Reinforcement in ABA (applied behavior analysis) is any consequence that follows a behavior and makes that behavior more likely to happen again. It’s the central mechanism behind how ABA therapy works: rather than focusing on what a person does wrong, practitioners identify what strengthens desired behaviors and build treatment around that. If a child uses words to ask for a toy and then receives the toy, that access to the toy is reinforcement, because the child is more likely to use words next time.
The key distinction is that reinforcement is defined by its effect, not by what it looks like. A sticker, a snack, or verbal praise only counts as reinforcement if the behavior it follows actually increases. Something that seems like a reward but doesn’t change behavior isn’t a reinforcer in ABA terms.
Positive vs. Negative Reinforcement
These two terms cause a lot of confusion because “positive” and “negative” sound like value judgments. They’re not. In ABA, positive and negative simply describe the direction of change: whether something is added or removed.
Positive reinforcement means a stimulus is presented after the behavior. A child completes a puzzle and gets a high-five. A student raises their hand and gets called on. The behavior increases because something was added.
Negative reinforcement means a stimulus is removed after the behavior. A child puts on sunglasses and the uncomfortable glare goes away. A student finishes their worksheet and the teacher removes the requirement to do extra problems. The behavior increases because something unpleasant was taken away. This is not punishment. Both positive and negative reinforcement increase behavior. The difference is only whether the reinforcing event involves something appearing or something disappearing.
In practice, many real-world behaviors involve both simultaneously. Research published in the Journal of the Experimental Analysis of Behavior has noted that even something as simple as turning on a light can be interpreted as positive reinforcement (light appears) or negative reinforcement (darkness is removed), depending on the framing. For clinicians, what matters most is identifying the specific consequence maintaining a behavior, not debating which label fits.
Primary and Secondary Reinforcers
Not all reinforcers work the same way. Some are built into our biology, and others are learned through experience.
Primary reinforcers (also called unconditioned reinforcers) satisfy basic biological needs and require no learning. Food, water, warmth, and sleep are primary reinforcers. A child doesn’t need to be taught that a drink of water is satisfying when they’re thirsty.
Secondary reinforcers (also called conditioned reinforcers) only gain their power through repeated pairing with primary reinforcers. Money is the classic example: a dollar bill has no biological value, but because it’s been associated with access to food, shelter, and other necessities, it becomes a powerful reinforcer. In ABA therapy, verbal praise, tokens, points, and stickers are all secondary reinforcers. They work because the learner has come to associate them with things they already find reinforcing.
Token economies are one of the most common applications of secondary reinforcement in ABA settings. A child earns tokens or stars for completing tasks or demonstrating target behaviors, then exchanges those tokens for preferred items or activities. The tokens themselves have no inherent value, but the system creates a bridge between the behavior and the meaningful reward.
How Reinforcement Schedules Shape Behavior
When and how often reinforcement is delivered matters just as much as what the reinforcer is. ABA uses four basic reinforcement schedules, each producing a different pattern of behavior.
- Fixed-ratio (FR): Reinforcement comes after a set number of responses. A child gets a token after every three math problems. This produces a high, steady rate of responding with brief pauses right after the reinforcer is delivered.
- Variable-ratio (VR): Reinforcement comes after an unpredictable number of responses. Sometimes after two problems, sometimes after five. This generates the most consistent responding because the person can never predict exactly when the next reinforcer will arrive.
- Fixed-interval (FI): Reinforcement becomes available after a set amount of time has passed. Checking the oven when a timer is almost done is a fixed-interval pattern.
- Variable-interval (VI): Reinforcement becomes available after an unpredictable amount of time. Checking your phone for messages follows this pattern, because you never know exactly when one will arrive.
In ABA therapy, new skills are typically taught using continuous reinforcement first, meaning every correct response is reinforced. Once the skill is established, therapists gradually thin the schedule, shifting to intermittent reinforcement so the behavior becomes more durable and more like how the real world works. This process involves progressively increasing the gap between reinforced responses across multiple sessions until the schedule is practical for parents and teachers to maintain at home or in the classroom.
Automatic vs. Socially Mediated Reinforcement
One of the most important distinctions in ABA is where the reinforcement comes from. Socially mediated reinforcement is delivered by another person: a therapist gives praise, a parent provides a snack, a teacher grants free time. The reinforcement depends on someone else’s behavior.
Automatic reinforcement, by contrast, is produced by the behavior itself without anyone else involved. Scratching an itch is automatically reinforced by the relief it produces. Humming might be automatically reinforced by the sensory input it creates. No one else needs to be present for the behavior to be reinforced.
This distinction is critical in treatment because automatically reinforced behaviors are harder to address. With socially mediated reinforcement, a therapist can control access to the reinforcer. If a child screams to get attention, the team can redirect attention toward more appropriate communication instead. But if a child engages in a repetitive behavior because it produces a pleasurable sensation, removing access to that internal reinforcer isn’t straightforward. Treatment strategies for automatically reinforced behaviors often involve providing alternative ways to access similar sensory input.
Why Context Changes What Works
A reinforcer that works perfectly on Monday might fall flat on Wednesday. ABA accounts for this through the concept of motivating operations, which are conditions that temporarily change how valuable a reinforcer is.
An establishing operation increases a reinforcer’s effectiveness. If a child hasn’t had access to their favorite toy all morning, that toy becomes a more powerful reinforcer during the therapy session. Deprivation is the most intuitive example: food is a stronger reinforcer when you’re hungry.
An abolishing operation does the opposite. If a child just spent 30 minutes playing with that same toy before the session, offering it as a reinforcer will be far less effective. This is satiation. Skilled ABA practitioners pay careful attention to these conditions, adjusting which reinforcers they use based on what the learner is currently motivated by rather than relying on a fixed list.
Even environmental changes can shift reinforcer value. Prolonged sun exposure, for instance, increases the reinforcing value of moving into shade or removing anything touching sunburned skin. The reinforcer hasn’t changed, but the context has made it more or less potent.
Reinforcement vs. Punishment in Practice
Reinforcement increases behavior. Punishment decreases it. Both can technically change behavior, but reinforcement-based approaches are the foundation of modern ABA for good reasons.
Research on functional communication training (a common ABA technique where learners are taught an appropriate way to request what they need) shows how reinforcement and other strategies interact. In a large review of cases, simply reinforcing an alternative communication response while continuing to reinforce problem behavior was successful in only 1 out of 10 applications. When problem behavior was no longer reinforced (placed on extinction) and only the communication response produced the reinforcer, success improved but still worked in roughly half of cases. The most effective approaches combined reinforcement for the desired behavior with consistent, structured consequences for the problem behavior.
This tells us something important: reinforcement is powerful, but it works best as part of a carefully designed system where the learner’s environment consistently signals which behaviors lead to which outcomes. The goal is always to build skills and increase adaptive behavior, using reinforcement as the primary tool to get there.