Dog training glossary
Differential Reinforcement
Quick answer
Reinforcing one specific behavior while withholding reinforcement for a different, unwanted one, so the dog gradually shifts toward the rewarded option.
Written by the TailorPup team, updated September 2026. See our editorial approach.
What it means
Differential reinforcement works by making an acceptable behavior consistently more rewarding than the problem behavior it competes with, rather than trying to punish the unwanted behavior directly. A common version rewards any behavior other than the problem one (for example, anything besides jumping on guests), while another version specifically rewards an incompatible behavior, like sitting, which physically cannot happen at the same time as jumping. Both rely on withholding reinforcement for the old behavior long enough for it to fade through extinction while the new one strengthens.
A real example
A dog that jumps on visitors for attention gets no attention at all while jumping, but an enthusiastic greeting and a treat the moment all four paws stay on the floor, so calm greetings become the more reliably rewarding option over repeated visits.
A worked example
Kai's dog barks for attention. Kai waits for a quiet moment, even two seconds, and rewards it. He ignores the barking and gradually extends the quiet time. After a few weeks, the dog lies down quietly when it wants attention.
Signs it is working
- The rewarded behavior becomes more frequent.
- The unwanted behavior declines.
- The dog offers the alternative to get attention.
What it is not
It is not just ignoring. You must actively reward the alternative, or the dog will find another way to get what it wants.
Common mistake
Rewarding the alternative behavior inconsistently while still occasionally giving in to the unwanted one, which keeps the old behavior alive on an intermittent schedule instead of letting it fade.
Why it matters
It gives the dog a clear choice: this behavior pays, that one does not. That is gentler and more effective than just trying to suppress the unwanted one.
How to use it
- Choose the behavior you want instead, one that cannot happen together with the problem.
- Reward it every time, and withhold reward for the unwanted behavior.
- Make the new behavior easy, so it wins most of the time.
- Once it is reliable, gradually reduce how often you reward it.
Does it work for barking?
Often yes. You reward quiet moments, or a calm alternative, and stop rewarding the barking. It works best if you also deal with whatever is causing the barking.
Related terms
Know the term.
Now build the plan.
TailorPup turns concepts like this into a personalized 12-week plan for your dog's breed, age, and behavior.