09/02/2026
Let's talk punishment.
Punishment means a lot of different things to different people. If you Google "punishment," a wide range of definitions will come up.
For the purpose of dog training and behavior, though, we're typically talking about B.F. Skinner's "four quadrants of operant conditioning":
• Positive Reinforcement: adding something pleasant to increase the likelihood a behavior is repeated
Dog sits ➡️ you give them a treat ➡️ sitting increases.
• Positive Punishment: adding something unpleasant to decrease the likelihood a behavior is repeated
Dog pulls ➡️ leash correction is applied ➡️ pulling decreases.
• Negative Reinforcement: removing something unpleasant to increase the likelihood a behavior is repeated
Leash pressure is applied ➡️ dog sits ➡️ pressure is released ➡️ sitting increases.
• Negative Punishment: removing something pleasant to decrease the likelihood a behavior is repeated
Dog jumps for attention ➡️ you remove your attention ➡️ jumping decreases.
**this isn't training advice, just examples**
Those terms can be confusing because most people hear "positive" and "negative" and think good or bad.
Scientists like to make things difficult though. 🫠
In this context:
Positive = something is added.
Negative = something is removed.
Reinforcement = the behavior increases.
Punishment = the behavior decreases.
And that's an important distinction because something isn't reinforcement or punishment just because we intended it to be.
If you give your dog a treat for sitting but sitting doesn't increase, technically it wasn't sufficient positive reinforcement.
If you correct your dog for pulling but pulling doesn't decrease, technically it wasn't sufficient positive punishment.
The consequence to the behavior determines whether something was punishing or reinforcing.
And if we're using these definitions, reinforcement and punishment are happening ALL THE TIME naturally. They don't necessarily have to be something we intentionally do.
Your dog jumps on the couch ➡️ finds a comfy spot ➡️ jumping on the couch increases.
Your dog steps in a puddle ➡️ hates having wet feet ➡️ starts avoiding puddles.
Their behavior is constantly being influenced by what happens before and after it.
This adds a lot of complexity to conversations about punishment in dog training.
It also opens the door for claims like:
"Positive reinforcement only/force-free doesn't exist."
"All four quadrants happen naturally, so we should use all four in training."
"Dogs need punishment to learn."
And on, and on, and on...usually to justify the use of harsh punishment.
But something happening naturally and intentionally using it as a training strategy are two different conversations.
There's also another part of punishment that I think gets overlooked:
Punishment can be reinforcing for the HUMAN using it.
Dog barks ➡️ human does something ➡️ barking stops.
The human just got relief from something they didn't want happening.
Which means the HUMAN may be more likely to do that same thing again the next time the dog barks.
Funny how that learning theory works both ways. 🫠
And that's where this conversation starts getting a whole lot more complicated.
Next we'll talk about the ethics of punishment and then what trainers mean when they talk about "fallout."
And if you're struggling with your dog's behavior, reach out, let's talk, and make a plan that works for you!