What is reinforcement in ABA?
Reinforcement is anything that follows a behavior and makes that behavior more likely to happen again, and it is the core mechanism ABA uses to build new skills.
Reinforcement is defined by its effect, not by intent. If a consequence increases the behavior it follows, it reinforced that behavior; if the behavior doesn't increase, it wasn't a reinforcer no matter how rewarding it was meant to be.
It takes two forms. Positive reinforcement adds something after the behavior - praise, access to a preferred item, a turn at a favorite activity. Negative reinforcement removes something aversive, such as ending a demand once the learner asks for a break.
What works is individual and shifts over time, so teams assess preferences regularly and vary what they use. Plans also work to move from contrived reinforcers toward the natural consequences a skill will meet outside of session.
Negative reinforcement is the term most often misread. It does not mean punishment - both forms of reinforcement increase behavior, and the words positive and negative describe whether something was added or removed, not whether it was pleasant. Punishment is the separate process of a consequence making behavior less likely, and it carries its own ethical and procedural constraints.
Timing and consistency do much of the work. A reinforcer delivered immediately after the response is far more effective than one delivered minutes later, because delay makes it unclear which behavior was reinforced. Early in teaching, reinforcement is usually delivered for every correct response, then thinned to an intermittent schedule as the skill establishes - which both maintains the behavior and makes it more durable.
Satiation is the practical limit. Anything used too heavily loses its value, sometimes within a single session, which is why teams rotate reinforcers and reassess preferences rather than relying on what worked last month. Preference is also not something to assume on a client's behalf; it is assessed by observing what they actually choose and work for.
The end goal is to need less of it. Skills maintained only by contrived reinforcement tend to disappear when the program does, so plans deliberately shift toward the natural consequences a behavior will meet in ordinary life - a request being granted, a conversation continuing - so the skill survives outside of session.