Frequently Asked Questions

If you want to train without punishment, essentially yes.

Most of what we ask dogs to do is not fun for a dog. Sitting still while a stranger walks past, coming away from a smell, lying quietly while guests eat dinner. Something has to make that worth doing, and praise and affection alone rarely carry a dog through learning something new. It is much the same reason you would not work a forty hour week for a compliment.

Food also does a second job that nothing else does as well: it guides movement. A piece of food at your dog's nose can draw them into a sit or a spin before they have any idea what the word means, and then it gets faded out. So food is both a lure at the start and the payment afterwards.

See Treats.

Yes, and far less than you fear. But we would gently push back on the goal itself.

Behavior depends on its consequences, so a behavior that never pays eventually fades. The usual advice from here is to move to an unpredictable, occasional payment, like a slot machine, because that makes behavior persistent: slow to disappear when the payment stops.

That is true, and it is worth asking what it buys you. Persistence is not the same as enthusiasm. A slot machine produces someone who keeps pulling the lever, not someone who is delighted to be there, and a dog on a very thin schedule often gives you exactly that: a sit that still happens, eventually, without much joy in it. Persistence is genuinely useful when a behavior has to survive long stretches without payment. But for the everyday things you will go on asking for, week in and week out, persistence is not the quality you are after. What you want is enthusiasm: a dog who answers quickly and gladly because answering you has always been worth their while. Nobody works enthusiastically for free, and dogs are no exception.

The behavior analyst Susan Friedman puts the practical version better than we can: you had to show up to cue it, so why miss the chance to pay for it (Friedman and Martin, Blazing Clickers). If you are there to ask, you are there to reinforce.

So the honest answer is not "wean your dog off food." It is:

• The food rate drops steeply on its own once a behavior is learned. You will get far more behavior for far less food without trying.
• Widen what counts as payment rather than shrinking the payment. A game, a sniff, the door opening, permission to greet a friend, the leash coming off. You are giving these away anyway.
• Keep paying the things you cannot afford to lose. Recall especially. A recall is the most expensive behavior you own, and it is the last one to put on a diet.

See Rate of Reinforcement.

Work out why, because the reason changes what you should do. Start with the possibility owners miss most often: your dog is frightened. A dog who will not eat is telling you they are over threshold, and no treat is good enough to fix that. Add distance, turn the trigger down, or stop for today. Do not go looking for better cheese.

After that, in order: too excited or distracted, not hungry, does not rate the treats, or unwell. A dog who suddenly stops eating food they normally love needs a vet rather than a training solution.

And the rule that runs the other way, worth memorising: a dog who IS eating is not necessarily comfortable. Eating tells you arousal is low enough to eat, and nothing more. See Stress Signals.

The full list, with what to do about each, is in Treats.

Start somewhere boring. Your living room, with nothing going on.

New behaviors need a place where your dog can concentrate and succeed. Once a behavior is reliable there, move it outward in steps: the hallway, the garden, the quiet street, the busier street, the park. Each new place makes the behavior harder, so make everything else easier when you arrive somewhere new. Shorter, closer, better treats.

The honest version of the rule: if you never practise at the park, do not expect it at the park. A sit learned in your kitchen is, as far as your dog is concerned, a kitchen sit.

See Generalization and Proof.

Short and often beats long and occasional. Five to fifteen minutes suits most dogs, and some steps want less than that.

Train as often as your life actually allows. Daily is lovely, and a day or two off is not lost ground: dogs consolidate in the gaps, and a rested dog often comes back better than they left. Two short sessions a week that genuinely happen beat a daily plan you abandon on Thursday.

The one rule worth keeping: stop while your dog is still keen. Ending a session with your dog wanting more is what makes them come back ready.

See Session.

No. This is one of the most common pieces of advice out there and it is backwards.

Starting with food and then "proofing" with a choke, prong or shock collar is not a natural next step, it is a different training method bolted onto the first one. It confuses a dog who has been told all along that getting it right pays, it damages the trust you spent months building, and the research links punishment-based methods to more fear, more aggression, and a worse relationship with the handler without producing better obedience. See Gear and Methods to Avoid and Positive Punishment.

What actually proofs a behavior is unglamorous: practise it in new places, around real distractions, with better treats and lower criteria until it holds. The app gives you four ways to do that, none of which involve a correction:

• Proof steps — replay just the hardest end of a finished plan somewhere new. See Proof Step.
• Proofing games — cues from different plans, shuffled, so your dog has to listen instead of guessing the pattern.
• Blue goals — anything blue on your home screen is the app naming the skill that wants review.
• Other people — let family and friends run proof steps or a game. A cue that only works when you are the one holding the treat pouch is not finished yet, and this is the cheapest way to find out.

It takes longer to start and it finishes faster, and nothing you build this way has to be maintained with fear.