent noncontingent attention, no demands | None; behavior is ignored | Comparison condition |
Ignore replaces alone when the behavior needs another person present, such as aggression. Tangible was added in later research.
Format
Conditions usually alternate rapidly in a multielement design (D.7), in sessions of 5 to 15 minutes, with distinct cues (such as different therapists or rooms) for each condition. A pairwise format, which alternates one test condition with play, can help when results are unclear.
Reading the graph
- Differentiated: behavior is consistently higher in one test condition than in play. That condition identifies the function.
- More than one elevated condition: the behavior is multiply controlled, for example by both attention and escape.
- High in alone or ignore, or high in every condition including play: suggests automatic reinforcement.
- Undifferentiated: may reflect automatic reinforcement, poor discrimination between conditions, or conditions that missed the relevant variable. Next steps include more sessions, a pairwise design, repeated alone sessions or redesigned conditions.
Variations
- Brief functional analysis (Northup et al., 1991): one or two short sessions per condition, often completed in about 90 minutes, sometimes followed by a contingency reversal to confirm. Useful in outpatient clinics, though fewer sessions mean less certainty.
- Trial-based functional analysis (Bloom et al., 2011): short trials embedded in classroom routines, each with a control and a test segment. It fits schools but offers less control than extended sessions.
- Latency-based functional analysis: each session ends at the first response, and the measure is time to that response. Shorter latencies in a test condition indicate function, and far fewer responses occur.
- Precursor functional analysis: consequences follow a behavior that reliably comes before the dangerous one, such as a whine before head hitting. It lowers risk, provided the precursor and the target share a function.
- Interview-informed synthesized contingency analysis (IISCA; Hanley et al., 2014): an open-ended interview and brief observation identify the situations that evoke the behavior. One test condition combines all the suspected reinforcers (for example, escape from work to toys and attention), delivered for the behavior or its precursors. It is compared with a matched control in which they are freely available.
Synthesized vs. isolated contingencies
Supporters of synthesized analyses cite speed, fewer occurrences of severe behavior, conditions tailored to the client’s life and good treatment outcomes. Critics argue that a combined condition can show a positive result even when some of its reinforcers would not maintain the behavior alone, so treatment may include unneeded components. Comparison studies have reported both agreement and disagreement with the standard FA, and the question is still open. For the exam, know the standard conditions and the rationale for the IISCA.
Safety and consent
Because an FA briefly reinforces problem behavior, plan for risk first (Standard 2.13, Selecting, Designing, and Implementing Assessments): obtain informed consent, rule out medical causes, set termination criteria (for example, stop if an injury of a defined severity occurs), arrange protective equipment or medical oversight for severe self-injury, and choose a format that limits exposure. Conduct an FA only within your competence or under supervision.
Seeing it in practice
- Jordan, 14, clinic. Hand biting is near zero in play, attention and tangible, moderate in alone and high in demand. The analyst concludes the behavior is maintained by escape and possibly by automatic reinforcement, and adds a series of alone sessions to check the second function.
- Talia, 7, public school. Separate 15-minute sessions are not possible at her school. A trial-based FA within class routines shows disruption only in the test segment of demand trials.
Common mix-ups
- FA vs. FBA. The FBA is the whole process; the FA is its experimental part.
- Alone tests automatic reinforcement, not attention. Behavior that continues with no one present does not depend on social consequences.
- Play is the control. It contains the reinforcers noncontingently and no establishing operations, so low responding there is expected.
- The demand consequence is escape, not a reprimand. In the demand condition, behavior produces a break from the task.
Exam tip: On an FA graph, compare each test condition with play. The condition that is consistently higher shows the function. High in alone, or high everywhere, points to automatic reinforcement.
Quick recap
- An FA manipulates contingencies to demonstrate function.
- Standard conditions: attention, demand, tangible, alone or ignore, and play as the control, usually in a multielement design.
- Differentiated results identify the function; high alone or high everywhere suggests automatic reinforcement.
- Brief, trial-based, latency-based, precursor and IISCA formats trade some precision for speed, safety or fit.
- Plan for safety: consent, medical screening, termination criteria and competent oversight.
Further reading
- Iwata, B. A., Dorsey, M. F., Slifer, K. J., Bauman, K. E., & Richman, G. S. (1994). Toward a functional analysis of self-injury. Journal of Applied Behavior Analysis, 27(2), 197-209. (Reprinted from Analysis and Intervention in Developmental Disabilities, 2, 3-20, 1982.)
- Hanley, G. P., Iwata, B. A., & McCord, B. E. (2003). Functional analysis of problem behavior: A review. Journal of Applied Behavior Analysis, 36(2), 147-185.
- Hanley, G. P., Jin, C. S., Vanselow, N. R., & Hanratty, L. A. (2014). Producing meaningful improvements in problem behavior of children with autism via synthesized analyses and treatments. Journal of Applied Behavior Analysis, 47(1), 16-36.
- Fisher, W. W., Greer, B. D., Romani, P. W., Zangrillo, A. N., & Owen, T. M. (2016). Comparisons of synthesized and individual reinforcement contingencies during functional analysis. Journal of Applied Behavior Analysis, 49(3), 596-616.
Key terms
- Functional analysis (FA, Analog functional analysis, Experimental functional analysis)
- An assessment in which the analyst systematically arranges antecedents and consequences, such as attention, escape from demands and access to items, and compares responding with a control condition to demonstrate which variables maintain a behavior.
- Functional behavior assessment (FBA, Functional assessment)
- The overall process of identifying the variables that reliably predict and maintain a behavior, using indirect assessment, descriptive assessment, functional analysis or a combination of them.
- Automatic reinforcement
- Reinforcement that occurs without another person delivering it: the behavior itself directly produces the consequence. It can be positive (adding stimulation) or negative (removing discomfort), and the reinforcer can be unconditioned or conditioned.
- Brief functional analysis (Brief FA)
- A shortened functional analysis with one or two brief sessions of each condition, often completed in about 90 minutes and sometimes followed by a contingency reversal to confirm the result (Northup et al., 1991).
- Trial-based functional analysis (Trial-based FA)
- A functional analysis made up of short trials embedded in a person's ongoing routines, each with a control segment and a test segment. Function is inferred by comparing the percentage of trials with behavior in each segment (Bloom et al., 2011).
- Latency-based functional analysis
- A functional analysis in which each session ends after the first occurrence of the behavior, and the measure is the latency to that response. Shorter latencies in a test condition suggest the function, and far fewer responses occur overall.
- Precursor functional analysis
- A functional analysis in which consequences are delivered for a precursor, a less dangerous behavior that reliably comes before the target behavior, so that function can be identified with fewer occurrences of the dangerous behavior. It assumes the precursor and the target share a function.
- Interview-informed synthesized contingency analysis (IISCA)
- A functional analysis format in which an open-ended interview and brief observation identify the conditions that evoke a behavior, and a single test condition combines all the suspected reinforcers, compared with a matched control condition in which those reinforcers are freely available (Hanley et al., 2014).
Check yourself
5 questions on F.6. Choose an answer to see whether you're right and why.
Show the answer
Answer: A. Hitting is consistently higher in one test condition than in play, so the results are differentiated and point to positive reinforcement by attention. An automatic function would show high rates in the ignore condition or across all conditions.
Show the answer
Answer: D. Behavior that persists when no social consequences follow it, including when the client is alone, suggests automatic reinforcement. Attention cannot explain high rates in the alone condition, where no one delivers it. A series of alone sessions can help confirm the pattern.
Show the answer
Answer: A. The play condition delivers attention and items noncontingently and places no demands, so no establishing operations are present and problem behavior should be low. The alone and ignore conditions test for automatic reinforcement; they are test conditions, not the control.
Show the answer
Answer: D. Ending each session at the first response limits exposure to as few occurrences as possible while still comparing conditions. A descriptive assessment adds no risk, but it cannot identify function. Extended sessions allow many more head bangs.
Show the answer
Answer: B. Critics argue that a single test condition combining several reinforcers can produce a positive result even when some of those reinforcers would not maintain the behavior alone. Both approaches use a control condition, and both use interviews or other information to design conditions.
Adds a step to your study record. Saved in this browser only.