Functional Communication Training (FCT) in ABA: Steps and Examples
How functional communication training works in ABA: examples by function, choosing the response, why extinction matters, and how to thin reinforcement.
Key Takeaways
- 01FCT teaches a communication response that earns the same reinforcer the problem behavior was getting. It is a form of DRA, and the response has to serve the same function as the behavior it replaces.
- 02Identify the function first, then pick the lowest-effort response the person can do, such as a card touch. Reinforce every response at first.
- 03FCT doesn't work reliably without extinction. In one inpatient study, FCT without extinction was tried 11 times and didn't bring problem behavior below 90% of baseline in any of them.
- 04Thin reinforcement gradually, using time delay, work requirements or signaled availability. Expect problem behavior to come back at some steps, then record it and tell your supervisor instead of changing the plan on the spot.
- 05On the RBT exam, FCT is named in task D.3 of the Behavior Reduction domain, which carries 14 of the 75 scored questions.
Functional communication training in four moves
- Find the reinforcer. Work out what the problem behavior gets the person.
- Teach an easier way to ask for it. The request has to take less effort than the problem behavior.
- Make the request pay, every time. Meanwhile, the problem behavior stops paying wherever that's safe.
- Thin it out. Slowly make the reinforcer less available until the plan fits real life.
A child who screams to get out of math learns to hand over a break card. It works. The screaming drops, everyone's thrilled, and by Thursday the child is handing over the card every time a worksheet comes out. No math is getting done.
That's not a failure of functional communication training. It's the expected middle of it. Most explanations of functional communication training stop at "teach the child to ask," which is the easy part. This guide is for the people running it: RBTs, students in an ABA program, special education staff and future BCBAs. It covers what FCT is, how to choose the response, why extinction changes the results, how to thin reinforcement, and where FCT in ABA shows up on the RBT and BCBA exams.
What Is Functional Communication Training?
Functional communication training is a differential reinforcement procedure. You teach a person a communication response that produces the same reinforcer their problem behavior has been producing. If hitting gets attention, you teach a way to ask for attention. If screaming ends a task, you teach a way to ask for a break.
The word "functional" is doing the work. Functional communication training isn't general language therapy, and it isn't teaching someone to say "please." The response is chosen because it serves the same function as the behavior it replaces. Teach a child to ask for a snack when their aggression is about escaping a task, and the aggression has no reason to stop.
In the language of differential reinforcement, functional communication training is a form of DRA, differential reinforcement of alternative behavior. Every FCT plan is a DRA plan, but not every DRA plan is FCT. Reinforcing a child for sitting with their hands in their lap can be DRA. It isn't FCT, because nothing was communicated.
Where Did FCT in ABA Come From, and Does It Work?
Edward Carr and V. Mark Durand described the approach in a 1985 study of four children with developmental disabilities. The children learned to ask adults for attention, help or both, and the authors reported "replicable suppression of behavior problems" (Carr and Durand, 1985).
The research has grown a long way since. A 2008 review found 91 articles across 19 journals, covering 204 participants who ranged from young children to adults. Of those participants, 195 had a developmental disability or intellectual disability and 81 were diagnosed with autism (Tiger, Hanley and Bruzek, 2008). A 2020 systematic review by the National Clearinghouse on Autism Evidence and Practice counted functional communication training as an evidence-based practice for autistic learners from preschool through high school, based on 31 single-case design studies (AFIRM Team, 2025).
Read the phrase "single-case" carefully. This evidence comes from many small, tightly controlled studies, often with experts running the treatment. That's strong evidence that functional communication training can work. It's weaker evidence for how well it holds up in a busy classroom, which is why the later sections of this guide matter.
Functional Communication Training Examples by Function
The form of the response changes with the person. What it earns depends on the function. These examples are illustrations, not protocols.
Escape
- Problem behavior
- Throws materials when a worksheet appears
- Response to teach
- Hands over a "break" card, or asks for "help"
- What it earns
- A short break from the task, or help with it
Attention
- Problem behavior
- Hits a staff member who is talking with someone else
- Response to teach
- Taps the adult's arm or says "look at me"
- What it earns
- A brief, warm interaction
Materials
- Problem behavior
- Screams when the tablet is put away
- Response to teach
- Signs or selects "tablet" on a device
- What it earns
- Time with the tablet
In the 2008 review, problem behavior was maintained by escape from demands in 43% of cases, attention in 32%, access to materials in 29%, escape from other aversive events in 4%, and access to other events in 3% (Tiger, Hanley and Bruzek, 2008). Those figures add up to more than 100% because some behavior served more than one function, so a plan can need more than one response.
All of those functions are social: another person controls the reinforcer. That's why functional communication training fits them. When a functional behavior assessment points to automatic reinforcement, there's no one to ask, and the team usually needs a different approach.
How Do You Choose the Communication Response?
This decision gets less attention than it deserves, because the response can fail even when the function is right. Researchers have used vocal responses, picture exchanges, sign language, gestures, and voice or text output devices (Tiger, Hanley and Bruzek, 2008). Three questions narrow the list.
- Is it easier than the problem behavior? Horner and Day (1991) compared signing a full sentence with signing a single word. The full sentence didn't beat the problem behavior. The single word occurred to the exclusion of it. Start with the lowest-effort response the person can do.
- Can you prompt it? Vocal responses are harder to prompt than a card touch or a hand movement. So the 2008 review suggests a motor response as the first FCT response, even for people who can speak, with vocal responses as the longer-term ideal because they reach people at a distance.
- Will other people understand it? A response only works if the community responds to it. Durand (1999) taught 5 people with disabilities to use voice-output devices, then took the devices into community settings, where people who hadn't been trained responded to them.
Picture systems have a tradeoff of their own. They're easy to prompt, but the person has to scan the choices and tell them apart, which is hard for some learners (Tiger, Hanley and Bruzek, 2008). One card that means one thing is a reasonable place to begin. Add choices once that card is reliable.
How to Do Functional Communication Training, Step by Step
- Identify the function. Of the 91 articles in the 2008 review, 71 used a functional analysis, and the authors recommend one before functional communication training for severe problem behavior. A wrong function means you teach a request for something the person wasn't after.
- Pick the response. Use the three questions above. Write down exactly what counts, so every adult reinforces the same thing.
- Teach it when the motivation is there. Set up the situation that usually triggers the behavior, then prompt the response before the problem behavior happens. Fade the prompts as the person starts responding independently.
- Reinforce every response at first. Every FCT study in the 2008 review started on a continuous schedule, where each response earned the reinforcer. Don't skip this because a break every time feels excessive. That's the job at this stage.
- Stop reinforcing the problem behavior, if that's safe. The next section explains why this step changes the outcome.
- Bring in the people and places of daily life. Add trainers and settings, and train caregivers. Teaching in one room with one adult doesn't reliably carry over.
- Thin the schedule. Make the reinforcer gradually less available, using one of the methods below.
- Graph it throughout. Track problem behavior and independent requests together. Rising requests and falling problem behavior is what you want to see. Either one alone doesn't tell you enough.
Does FCT Work Without Extinction?
Not reliably. Extinction means the problem behavior no longer produces its reinforcer. If screaming still ends the math, a child has two ways to get out of math, and asking isn't always the one they'll pick.
The clearest numbers come from an inpatient unit treating severe behavior, in a study of 21 clients. Across 27 applications, functional communication training without extinction was tried 11 times and didn't bring problem behavior below 90% of baseline in any of them. FCT with extinction was used 25 times and produced reductions of at least 90% in 11 (Tiger, Hanley and Bruzek, 2008, summarizing Hagopian and colleagues, 1998). So extinction made a large difference, and it still left more than half of those applications short of a 90% reduction.
The same study found that adding punishment produced at least a 90% reduction in every case where it was used. It also found that adding demand or delay fading to FCT with extinction reduced how well the treatment worked in about half of the applications (Hagopian and colleagues, 1998). These were severe cases on a specialized unit, and the 2008 review notes that therapeutic punishment raises ethical issues. Don't read this as a general recommendation. Whether to use punishment is a clinical and ethical decision for the supervising BCBA, within the rules of the setting.
Three practical points follow for anyone running the plan:
- Extinction has to match the function. Ignoring works for attention. For escape, extinction means the demand stays in place.
- Extinction isn't always possible. You can't ignore self-injury or furniture climbing when ignoring would be dangerous (Brown and colleagues, 2020). The behavior plan should say what to do in that case.
- Expect things to get worse before they get better. An extinction burst is one of the secondary effects the RBT outline lists under task D.6. If you give in during a burst, you've reinforced a more intense version of the behavior.
How Do You Thin Reinforcement in FCT Without Losing Progress?
This is the part that decides whether functional communication training survives outside the training room. No teacher can give a break on demand all day. But if a child who always got a break suddenly stops getting one, the request itself goes on extinction, and that sets the stage for the problem behavior to come back (Tiger, Hanley and Bruzek, 2008). So the change has to be gradual.
Time delay
- How it works
- A wait between the request and the reinforcer, starting small. One study went from 1 second up to 30.
- What to watch
- It's the most commonly reported method, but long waits weaken the link between asking and getting.
Work requirement
- How it works
- The person finishes more of the task before a request is honored. One study used a 16-step task.
- What to watch
- The request and the reinforcer stay close together, because the reinforcer comes right after the request that's honored.
Signaled availability
- How it works
- A visible cue shows when requests will pay. One study used a red card for reinforcement periods and a white card for extinction periods.
- What to watch
- It worked with reinforcement available only 20% of the time, and it gives caregivers real gaps.
The three methods above come from Fisher and colleagues (2000), Lalli, Casey and Kates (1995), and Hanley, Iwata and Thompson (2001), as summarized in the 2008 review. They aren't mutually exclusive. Time delay is the simplest to explain, and signaled availability lets the card do the telling when an adult is busy. Signaled availability is a multiple schedule; the other compound schedules are in our guide to schedules of reinforcement.
A Worked Thinning Example
Devon is a fictional 7-year-old who screams to get out of math. The steps below are illustrative. A real plan moves on the client's data, not on a calendar.
- Training. Devon hands over the break card and gets a short break right away, every time. Screaming doesn't end the task.
- A small requirement. After Devon asks, the teacher says "One problem, then break." Devon finishes one problem and gets the break.
- A bigger requirement. The number of problems climbs only while screaming stays low and Devon keeps asking.
- A visible cue. A card on the desk shows when breaks are available, so "not yet" is clear without an argument.
- Something to do while waiting. If screaming returns during waits, the 2008 review suggests preferred items during delay periods to compete with it.
Why Relapse Is the Normal Case
Plan for problem behavior to come back. A study of 25 consecutive functional communication training treatments for severe destructive behavior found resurgence in 76% of treatments, meaning the behavior returned when reinforcement for the replacement got worse. It found renewal in 69%, meaning the behavior returned after a change of context, such as a new setting (Mitteer and colleagues, 2022). Each of those happened in about a third of the individual changes, not at every step.
The 2008 review says the same thing plainly: problem behavior "should be expected to occur during the reinforcement thinning process, regardless of the strategy used." So a spike at a new step isn't proof the plan has failed, and an RBT who sees one should record it and tell the supervisor instead of changing the plan on the spot. The BCBA decides whether to wait, step back or add a component.
Where FCT Goes Wrong in Real Sessions
- The request only works with one person. If the clinic honors the card and the classroom doesn't, the classroom is teaching the card doesn't work.
- Adults require "good" asking too soon. Asking for a full sentence when the plan calls for a card touch raises the effort, and the problem behavior becomes the easier option again.
- The card goes missing. A response the person can't reach isn't a response. Put the device or card wherever the behavior tends to happen.
- Problem behavior and asking get chained. If a child screams and then asks, and gets the break, the team may be reinforcing the scream too. The plan should say what happens in that sequence.
- Thinning starts on someone's schedule, not the data. Pressure to get back to full academics is real. Moving before the data supports it invites the relapse you were trying to avoid.
These details belong in the behavior intervention plan, so that every adult responds the same way at the same step.
FCT on the RBT and BCBA Exams
On the RBT exam, functional communication training is named in task D.3, "Implement differential reinforcement procedures (e.g., DRO, DRA, DRI, DRL, FCT)." It sits in the Behavior Reduction domain, which carries 14 of the 75 scored questions. Task D.6 covers the secondary effects you'll see during functional communication training, including extinction bursts and resurgence. Expect a short scenario and four procedure names. The RBT practice exam has an FCT question of exactly that kind.
On the BCBA exam, the matching task is G.2: "Design and evaluate differential reinforcement (e.g., DRA, DRO, DRL, DRH) procedures with and without extinction." Behavior-Change Procedures carries 25 scored questions. Two tasks in Domain H, which carries 20, apply as well: H.3, selecting a socially valid alternative behavior, and H.5, planning for relapse. Functional communication training isn't spelled out in the BCBA outline, but it's a natural test case for all three. Our BCBA task list guide has the full list.
For the job itself, the split is simple. The BCBA chooses the response, writes the thinning steps and decides when to move. An RBT usually delivers them session after session, which means the RBT is often the first to see the plan working or slipping.
Frequently Asked Questions
What is functional communication training in ABA? +
What is an example of functional communication training? +
Who developed functional communication training? +
Is FCT the same as DRA? +
Do RBTs implement functional communication training? +
Sources
- Tiger, Hanley and Bruzek, Functional communication training: a review and practical guide (Behavior Analysis in Practice, 2008)
- Carr and Durand, Reducing behavior problems through functional communication training (JABA, 1985)
- Hagopian, Fisher, Sullivan, Acquisto and LeBlanc, Effectiveness of functional communication training with and without extinction and punishment: a summary of 21 inpatient cases (JABA, 1998)
- Mitteer, Greer, Randall and Haney, On the scope and characteristics of relapse when treating severe destructive behavior (JABA, 2022)
- Brown and colleagues, Resurgence following differential reinforcement of alternative behavior implemented with and without extinction (2020)
- AFIRM Team, Functional Communication Training brief packet (updated 2025), citing the 2020 NCAEP evidence-based practice report
- BACB, RBT Test Content Outline (3rd ed.) - tasks D.3 and D.6, Behavior Reduction domain weight
- BACB, BCBA Test Content Outline (6th ed.) - tasks G.2, H.3 and H.5
Keep going
Explore ABA programs
Ready to take the next step? Browse our expert-ranked program listings.