Positive Reinforcement To Stop An Unwanted Behaviour?
Why Consequences Matter in Learning — and Why This Matters to NAPET
There is a persistent belief within modern dog training that unwanted or dangerous behaviours can be stopped using positive reinforcement alone. It is a comforting belief. It is marketable. And it fits neatly into a particular ideological narrative.
Unfortunately, it is also wrong.
Positive reinforcement, by definition, increases the likelihood of a behaviour occurring. If a consequence strengthens behaviour, it is reinforcement. If it reduces behaviour, it is operantly defined as punishment. That is not opinion or rhetoric, it is the foundational definition within operant conditioning.
Because of that, you cannot reliably stop a dangerous or self-reinforcing behaviour using reinforcement alone. You can influence behaviour, compete with it, or redirect it under certain conditions, but you cannot eliminate serious unwanted behaviour without consequences. That is simply not how learning works.
The Myth of the “Alternative Behaviour”
One of the most common defences of reinforcement-only approaches is the idea of teaching an alternative behaviour. The claim is that if you reinforce the “right” behaviour enough, the “wrong” one will disappear.
It sounds progressive. It sounds humane.
But operantly, it does not hold up in the real world.
Teaching an alternative behaviour does not teach a dog not to perform the original behaviour. It simply adds another behaviour to the dog’s repertoire, one that may or may not compete effectively when arousal is high or consequences are meaningful.
This approach may work in controlled, low-stress environments. But dogs do not live their lives in training halls or social media clips. Dangerous behaviours occur under pressure, excitement, fear, or instinct. In those moments, reinforced alternatives often collapse.
Without a clear consequence that communicates “do not do this”, the dog has been given only half the information.
Why Extinction Is Rarely Ethical or Realistic
Another commonly misunderstood concept is extinction, the idea that if a behaviour is not reinforced, it will simply disappear.
In practice, extinction is rarely achievable. Dogs almost never experience a situation where a behaviour produces zero reinforcement. Even intermittent reinforcement makes behaviour extremely resistant to extinction.
Worse still, extinction commonly produces an extinction burst, an escalation in intensity before any reduction occurs.
Allowing dangerous behaviour to escalate while waiting for extinction is not ethical, not responsible, and not realistic.
Positive Reinforcement Is Essential, but Incomplete
Positive reinforcement is not the enemy here.
It is absolutely critical.
It is essential for:
teaching new behaviours
reinforcing correct choices
building engagement and motivation
creating learning clarity
No trainer with any real understanding of behaviour disputes this.
But reinforcement is only one half of the equation.
Behaviour modification requires:
information about what to do
and information about what not to do
When trainers deliberately omit the “don’t” part because it is uncomfortable or ideologically inconvenient, they give dogs incomplete information.
Why Operant Conditioning Is Being Quietly Abandoned
Some trainers now claim they are “moving away” from operant conditioning, describing it as outdated or irrelevant.
This is not because operant conditioning has been disproven. It hasn’t.
It remains the foundation of every serious academic and vocational qualification involving learning across species. It is rejected not because it is wrong, but because a full understanding of it does not align with certain narratives.
Operant conditioning is morally neutral. It describes how behaviour changes in response to consequences. Nothing more.
Much of the fear centres on the word punishment, which has been deliberately misrepresented. In operant terms, punishment is anything that reduces the likelihood of a behaviour occurring.
That could be:
withholding a reward
a change in tone
body language
physical intervention
removal of access
It does not automatically mean cruelty or abuse. Claiming otherwise is a rhetorical tactic designed to shut down discussion.
A Real-World Example: Why Consequences Matter
Imagine teaching a ten-week-old Labrador puppy to sit. Positive reinforcement is absolutely the best way to do this. You lure, capture, mark, and reinforce.
Now fast forward a year.
That same dog is standing beside a busy road. You ask for a sit. The dog breaks position and moves toward traffic.
At that moment, the fact that you once reinforced the sit does not teach the dog not to step into the road.
You will need immediate, consequential intervention, your voice, physical restraint, body blocking, or other operant punishment.
To pretend otherwise is fantasy.
If we do not apply ethical, well-timed training-based consequences, the environment will. And the environment does not warn, negotiate, or care about ideology.
Ethical consequences in training are not cruelty.
They are preparation for reality.
The Welfare Question Nobody Wants to Ask
If a methodology keeps a dog and owner locked in prolonged stress for months or years, sometimes costing thousands, while gently managing a behaviour that could be resolved through a short, clear learning experience, is that really humane?
Is it kinder to allow ongoing confusion, frustration, and risk simply to avoid momentary discomfort?
Or is the word humane being used to protect ideology rather than dogs?
We see the consequences of this constantly:
reinforcement-only approaches failing serious cases
owners told to “avoid triggers” indefinitely
dogs placed on heavy psychotropic medication
behavioural euthanasia framed as the final option
Over many years, countless dogs on the brink of euthanasia have gone on to live stable, normal lives, not because of force, but because of clarity.
The difference was not cruelty.
It was complete information.
The Science Problem: Cortisol Without Context
Many studies used to argue against non-reinforcement-only methods rely heavily on salivary cortisol as a welfare indicator.
Cortisol is not a “bad hormone.” It is a general arousal hormone. It rises during anticipation, excitement, play, work, frustration, fear, and pleasure.
Dogs experience cortisol spikes during:
play
hunting
tracking
problem-solving
training
excitement
Cortisol tells us something matters.
It does not tell us whether that thing is good or bad.
Using cortisol alone to label training as abusive, without context, duration, recovery, behaviour, or skill of application, is scientifically weak.
Elevated cortisol does not equal abuse. Treating it as a moral verdict rather than a physiological signal strips away reality.
Why This Matters to NAPET
This debate goes to the heart of why NAPET exists.
When correction is treated as taboo:
standards collapse
trainers soften language instead of improving practice
owners are misled
dogs are placed at risk
tools become political targets
bans become inevitable
NAPET rejects both extremes:
punishment-only training
reinforcement-only ideology
We stand for:
ethical, proportionate consequences
professional competence
welfare measured by outcomes, not optics
prevention rather than apology
Dogs live in the real world. Roads, livestock, law, and risk do not pause for ideology.
Correction is not abuse.
Avoiding it at the expense of clarity is.
Positive reinforcement is vital, but on its own, it cannot stop dangerous or unwanted behaviour. Anyone claiming otherwise is either deeply misled or deliberately misleading.
Dogs need clarity.
Clarity is welfare.
And ensuring that clarity is delivered ethically, competently, and responsibly is exactly why NAPET exists.