Tuesday Training Byte: Better methods and better results
Today I was under anesthesia for a medical procedure, so am sharing something from a fellow trainer- George Van Huyssteen- regarding rewards and training methods. I could not have said it any better. He clearly explains the many questions that I commonly get.
Today I heard another person spreading the word that using food for behaviour modification in dogs is ridiculous, so here is another solid explanation of why we use food when working with a dog's behaviour
I actually get asked this a lot: "Isn't using food just bribery?" or "Will my dog only behave if I've got treats on me?" It's a fair question, so let me actually answer it properly instead of just saying "trust me, it works."
Food isn't just a reward we've picked because dogs like it. In learning theory, there's a real distinction between what's called a primary reinforcer and a secondary one. Food, along with things like play and social contact, is primary — a dog is wired to value it without any learning required. Praise and clicker sounds are secondary, and they only actually mean anything once they've been reliably linked to something primary like food. That's part of why a cheerful "good boy" often does nothing for a dog who's properly stressed, while a bit of chicken gets through.
But here's the bit I actually find fascinating, and the part that changes how I explain this to clients. Most of what we're doing with a fearful or reactive dog isn't really "training" in the sit-and-stay sense. We're not teaching the dog what to do. We're changing how the dog feels. That's classical conditioning, not operant conditioning, and it's the backbone of counter-conditioning work. When we feed a dog for noticing another dog, a car, a stranger — whatever the trigger is — we're not rewarding the reactive behaviour. We're pairing that trigger with something good, over and over, until the dog's gut-level emotional response to it starts to shift. It's been tested properly too, not just something trainers say to sound scientific: dogs who went through structured counter-conditioning for vet-visit fear showed real, measurable drops in fear over just four weeks.
There's actually a neurological reason this works, and it's worth knowing. The brain has what's known as a SEEKING system — dopamine-driven, switched on by anticipation of something rewarding, food included. It sits in a different network from the FEAR system, and the two appear to work in "opposition" to each other (it is more complicated than that, but let us keep it simple). So when a dog is calmly taking food around something that used to send him over the edge, that's not a trick or a distraction. You may genuinely be activating a brain system that competes with the fear response itself. And on the flip side, a dog who won't eat at all is telling you something important—that it's too far over threshold for any learning to happen right then.
This is also why we lean so hard on reward-based methods and steer well clear of corrections and punishment in behaviour work. It's not just a philosophical stance. A growing pile of research shows aversive methods are linked to higher stress hormones, more fear and aggression, damage to the dog-owner relationship, and the dog's ability or wanting to learn new things becomes worse over time. Some dogs subjected to repeated, unpredictable punishment shut down completely and stop trying — and that flat, quiet dog often gets mistaken for "well behaved" when really it's given up. That's not a state any of us should want to create in a dog we love.
So no, it's not bribery. Bribery is showing food beforehand to lure a response. What we're doing is reinforcement and conditioning — food delivered to strengthen a behaviour or reshape an emotional response, faded systematically as the new pattern takes hold on its own. It's a bridge to get the dog to a better place, not a permanent crutch.
So next time you see a dog working through fear or reactivity with a treat pouch on the handler's hip, know that it's not a shortcut. It's the science doing exactly what it's meant to do. Happy Training!
Comments