Some say using food to train dogs is bribery. It isn’t, it’s pay. Unless you’re working to help a dog overcome a fear, treats are simply fair reward for services rendered. If you use them as a "bribe" or advance payment, you can run into all sorts of trouble. Here’s a cautionary tale from very close to home. But I’m going to let my partner, Nina, tell it because I’ve got a cast iron alibi.
Outmanoeuvred by a dog's long con
I have to make clear from the outset that I am not the professional dog trainer in the family. I have other skills. Really. But since I live with one, I assumed I’d picked up enough along the way through a process of informational osmosis. As things turned out, I should have been paying more attention.
Si was in hospital, leaving me sober in charge of our rescue Rottweiler. Murphy was a clever dog. I knew this. He was also a world-leading, world-beating, world class grifter. I knew this, too. It didn’t save me.
Day One and I was working from home, but had a chunk of time before the next Zoom meeting to take Murph out for a run. We got in the car and headed for the hills. Walk over and back at the car, I opened the boot and said confidently: “In”. Nothing. Not a muscle twitch. So I chucked a couple of treats in, just to hurry things along. It didn’t. He tilted his head and gave me the long look of a con artist sizing up a potential mark. Then he sat down. I looked at the time and had a small prickle of panic.
I am, in truth, a bit of a catastrophist. This might have been just a small delay in getting a dog into the car. But spooling out in my head were images of still being on a track in the middle of nowhere as night fell, locked in a battle of wills, Zoom meeting missed, fired from the contract, never working again etc, etc. I threw the entire contents of the treat bag into the back of the car. Murphy jumped in. Crisis averted. Zoom call acquired. The triumph of pragmatic reinterpretation of positive reinforcement methods.
But no. The next day, the price of getting back into the car at the end of the walk had risen. The handful of treats wasn’t enough. I had to add the bonus of a Nairn’s cheese oatcake from a stash kept in the car for energy dips. Within no time, treat inflation was out of control and the ‘in the car’ price had rocketed to a handful of treats, a piece of Manchego cheese, two cheese oatcakes and quarter of a lamb and rice stick.
After reaching this peak, the rate of inflation did sink back to zero. Murphy was, after all, a smart cookie who knew that going for too reckless a plan for growth would crash the household economy. But the price of getting him back in the car was still eye-wateringly high. I’d set out, treat pouch bulging with food, just in case. Still, it was all under control.
Then after one bright, windswept walk, I paid the advance, got him back into the car, drove home, opened the boot and said, cheerfully: ‘Out you go.’ Nothing. Murphy had decided to go with the growth plan after all. Getting out of the car should have its price, too.
Sadly (for Murphy, at least), he hadn’t done the situational analysis. We were home. It wasn’t hot. He could sit there all day if he wanted. Besides, it was Saturday and I was in no hurry. I shut the boot and leaned on the bonnet, chatted to passing neighbours, checked a few emails on my phone. Then I opened the boot. He got out.
After this, things chugged along for a week or so. The price of getting into the car was still extortionately high, but getting out was thrown in as part of a ‘two for one’ deal for loyal customers. Until the day came when Si was up and about and back on walking duties when normal service was, more or less, resumed. Not, I think, because of some magical, mystical dog whispery thing. But the joy of having Dad back in the game was reward enough for all car-related manoeuvres, at least in the short term. Nevertheless, it had been a wake-up call. We needed to reset the rules.
This being a dog training blog, not a ‘treat and tell’ memoir, we’d best move onto the advice. It basically comes down to this: positive reinforcement is just that. Food rewards are pay, not bribery or distraction. It's OK to use food as a lure in the first stages of teaching some behaviours. In fact, it’s the most efficient way of starting. But you then need to move pretty rapidly to a performance-based system of pay for productivity. And if you have a dog who is accustomed to playing the odds, like Murphy, your best bet may be to randomise when the big pay-out happens.
As far as Murphy and the economy-threatening treat inflation was concerned, we had to go back to established principles. We realised that when we adopted him, we’d never actually trained getting in the car because he just did. I‘d slipped into the habit of the constitutionally impatient, luring him with a treat when he was distracted or dawdled and delayed. But it was no big deal. Until it was.
We had to knuckle down and do the training: asking him to get in the car, rewarding, asking him to get out. We mixed it up, taking him for a short walk, then back in the car, out again and then off for another walk. Sometimes we gave small rewards for ‘in the car’ obedience and bonuses for snappy entries, but, for exits, no rewards other than thanks. And because he could never predict when walks were over or what would be on offer for 'in the car' compliance, even for a master grifter like Murphy, it was always worth putting in the behaviour in the hope of a good payday.
So here ends our cautionary tale. If you’ve edged into using treats as bribes, it's maybe time to do a training reboot. Otherwise, well, we hope your dog isn’t a loveable rogue like Murphy. Because they will always want to see the colour of your money before they give you anything. And it might just get very, very expensive.