Continuous vs intermittent reinforcement
Continuous reinforcement delivers a reinforcer after every instance of the target behavior, while intermittent reinforcement delivers it after only some instances, on a ratio or interval schedule. Continuous reinforcement is best for teaching new skills; intermittent reinforcement keeps learned skills going with fewer reinforcers.
What is the difference between continuous and intermittent reinforcement?
With continuous reinforcement, written as CRF or FR 1, every correct response earns the reinforcer. With intermittent reinforcement, only some responses do, based on a count or time requirement that may be fixed or variable. Continuous reinforcement builds a new behavior quickly, while intermittent reinforcement makes a behavior more durable and less likely to stop when reinforcement pauses.
The move from one to the other is called thinning. A supervisor may start a new skill on CRF, then move to FR 2, then VR 3 and so on as the skill becomes reliable.
Intermittent reinforcement also works against the team when it happens by accident. If a problem behavior on extinction is reinforced now and then, it becomes more persistent than before.
Continuous and intermittent reinforcement side by side
| Continuous (CRF) | Intermittent | |
|---|---|---|
| How often reinforced | Every response | Some responses |
| Schedule names | FR 1 | FR, VR, FI, VI |
| Best for | Acquiring a new skill | Maintaining a learned skill |
| Resistance to extinction | Low | Higher |
| Satiation risk | Higher | Lower |
| Example | Praise and a chip for every correct color | A token after every third correct color |
How do you tell continuous and intermittent reinforcement apart on the exam?
Ask whether every single correct response is followed by reinforcement. If yes, it is continuous. If any correct responses go unreinforced by design, it is intermittent, whatever the specific schedule. Then, if the question asks why, link continuous to fast learning and intermittent to durability, thinning and resistance to extinction.
- Every response reinforced: continuous.
- Some responses reinforced: intermittent.
- New skill: continuous is usual.
- Mastered skill being maintained: intermittent is usual.
Worked scenarios
Scenario 1. A learner is just starting to point to request. Every point is honored immediately. Answer: continuous reinforcement.
Scenario 2. A learner who reliably says "hi" gets an enthusiastic reply about every third time and a neutral nod otherwise. Answer: intermittent reinforcement, used to maintain the skill.
Scenario 3. A learner's tantrums are on extinction, but a relative occasionally gives in. Tantrums become harder to stop. Answer: accidental intermittent reinforcement of problem behavior.
Scenario 4. A supervisor changes a target from reinforcing every correct answer to reinforcing every other one. Answer: thinning from continuous to intermittent.
What are the common traps with continuous and intermittent reinforcement?
A trap is assuming more reinforcement is always better. Continuous reinforcement can cause satiation and fragile skills if it is never thinned. Another is thinning too quickly, which can lead to the skill breaking down. A third is missing that intermittent reinforcement of problem behavior, like occasional giving in, strengthens the problem rather than fading it.
- Thinking CRF should be used forever.
- Thinning without a supervisor's plan.
- Overlooking occasional reinforcement of problem behavior.
When would you see each in a session?
Continuous reinforcement appears in new acquisition targets and in early functional communication training, when every appropriate request should work. Intermittent schedules appear in maintenance programs, token boards and once skills are reliable. Your supervisor sets the schedule and decides when to thin it. You follow it exactly and report if a skill starts slipping.
Consistency matters most during thinning. If one technician reinforces every response and another every fifth, the learner experiences a confusing schedule nobody planned.
Keep data while the schedule changes. If correct responding drops after a thinning step, record it and tell your supervisor, who may return to a denser schedule for a while before trying again.
Practice questions
A supervisor is introducing a brand-new target, naming coins, to Priya, 8. Which reinforcement schedule is usually best while she is first learning the skill?
After Omar, 12, learns to check his visual schedule, the team moves from reinforcing every check to reinforcing only some checks. Why is this change helpful for maintenance?
A parent is using planned ignoring for attention-maintained bedtime crying, as the plan describes. She tells the technician that she ignores the crying most nights but goes in to comfort her son about once a week. What is the most likely effect?
Lena, 9, is sorting silverware at her group home. Her technician gives her a point after every fourth item she sorts correctly, every time. Which schedule of reinforcement is this?
Questions people ask
Is FR 1 continuous reinforcement?
Yes. FR 1 means every response is reinforced, which is continuous reinforcement.
Who decides when to thin reinforcement?
The supervising BCBA decides, based on data. The technician follows the updated schedule.
How ready are you?
Ten questions, about 3 minutes, no account. Get your readiness score and your weakest domains.
Check my readiness