Most problems we see between dogs and owners are not about stubborn dogs. They are about two species trying to talk to each other without a shared language. Once you understand how your dog takes in the world, how it learns and how it tells you what it is feeling, training gets a lot simpler and a lot fairer.
A world made of smells
Your dog's nose is its main window on the world. Researchers who studied the airflow inside a dog's nose found that it splits each sniff, sending part of the air to a separate area packed with scent-detecting tissue, where it is not simply breathed out again (Craven, Paterson and Settles). That is a big part of why dogs can pick up smells we will never notice.
Hearing comes next. Dogs hear higher pitches than we do, which is one reason a high, happy voice can bring a dog in when a deep, flat one does not. Sight matters too, but in a different way than ours. Dogs have two types of colour cones instead of our three, so they see a smaller range of colour, closer to what a person with red-green colour blindness sees (Neitz, Geist and Jacobs). They are very good at picking up movement, which is why a small shift of your hand or shoulders can mean more to your dog than the word you said.
In training, that means the walk is not just exercise for your dog. It is the news of the day. Sniffing is one of the best rewards you have, and you can use it, as we explain in our article on the Premack principle.
How the dog mind works
Dogs live in the moment. They are very good at learning what goes with what: the leash means a walk, the car keys mean you are leaving, the cupboard door means supper. That is association, and it is how almost everything your dog knows got there. What dogs are not doing is sitting and reasoning out a plan, or thinking back over what they did an hour ago.
The "guilty look" is the best example. Most owners are sure their dog knows it did wrong. A study at Barnard College tested this by having owners leave the room and then telling them, sometimes falsely, whether their dog had eaten a forbidden treat. The dogs looked most "guilty" when their owners scolded them, whether or not the dog had actually broken the rule (Horowitz). The look is a reaction to you, not a confession.
So coming home and scolding a dog for something it chewed hours ago does not teach it anything about chewing. It only teaches the dog that you are unpredictable when you walk in the door.
How dogs learn
There are two main ways dogs learn, and good training uses both.
Classical conditioning
This is learning by pairing. Something that means nothing to the dog, like a word, becomes meaningful because it always comes right before something that matters. That is exactly how a marker works. We say "yes" and then pay with food or a toy, again and again, until "yes" itself makes the dog light up. Brain research backs this up. Dogs trained to lie still in an MRI scanner showed activity in the brain's reward area when they saw a hand signal that meant food was coming (Berns, Brooks and Spivak). Another scanner study found that dogs process the words we say and the tone we say them in, and the reward area responded most when praise words were said in a praising tone (Andics and colleagues). Your words count, and so does how you say them.
Operant conditioning
This is learning by consequences. Behaviour that pays off happens more. Behaviour that does not pay off, or that costs the dog something, happens less. If the terms are new to you, our four quadrants article walks through them in plain English.
Timing, markers and motivation
Because dogs learn by association, timing is everything. The marker has to land the moment the behaviour happens, not a few seconds later when your dog has already moved on. That is the whole point of a marker. It lets you say "that, right there" with perfect timing, even if the reward takes a second to arrive.
Motivation is the fuel. Our approach is built on the motivational, marker-based training taught by Michael Ellis, which we follow in our own way at ACA. A dog that wants what you have, whether food, a tug or the chance to go sniff, learns faster and works happier. We spend a lot of time building that drive and figuring out what each dog truly values.
Fair corrections
Corrections have a place, but only on behaviours the dog already knows well. A dog that has not learned a behaviour is not disobeying, it simply does not understand yet, and correcting it is not fair. When a dog that knows a behaviour well chooses to ignore it, we say "no" first, then give a correction matched to the situation, then help the dog get it right and pay it. We never use the e-collar for corrections on puppies.
How dogs communicate
Dogs talk all the time, mostly with their bodies.
- Ears. Forward usually means interest or focus. Pinned back can mean worry, fear or appeasement, depending on the rest of the body.
- Tail. A wagging tail does not always mean a happy dog. Height, speed and stiffness all matter. One study even found that dogs tended to wag more to the right when they saw their owner and more to the left when they saw an unfamiliar, dominant dog (Quaranta, Siniscalchi and Vallortigara).
- Eyes. Soft eyes are relaxed. A hard stare is a warning. Seeing the whites of the eyes, often called whale eye, is a sign of stress. Dogs have also developed a muscle that raises the inner eyebrow, giving that "puppy dog eyes" look. Wolves mostly lack it, and dogs use it more when a person is looking at them (Kaminski and colleagues).
- Voice. Barks, whines, growls and groans all carry meaning, and the same sound can mean different things in different situations. A growl is information. Punishing it can take away your dog's warning without changing how it feels.
Calming and stress signals
Turning the head away, licking the nose, freezing, yawning and sniffing the ground are often called calming signals. A study of off-leash meetings between dogs found these behaviours mostly showed up while the dogs were interacting, more often with unfamiliar dogs, and that when one dog showed them after the other acted aggressively, the aggression usually went down (Mariti and colleagues). The researchers called it a pilot study, so there is more to learn, but it matches what we see every day. When you see a lip lick or a look away during training, your dog may be telling you it is under pressure. Slow down, add distance or make the task easier.
The warning ladder
Most dogs climb a ladder of warnings before they bite: calming signals first, then a hard stare, raised hackles, a growl or teeth, an attempt to leave, then a snap or bite. When people say a dog bit with no warning, a step was usually missed or punished.
Reading us
Dogs are remarkably good at reading people. In a well-known study, dogs used human pointing and gaze to find hidden food better than chimpanzees, and better than wolves raised by people (Hare and colleagues). Research groups like the Duke Canine Cognition Center and the Family Dog Project in Budapest have spent years studying this skill. It is a gift for training, but it cuts both ways. Your dog is watching your body all the time, so if your hands, posture and words say different things, your dog will usually believe your body.
The practical rule: give the word first and the hand signal a second later, so your dog learns both. Given together, it learns only the hand.
Genetics shape how easily a dog becomes fearful or reactive; training changes what it does with that.
How this shapes the way we train
Everything above is why we teach in the same order every time:
- Lure. We use food or a toy to help the dog into the position, so it does not have to guess.
- Mark. "Yes" tells the dog the exact moment it got it right, then the reward follows.
- Add the word. Only once the dog is doing the behaviour reliably do we put a word to it, so the word gets paired with the right picture.
- Repetitions. Short, upbeat sessions build the habit without wearing the dog out.
- Proofing. We add distractions, distance and new places one at a time, because dogs do not automatically carry a lesson from the kitchen to the park.
We also do not teach a separate stay. Since dogs learn from clear, consistent patterns, we keep it simple: sit means sit and down means down, until we release with "okay". One rule, every time, is far easier for a dog to understand than a list of exceptions.
When you train with how your dog actually thinks, you stop fighting your dog and start working with it. That is where the real progress happens.
Keep reading
For more on rewards in real life, read The Premack Principle: Using What Your Dog Wants as the Reward and When the Squirrel Wins: Competing Reinforcers in Plain English. To work on this with us in person, see our training courses, from Puppy Foundations to Basic Obedience and beyond.