By Heather Pyne
I recently found myself in the position where I needed to explain why we use food rewards in our training. If you’ve trained with Pawsitive Squad you will already know that we are a positive reinforcement training company because:
a) It is categorically the most welfare friendly way to train a dog. Fear and punishment frankly have no place in a working relationship between handler and dog.
b) It’s an evidence-based practice unlike dominance/alpha training which is based on debunked research and outmoded principles.
So, for training using positive reinforcement you need a reinforcer, something positive to your dog which makes them more likely to repeat the behaviour. For humans MONEY is a big reinforcer as is feelings of doing good, receiving gratitude, stars and stickers for younger ones and general praise. For dogs we use 4 main types of reinforcer at Pawsitive Squad – food (treats), toys (usually a tennis ball or tuggy), tactile (stroking, petting) and verbal (saying “good dog!).
Unlike most organisations who use positive reinforcement we aim not to have the reinforcer in sight because that can make the dog less likely to be focused on the behaviour and more focused on the reinforcer and be coercive so we don’t do things like luring with treats. BUT if we want our dogs to work for us, they have to get ‘paid’ (rewarded, behaviour positively reinforced) and very quickly after they do something positive so they make the association between the behaviour and the reinforcer (you can use clickers to tide over somethings but this comes with its own issues in the assistance dog world). This is why often treats are the reinforcer of choice when we are training our dogs in working environments.
Let me give you a scenario to help explain…
You’re in a busy shop, it’s crowded and noisy. Your dog spots that someone has dropped some crisps on the floor – YUM! You stand still and your dog CHOOSES to look back at you and away from the crisps. At this moment you need to reinforce that leave behaviour but how? You give verbal praise but that’s a low value reinforcer in such a high distraction situation and frankly your dog would rather eat those crisps than have your ‘good dog’ praise, not that they can really hear you over all the background noise anyway. Your dog finds tug toys highly rewarding but you can’t give a tug toy in the middle of a busy shop, that’s not appropriate. You can stroke your dog but again this is low value to most dogs when they’re in a high distraction environment and when they are doing something which is taking masses of impulse control. So, you offer your verbal praise and stroke and move your dog away and your dog thinks “leaving those yummy crisps was not a beneficial thing for me to do”. So next time you go into a shop your pup spots something on the floor and they lunge and snatch it “yum yum yum” Now eating food dropped on the floor is reinforced and the dog only tried to get it because you hadn’t taught them how amazingly valuable leaving dropped food is to them.
Now, what is in that initial scenario when your dog chose to leave the crisps, you’d have given your dog a small bit of chicken? Or similar high value treat? Well then your dog is thinking “wow! Leaving stuff which is on the floor is so rewarding for me!” Next time they see dropped food they look up at you enthusiastically leaving because they know that you want them to leave the dropped item AND that leaving it is rewarding for them. Doing this doesn’t mean you are always giving treats because we quickly move into a jackpot reward system AND we are only using treats when they are the most appropriate reinforcer for the dog and the environment which means you have a reliably trained dog who wants to learn and is eager to please.
ANOTHER reason we often do use treats in training is because of something called contra-freeloading which is an observed behaviour which is reliably seen in dogs (and other animals!) supported by scientific evidence. Contra-freeloading is where an animal would rather work for food than be given food without having to work for it. Meaning that your dog would CHOOSE to get food out of a food puzzle even if a bowl of food is present and easily accessible. With this in mind we like to use food rewards because it is what our dogs want! They don't want boring food bowls and easy scoffing, they want to earn their food so using food as a reinforcer is working more with their natural behaviour & drive.
Now, when we are doing our home practising and practising things like recall, we often use other reinforcers especially play (tuggies & balls especially) to make training and working valuable for our dogs. For assistance dogs and the complex, busy and ‘non dog play friendly’ spaces they’re often in we often need food rewards to occasionally reinforce the desirable behaviour and that is the right thing to do because we are using an evidence based, welfare friendly (ethical) approach to training.