Training Framework

Differential Reinforcement

Your dog jumps for attention, so you teach them to sit instead. When barking gets excessive, you reward quiet moments. When your retriever finally drops the tennis ball, they get an even better treat. You’re already using differential reinforcement — you just might not know its name.

What Differential Reinforcement Actually Does

Differential reinforcement means you reinforce certain behaviors while withholding reinforcement from others. Instead of simply trying to stop unwanted behavior, you deliberately build up specific alternatives. The key is in the word “differential”: you’re making a clear choice about which behaviors earn reinforcement and which don’t.

This isn’t about dominance or correction. It’s about giving your dog a clear path to success by making the behaviors you want more rewarding than the ones you don’t. When applied well, your dog starts offering better choices because better choices produce better outcomes.

The Four Types You Need to Know

There are four main differential reinforcement procedures, each suited to different training situations:

DRA

Differential Reinforcement of Alternative Behavior

Teach a specific replacement that serves the same function as the problem behavior. If your dog jumps for attention, teach them to sit for attention instead. The key is matching the function — both behaviors get the dog what they actually want.

DRI

Differential Reinforcement of Incompatible Behavior

Reinforce a behavior that physically cannot happen at the same time as the problem. A dog with four paws on the floor cannot be jumping. A dog lying on their mat cannot be counter-surfing. This gives you a structural guarantee.

DRO

Differential Reinforcement of Other Behavior

Reinforce the absence of the problem behavior for a specific time period. “No barking for 30 seconds earns a treat.” Your dog can do anything except the target behavior. Simple to implement, but it doesn’t teach a specific replacement.

DRL

Differential Reinforcement of Low Rates

Use when some level of the behavior is acceptable, but the current rate is too high. Alert barking is useful; continuous barking is not. Reinforce when your dog barks fewer than three times per visitor instead of the usual fifteen.

Why Function Matters More Than Form

The most common mistake is focusing on the behavior’s form (what it looks like) instead of its function (what it accomplishes for the dog). Teaching a dog to sit when they want attention will work. Teaching them to lie down when they want attention will fail if lying down doesn’t reliably produce attention.

This is why DRA is usually your best choice — it directly addresses why the dog was doing the problem behavior in the first place. If jumping gets attention and sitting gets attention, sitting becomes the more efficient option. If jumping gets attention but sitting only gets treats, you haven’t solved the attention need.

The Function-Matching Rule

Your replacement behavior must produce the same outcome the problem behavior was producing. If you’re not sure what function the behavior serves, ask: What does my dog get or avoid when they do this? Your replacement must provide the same thing.

Practical Application Examples

Here’s how to choose which type of differential reinforcement fits your situation:

Problem: Dog jumps on guests for attention. Use DRA. Teach sitting for greetings. Both jumping and sitting can get attention, but you’ll only give attention for sitting. The function (getting attention) stays the same; the form (sitting vs. jumping) changes.

Problem: Counter-surfing in the kitchen. Use DRI. Train “four on the floor” and reinforce heavily when all four paws stay on the ground near the counter. The dog physically cannot surf the counter while maintaining four paws on the floor.

Problem: Excessive barking during TV time. Use DRO if you want any quiet behavior, or DRL if some barking is acceptable. DRO: “Every two minutes of quiet gets a chew.” DRL: “Fewer than five barks during this 30-minute show gets a special treat.”

The Extinction Component

Differential reinforcement works best when you pair it with extinction — stopping reinforcement for the problem behavior. If you reinforce sitting for attention but still occasionally give attention for jumping, you’re running a lottery system where jumping sometimes pays off. Consistency matters.

Research shows that when extinction is combined with differential reinforcement, Extinction bursts drop from 36% to about 12% (Lerman & Iwata, 1995) of cases. The alternative behavior gives your dog something to do instead of escalating the problem behavior.

Common Implementation Error

Don’t default to DRO (rewarding the absence of behavior) when DRA (teaching a specific replacement) would work better. Telling your dog what NOT to do without showing them what TO do often creates a confused animal cycling through random behaviors until they stumble onto reinforcement.

When Differential Reinforcement Fails

The most common failure point is mismatched functions. If your dog’s jumping is actually about accessing something you’re holding (tangible reinforcement) but you treat it as attention-seeking, no amount of training sits for attention will solve the problem. Always start by understanding why the behavior is happening.

Another failure point is insufficient reinforcement value. If the problem behavior has been heavily reinforced for months or years, your alternative behavior needs to be significantly more rewarding, at least at first. Don’t expect a dog that’s been getting attention for jumping to suddenly prefer sitting for brief praise.

The Power of Strategic Choice

Differential reinforcement isn’t about controlling your dog — it’s about structuring their choices. When you make appropriate behaviors more rewarding than inappropriate ones, your dog will choose appropriately. Not because they have to, but because it works better for them.

Research foundation from Vollmer & Iwata (1992), Athens & Vollmer (2010), and Protopopova et al. (2016).