Token Economy in ABA: How It Works and Where It Fails

A token economy pays learners with tokens they trade for rewards. Learn how BCBAs set one up, condition tokens, and avoid common mistakes.

Key takeaway

A token economy in ABA is a system where learners earn tokens for target behaviors and later trade those tokens for rewards. The tokens might be stars, tickets, points, punch cards, or app points.

ABA Beyond Autism
Watch the full CEU recording

ABA Beyond Autism

Nicole Parks · 1 CEU · 60 min
Watch on openceu.com →

Free learning path on this topic

PDA and Demand Avoidance

B. Kuereine Gray's three PDA talks plus Clelia Sigaud on the logic of anxiety. What PDA is and isn't, how to collaborate with caregivers, how to handle complex profiles and replacement behaviors, and why demand-avoidant learners respond the way they do.

Start the path →4 courses · 5 CEUs · Watch at your own pace

Experts who cover this

10 recordings on openceu cover token economy.

A token economy in ABA is a system where learners earn tokens for target behaviors and later trade those tokens for rewards. The tokens might be stars, tickets, points, punch cards, or app points. The rewards they buy are called backup reinforcers. That just means the real thing the learner wants, like a song, a snack, or a special event.

Token economies show up in clinics, classrooms, homes, and hospitals. They look simple on paper. In practice, the experts below agree that many systems are set up poorly. They also differ on how much clinicians should lean on tokens at all. This page pulls together what ten presenters said, plus what the research adds.

How a token economy works in real settings#

Every token economy has the same basic parts. There is a target behavior. There is a token the learner earns for that behavior. And there is a trade, where tokens turn into something the learner values. Claudia Segoe, speaking about Tier 1 school support, the universal practices for every student, notes that these systems go by many names. She lists token systems, level systems, and school stores as members of one family. She calls them "structured systems of acknowledgement."

Dr. Kaci Ellis gives a tour of what this looks like in schools. Some schools use a schoolwide ticket that students trade in for events. Some teachers use Class Dojo, which is an app that tracks points. Others use a punch card. When the card is full, the student picks from a treasure box or earns lunch with the teacher.

Ellis also shares two of her favorite rewards. Students can earn a coupon to be "Class DJ" and pick a song for a brain break. Or they can earn the right to draw a picture on her whiteboard that stays up. These rewards cost almost nothing. They work because students actually want them.

Having a token board is not the same as running one well#

Dr. Kaci Ellis makes a point that every school-based BCBA, a board-certified behavior analyst, will recognize. Walking into a classroom and seeing a token chart on the wall tells you very little.

Just because a teacher has a token economy doesn't mean it's going correctly.From the talk — Dr. Kaci Ellis

Her fix is a step-by-step process she walks through with each teacher. She starts by explaining what a token economy is, in plain words. She does not assume the teacher already knows. Many teachers have heard the term but never learned why the system works.

Ellis also tackles a common teacher objection head on. Some teachers feel they are paying kids to do what they should already do. Ellis frames the system as having a clear point beyond that. She asks teachers to try it with her and build it together. She also makes sure there is a plan to monitor it over time. In her view, a token economy without monitoring is just a chart on a wall. The teacher and the BCBA both need to check whether behavior is actually changing.

Tokens only work inside a good classroom#

Claudia Segoe adds an important limit. A token system cannot carry a classroom by itself. She places token economies alongside other Tier 1 practices, like a high rate of praise. She cites a target of about six praise statements per 15 minutes of group instruction. For kids at higher risk of challenging behavior, she suggests a praise ratio well above four to one.

Segoe also stresses that the backup reinforcers must be a "meaningful array." That means a real menu of choices that fits the learner's age, stage, and interests. Teachers can get stuck here. But a weak menu makes the whole system weak.

If everything is terrible, except that I get tickets, which I occasionally can redeem for a prize, I'm not going to have a meaningful school experience, right?From the talk — Claudia Segoe

Her point is that tickets might help a little. But they cannot fix a classroom that lacks warm relationships and skill building. Segoe wants token systems used inside a bigger plan, not in place of one. For a BCBA consulting with a school, this is a useful lens. Before tweaking the token board, look at the rest of the room.

A big store does not guarantee good use#

Mark Malady describes a program that spent heavily on its token economy. The rewards came from donations, and the store was a big draw for learners.

We had this crazy token economy that we spent about $20,000 per year that we got from donationsFrom the talk — Mark Malady

Malady calls it a "highly functional token economy for learners." Yet he is candid about what went wrong. The team did not use the store for the right things. Over time, they slipped into bad patterns with it. A rich reward menu did not stop drift in how staff ran the system.

He also names a plain, practical drawback. Physical tokens get lost. He later built a card-based feedback system for staff and saw the same issue. Cards went missing, and they had to be printed. He tried a point system, ran into other problems, and went back to cards. The lesson for clinicians is simple. Plan for lost tokens and for staff drift before they happen.

The token has to mean something first#

Michael Carrero and Sean Yocum, BCBA hosts of the Rad N Bad podcast, push on a basic question. Is the token actually working as a reinforcer? A token is supposed to become a generalized conditioned reinforcer. In plain terms, that means it gains value because it has been paired with many good things. Yocum argues many boards never get there.

those token boards aren't even acting as generalized condition reinforcers, like I said.From the talk — Sean Yocum

Yocum says that, half the time, the board is just a visual countdown timer. It tells the child how long until they can escape the clinician. Carrero agrees and goes a step further.

I would argue that the vast majority of people who have token economies in place have not done a reinforcer assessmentFrom the talk — Michael Carrero

A reinforcer assessment tests whether an item actually increases behavior. It goes beyond a preference assessment, which only shows what a learner picks. Carrero says the assessment should set how much the reward is worth and how many responses it should take to earn. Even when teams do this step, he says staff are often not trained to the standard needed. He wants staff to know how and when to bring a token economy in and when to take it out.

The five-token board problem#

In a second Rad N Bad episode on supervision, Michael Carrero names a habit he sees in new BCBAs. They carry the same board everywhere. It has five tokens, a sticky note "I am working for" spot, and a generic choice board. Every learner gets the same prize box.

Carrero explains where the number five often comes from. Five tokens make percent math easy. Four out of five is 80%, which fits the common 80% mastery rule. That is a data habit, not a clinical reason. He calls supervisees who copy this setup "procedural parrots." They can repeat a procedure but cannot problem solve or adapt it.

It's not I'm working for my preferred items that could potentially be reinforcing. It's I'm working because I'm working for tokens, and I'm really working to get away from you, right?From the talk — Michael Carrero

In the NET episode, Carrero discusses natural environment teaching, or NET, teaching built into everyday moments rather than drills. He adds that many boards are really fixed ratio 1 at heart. That means the schedule always pays for one response and never asks for more. Five tokens buys the iPad, and every break is the same length. He asks why every learner is on the same numbers. His answer is that individualizing takes real assessment, and many teams skip it.

Starting dense, then thinning the schedule#

Morgan Johnson, discussing a study on physical activity, uses token economies as a familiar example. Her point is about how a new system usually starts.

right when you introduce a token economy. You're doing it on a continuous reinforcement schedule.From the talk — Morgan Johnson

Continuous reinforcement means the learner gets a token for every correct response. From there, the team asks for more responses per token. It might move to every three, then every five. The group also mentioned moving to a variable schedule, where the number changes. This process is called reinforcement thinning. In plain terms, that means asking for more work before the reward each time. It has its own page.

Michael Carrero adds the warning sign to watch for while thinning.

Ratio strain. Like figure out where that ratio strain is and thin the schedule.From the talk — Michael Carrero

Ratio strain is when the learner's behavior breaks down because the work asked is too much for the pay. Carrero also points to shaping, building a skill through small rewarded steps, and "magnitude or value manipulation." That means a bigger or better reward for a more complete response. He worries that learners who leave a clinic on dense schedules will struggle in a general education classroom. That setting, he says, has a much thinner schedule of reinforcement.

A hospital-wide token economy: the Nicole Parks case#

Nicole Parks shares the most detailed case in this set. She is a BCBA who personally built and ran a token economy at a Georgia state psychiatric hospital.

this token economy was introduced following legal changes, including a settlement in 2009From the talk — Nicole Parks

Parks says she was the only behavior analyst at her hospital at the time. Each patient had a treatment team, including a nurse, a social worker, and an activity therapist. Many patients were not motivated to take part in therapy. The token economy aimed at that gap. Parks was promoted and asked to roll it out across the whole hospital.

Patients earned points for things like making the bed, taking medication, showering, and going to groups. Rewards included items such as t-shirts or pajama pants, after a review for safety. Later, the hospital added a themed monthly event off the unit, with themed food. Patients could save points to buy a ticket.

The outcomes were strong. Parks reports more group participation and fewer restrictive interventions. Restrictive interventions here means seclusion, restraint, and emergency shots. Staff morale also rose, because patient aggression dropped. Staff who resisted the program at first came to buy in.

Giving the learner a say in how points are earned#

One design choice in Parks's program deserves its own look. Patients did not just receive a plan. They helped write it.

we had cards, we had a point system and each of the, um, patients was able to allot 50% of their points and their treatment team was allowed to allot the other 50% of their points across some different activities throughout the day.From the talk — Nicole Parks

Parks is frank about how this played out. Patients put their points where they knew they would succeed. The team put points where they felt the patient most needed to grow. That split gave patients early wins while still pushing on harder goals.

Her favorite outcome involved one man who had been on the unit for 15 years. He had refused to leave or take part in anything.

So for a long time, he didn't really care. He had a card. He never got it signed off on. And then eventually after he kept hearing people come back to the unit, talking about all the things that they were getting and going to the events and stuff like that, he started participating.From the talk — Nicole Parks

He even left the unit to attend a social event. Staff were shocked. Parks says nothing else in over a decade had worked for him. Notice the path here. He did not respond to the card alone. He responded after hearing peers describe the rewards they earned.

Building a family-wide system from the learner's values#

Brian Middleton describes a home token economy built around a 12 to 13 year old named Paul. Teachers and parents had called Paul antisocial. Middleton disagreed. In their talk, Paul lit up about things he valued: challenge, achievement, and progression. Middleton then turned to Paul's mom with a suggestion.

sounds like Paul needs a token economy at home. And everybody in the home needs to be a participant in this token economy. It's not just him. It's everybody.From the talk — Brian Middleton

Mom and Paul designed the system together, with Middleton guiding. They saw a change in behavior early on. The family later transferred to a closer provider due to distance and other factors.

Middleton also notes that this system still had limits. He used limited holds on some items Paul could access. A limited hold means the reward is only available for a set window of time. Middleton frames this as part of shaping, not as restriction for its own sake. He says limited holds "have a time and a place," because a learner cannot have access to everything all the time.

Teaching staff to use tokens beyond one setting#

Mellanie Page looks at token economies from a trainer's view. She uses them to show RBTs, the front-line staff who carry out behavior plans, how to stretch one skill into many settings. Her approach draws on derived relational responding. In simple terms, that means helping trainees connect what they learned to new cases they were never directly taught.

Page might first train RBTs on token economies in a classroom. Then, through guided talk, she asks them to adapt it. Her challenge question is practical.

How would I use it in a grocery store if I don't want to lug around the token economy?From the talk — Mellanie Page

The goal, she says, is that staff do not think a system only works in one place. She wants them to understand they can use it anywhere. This lines up with Carrero's concern about "procedural parrots." Both presenters want staff who can reason about a token system, not just copy one. Page's method gives supervisors a way to build that skill on purpose.

The ADHD caution: when tokens build dependence#

Nicole Stewart, a BCBA who works with ADHD learners, raises the sharpest warning in this set. ADHD stands for attention-deficit/hyperactivity disorder. In her Best Practices talk, she asks her audience a direct question.

why should we be cautious using a token economy system with people with ADHD?From the talk — Nicole Stewart

Her own answer is clear. She knows token systems are often recommended for ADHD. She still uses them sparingly.

I don't like to use a lot of token economy systems, despite the fact they're recommended. Because they create a dependency for the ADHD brain on those reward systems.From the talk — Nicole Stewart

Stewart describes what this looks like at home. Parents tell her their child won't do anything unless there is something in it for them. A child may refuse to put away dishes because it is "not on my chore chart." Another may finish ten required tasks and then stop cold. Stewart says this builds rigid thinking instead of flexible thinking. Flexible thinking means adjusting when plans or demands change.

Her view is not all or nothing, though. She says there are fitting uses for reward systems. On vacation, routines break down, and that is hard for many kids with ADHD. So she has used a tally counter with her own kids. Each good choice earns a point, and points add up to screen time. She also says a chore or sticker chart can help in another way. A child can learn that a task, like brushing teeth, is easier than it seems. Once the habit forms, the chart has done its job.

In a separate talk on autonomy, Stewart adds a point about fading. Fading means slowly removing a support until the learner no longer needs it. She warns that leaning on token economies and other reward systems mainly to get compliance creates a problem. Those systems are hard to generalize or fade across different situations.

She also flags a habit she sees often. When a child does not comply, the first move is to add a reward. Stewart calls this a way to coerce the child, not to build independence. In a third talk on classroom management, she notes that a token system will likely reduce the problem behavior. She says these tools "aren't terrible." But a token system may not teach the skill the child is missing. So her message to clinicians is to use tokens on purpose and plan the exit early.

Where the experts differ: helpful structure or a crutch?#

The presenters agree that poor setup is common. They differ on how central token economies should be in good practice.

Nicole Parks and Claudia Segoe present token systems as valuable tools. Parks credits one with reaching a patient that nothing else had reached in years. Segoe treats structured acknowledgement systems as a core Tier 1 practice, as long as they sit inside a warm classroom. Brian Middleton built one around a teen's own values, with strong early results.

Nicole Stewart sits on the other side of the line. She says token systems are recommended for ADHD, yet she avoids using many of them. Her worry is dependence on the reward and rigid, transactional thinking. That is a direct tension with the more positive picture from Parks and Segoe.

Michael Carrero, speaking about natural environment teaching (NET), makes a related point. He prefers reinforcers tied to the behavior itself. If a child asks for a push on the swing, the push is the reward. He argues this avoids a "bizarre reinforcer dependency" on a laminated board. So the split is real. Some experts see tokens as a strong bridge to engagement. Others see them as a crutch that clinicians reach for too fast.

The Rad N Bad hosts expand on this in Rad N Bad: The Middle Ground of NET.

What the research says#

The published research supports token economies broadly, while echoing the presenters' worries about setup. Token economies are described as among the oldest and most successful programs in ABA (degli Espinosa & Hackenberg, 2024). The same authors name a gap between research and daily practice. They offer step-by-step guidance for building a system from scratch.

An app-based token economy raised task engagement for every adult with developmental disabilities in one study (Beahm et al., 2023). Gains held up to 6 weeks after the intervention ended. Some staff, though, found it time-consuming and hard to run. That matches what Ellis and Malady say about implementation.

Research on ADHD adds a useful counterpoint to Stewart's caution. A game-based digital program used a self-directed token economy for children with ADHD (Kim, 2025). After four weeks, the token group scored higher on an attention task than the control group. So tokens can help with ADHD. The open question is how to fade them well.

Critics have long called token economies too rigid and restrictive (Pritchard et al., 2018). These authors studied a school-wide point and level system that used no punishment. Higher point use was linked with fewer severe problem behaviors in a residential program.

References

Beahm, L. A., Ingvarsson, E. T., Funk, N., Haskins, L., & Frazier, J. (2023). Using an App-Based Token Economy to Increase Engagement in Daily Living and Vocational Tasks with Adults with Developmental Disabilities. Behavior Analysis in Practice, 16(4), 1047-1060. https://doi.org/10.1007/s40617-023-00774-4

degli Espinosa, F. & Hackenberg, T. D. (2024). Token economies: Evidence‐based recommendations for practitioners. Behavioral Interventions, 39(4). https://doi.org/10.1002/bin.2051

Kim, S. (2025). Verification of the Effectiveness of a Token Economy Method Through Digital Intervention Content for Children with Attention-Deficit/Hyperactivity Disorder. Bioengineering, 12(10). https://doi.org/10.3390/bioengineering12101035

Pritchard, D., Penney, H., & Mace, F. C. (2018). The ACHIEVE! program: A point and level system for reducing severe problem behavior. Behavioral Interventions, 33(1), 41-55. https://doi.org/10.1002/bin.1506

FAQ#

What is a token economy in ABA? It is a system where a learner earns tokens for target behaviors. The learner later trades the tokens for backup reinforcers, which are items or activities they value. Tokens can be stars, tickets, punch cards, or app points.

Why do token economies sometimes fail? Presenters point to a few common causes. Teams skip the reinforcer assessment, so the token never gains real value. Every learner gets the same five-token board and prize box. And staff drift over time without monitoring. A token economy also cannot make up for a classroom that lacks warmth and good teaching.

Should a token economy be faded over time? Most systems start with a token for every response, then ask for more responses per token. Experts like Nicole Stewart warn that systems used mainly for compliance can be hard to fade. Planning the fade from day one, and watching for ratio strain, helps avoid dependence on tokens.

Turn this topic into a CEU

You just studied this. Now get credit for it.

Watch ABA Beyond Autism with Nicole Parks and earn 1 free BCBA CEU. Detailed certificate, delivered the moment you finish.

Watch and earn the CEU →Free account · No card · detailed cert

Keep going with a free learning path

PDA and Demand Avoidance

B. Kuereine Gray's three PDA talks plus Clelia Sigaud on the logic of anxiety. What PDA is and isn't, how to collaborate with caregivers, how to handle complex profiles and replacement behaviors, and why demand-avoidant learners respond the way they do.

Start the path →4 courses · 5 CEUs · Watch at your own pace

Want the primary literature? Read the Token Economy in ABA research roundup on our sister site, Behaviorist Book Club.