Modern dog training: an introduction to positive-reinforcement methods
If you are new to dogs, or new to thinking carefully about how you train the one you already have, the most useful sentence you can absorb early is this: dogs do what works. They do not think about right and wrong. They notice, very quickly, which behaviors make their world better and which behaviors make their world worse, and they repeat the ones that pay off. Modern training is built on that small, unsentimental fact. Once you accept it, almost everything else becomes simpler.
A short word on what training is
Training, in the modern sense, is not about obedience and it is not about dominance. It is about teaching your dog which behaviors reliably pay off so that the dog volunteers those behaviors more often. Positive reinforcement — reward the behavior you want, withhold reward when the dog gets it wrong — has been the consensus of veterinary behaviorists and certified trainers for more than two decades. The dominance-based, correction-heavy approach popularised by some early television shows on both sides of the Atlantic is not just outdated; the peer-reviewed evidence is consistent that it produces worse outcomes and higher rates of anxiety and aggression.
If a trainer recommends prong collars, e-collars, alpha rolls, or “showing the dog who is boss,” find a different trainer. The British Veterinary Behaviour Association (BVBA) and the American Veterinary Society of Animal Behavior position statements on this is unambiguous.
The three ingredients
Marker. A short, distinct sound that means “what you just did is the thing I am about to pay you for.” A clicker works beautifully because it is sharp and consistent, but a single, sharp word — most people use “yes” — works almost as well. The marker has to come within a second of the behavior, or the dog will reward the wrong thing.
Reward. Something the dog actually values, delivered immediately after the marker. For most dogs, small, soft food rewards that disappear in a second are best, because they let you keep training without breaking the rhythm. Cut tiny cubes of cooked chicken, hot dog, or a low-fat training treat. Use better rewards in harder environments. A dog who works for kibble at home may need cheese in the park.
Patience. Five short sessions a day are worth more than one long one. Two minutes of training, ten times a week, will outperform an hour-long Saturday session every time. Dogs learn in small, repeated exposures, and they fatigue mentally faster than we realise.
What to teach first
Most people start with sit. There is nothing wrong with sit — but the more useful early skills, by a wide margin, are name response, look, and settle on a mat. Together those three behaviors prevent more daily friction than any number of advanced obedience cues.
Name response is exactly what it sounds like: say the dog’s name, the dog turns their head toward you, you mark and reward. Repeat until the dog’s name is the most reliable sound in their life. This is the foundation for recall later. Many dogs ignore their name because their owners use it as a constant low-level background noise; train it the way you would train any other cue.
Look is a sustained version of name response: the dog makes and holds eye contact with you for one second, then three, then five. It is the single most useful skill on a busy sidewalk.
Settle on a mat teaches the dog that the mat is the calm-and-quiet place. Put the mat down, reward the dog generously every time they choose to lie on it of their own accord, and within a week most dogs will choose the mat over the couch when something stressful happens.
The single biggest mistake
Almost everyone goes too fast. They train a behavior in their living room, decide the dog “knows it,” and then expect it to work at the park, with squirrels, off leash, after three days. It will not. Behaviors generalise slowly. A dog who reliably sits in the kitchen is not the same dog as one who reliably sits at the entrance to a dog park.
Add difficulty one variable at a time: first distance, then duration, then distraction. If you fail at one level, drop back to the previous level for a few easy reps and rebuild the dog’s confidence. This is the rhythm of every good training plan.
When to call a professional
Some behavior — fear-based aggression, severe resource guarding, separation anxiety that escalates rather than improves — needs a credentialed professional and sometimes a veterinary behaviorist. Look for the letters CPDT-KA, KPA-CTP, or DACVB after a trainer’s name. Avoid anyone who guarantees results in a week or recommends correction tools.
Where to go next
If you are still in the first week with a new dog, our companion guide to a calm, practical first seven days is here. For our broader approach to writing about training and behavior, see our editorial standards. The newsletter will let you know when the next pieces in this series — on recall, on loose-leash walking, and on the behavior problems most owners create accidentally — go up.