Translated from the Portuguese original. Read the original (Português)
A functional analysis is the experimental method that uncovers the function of a behavior — that is, what it "gets" as a consequence. Instead of just observing, the professional deliberately manipulates the conditions around the person (giving attention, presenting tasks, leaving the person alone) and compares in which one the behavior shows up more. The condition that makes the behavior spike points to what maintains it. Knowing the function makes it possible to teach a better path to the same goal, instead of simply trying to suppress the behavior.
In Applied Behavior Analysis (Cooper, Heron & Heward; Iwata et al.), a functional analysis is an experimental procedure in which environmental variables — antecedents and consequences — are systematically manipulated to identify the variable that maintains a behavior. By arranging and comparing test conditions, a functional (causal) relationship between the behavior and its consequence is demonstrated, not merely a correlation. It is the experimental step — and the gold standard — within the functional behavior assessment.
Why find out the function?
Two behaviors that look identical on the outside can exist for completely different reasons. One child may bang on the table to get an adult's attention; another may do the exact same thing to escape a difficult task; a third one, because the sensation itself feels good. If the reason is different, what helps is also different. Treating a behavior without knowing its function is like giving medicine before a diagnosis: it may not work — or even make things worse. The functional analysis exists precisely to answer the question "what is this behavior for?" in a way that can be tested.
The four classic conditions
The best-known methodology arranges a few short, repeated test conditions, each one checking a hypothesis about function. Tap to explore:
Attention — tests social positive reinforcement
The adult stays nearby, but "busy," and only gives attention (a glance, a comment, a "don't do that") when the target behavior occurs. If the behavior increases in this condition, the hypothesis is that it is maintained by attention. A similar variation tests access to a preferred item or activity (the tangible condition).
Escape — tests social negative reinforcement
Tasks or requests are presented, and every time the target behavior occurs, the demand is withdrawn for a moment — the person "escapes" what was being asked of them. If the behavior rises here, the likely function is escaping or getting out of demands.
Alone — tests automatic reinforcement
The person stays alone, with no materials and no attention available. If the behavior remains frequent even with no one nearby, the consequence that maintains it is produced by the person's own body (for example, a sensation) — what's called automatic reinforcement, which doesn't depend on another person.
Play — the control condition
A rich, pleasant environment: free attention, available items, and no demands. Nothing special happens when the behavior occurs. It serves as a comparison (a kind of baseline): the behavior is expected to stay low, and it's against this baseline that the rise seen in the other conditions is read.
Observing is not the same as testingThe ABC record (see three-term contingency) describes what happens naturally and reveals correlations. The functional analysis goes further: it deliberately arranges the conditions and, because of that, demonstrates a causal relationship. It is this controlled manipulation that makes it the most reliable method for identifying function.
The three possible functions
However varied behaviors may be, the consequences that maintain them tend to fall into three broad categories. Flip the cards:
One of the field's most studied methodsThe modern experimental methodology was born with the work of Iwata and colleagues (1982/1994) on self-injury, and has since been replicated and refined in hundreds of studies. Finding the function makes it possible to build a function-based treatment — for example, teaching the person to ask for attention or a break in an effective way — a more effective and less restrictive approach than simply trying to suppress the behavior.
Done with care
Because the functional analysis needs to evoke the behavior for a few moments in order to measure it, it is conducted by trained professionals with safeguards in place: short sessions, clear stop criteria, a safe environment, and consent. There are versions designed to reduce risk and time — such as the brief functional analysis, the trial-based version (usable in the classroom or at home), and interview-based formats. Not every case requires the full version.
Understanding is not "excusing"Finding out what a behavior is for does not mean approving of it or justifying it. The function is working information: it points to the better path to teach. If a child hits to escape a task, the goal becomes teaching them to ask for a break and making the task more manageable — not simply "putting up with" the hitting.
Test your understanding
Quick quiz
Three questions to reinforce the concept. You'll see the explanation right away.
Further reading
- Iwata, B. A.; Dorsey, M. F.; Slifer, K. J.; Bauman, K. E.; Richman, G. S. "Toward a functional analysis of self-injury". Analysis and Intervention in Developmental Disabilities, 1982 (reprinted in Journal of Applied Behavior Analysis, 1994).
- Cooper, J. O.; Heron, T. E.; Heward, W. L. Applied Behavior Analysis. 3rd ed., 2020 — chapter on functional behavior assessment.
- Beavers, G. A.; Iwata, B. A.; Lerman, D. C. "Thirty years of research on the functional analysis of problem behavior". Journal of Applied Behavior Analysis, 2013.
- Hanley, G. P.; Iwata, B. A.; McCord, B. E. "Functional analysis of problem behavior: A review". Journal of Applied Behavior Analysis, 2003.
This entry is educational and does not replace assessment, diagnosis or treatment by a qualified health professional. Every person is unique — for a specific case, seek individual guidance.