25/05/2024
Some really great info here about rewarding behaviour, positive reinforcement and also accidentally reinforcing unwanted behaviours
Training Rewards: Are They Helping or Hurting?
Of course, we want to teach our dogs good manners and behaviour through rewards, particularly treats or food. Rewards aren’t just treats; rewards can be any food, praise, affection, toys, or anything your dog finds rewarding. Rewards are tricky, though, and being aware of the nuances is important. There are circumstances where rewards are harming your dog more than helping them. Let’s talk about that.
REWARDS ARE REINFORCEMENTS
It's obvious that the bigger reward your dog gets for any behaviour or action, the more they will do whatever behaviour they’re doing (or whatever state of mind they’re in).
There are various ways your dog can get a reward, including self-rewarding. Self-rewarding is when your dog gets rewarded for doing a behaviour from a natural drive, such as chasing a cat, jumping up at a stranger, or taking food off the side or plate. So some important points to remember are;
You will struggle to give a bigger reward or have better timing than the one a dog gets for the types of behaviour described above. You will find it hard to give bigger ‘consequences of actions’ than something for which they get a high-value reward. So it is very important not to allow your dog to do any behaviours you don't want them to do, such as jumping at work surfaces, running up to other dogs on walks or chasing pheasants in the woods. To reiterate, the more you allow your dog to do these things, the more ingrained the behaviour becomes.
To prevent these self-rewarding behaviours, it's important to interrupt or anticipate as quickly as possible, ensuring your dog understands what you want and redirecting them to the behaviour you are looking for. The other option is to manage your dog in any situation where unwanted behaviours might happen - there is no point recalling your 8-month-old puppy as it is tearing across the field when that command hasn’t been fully taught or reinforced, so you must keep them on a longline.
Affection is a reward, and all rewards tell your dog that you like that behaviour/state of mind and that you’d like to see more of it. It’s easy to accidentally reward your dog at the wrong moment and reinforce the wrong behaviour.
One of the most common mistakes is when visitors to the house or strangers out on walks encourage your super cute puppy to come bowling over for lots of excitable fuss and then wonder why our one-year-old 30-kg Labrador wants to throw themselves at every person they see. If only we were all as good at training and rewarding our dogs for everything as we are at getting them to react to someone coming in the door!
Other examples are;
Sweetly telling your dog “it’s okay” and/or petting them while they are afraid (this is a controversial and nuanced one. There is a difference between rewarding your dog in this manner and calmly supporting your dog through a scary moment)
Pulling your dog close to you and petting them while they are growling at strangers
Engaging with a dog who jumps up on you
In all of these cases, we’re not trying to reward bad behaviour, but unfortunately, that’s what’s happening! It’s important to understand that praise, affection, food, and toys will always be perceived as rewards by your dog, and therefore, it’s important to stay aware of when you give these rewards.
REWARDS AND STATE OF MIND
The other common scenario where rewards are more harmful than helpful is when they get a dog super amped up. This is frequently a concern during training. For example, if I work with a super ball-driven dog, the ball is his ultimate reward. But if I’m working with him on place command a ball is going to send his brain right into a state of arousal, which is the opposite of what we want during a calming exercise like duration.
This can also happen with food and affection. I love to use a dog’s daily kibble when I’m teaching them new skills, like how to sit, down, or place. However, some dogs get so excited by food that it becomes counterproductive to use food as a teaching reward. Their state of mind is so elevated that they’re not thinking and, therefore, not learning—they’re just frantically reacting to figure out how to get the food. Frantic is not a great state of mind for learning.
I’ll usually switch to affection as a reward for dogs super amped by food while teaching. Sometimes dogs can’t handle affection either, so if that also sends their state of mind through the roof, then I’ll just use praise—a soft, calm “Good girl!” is all it takes to send the message that yes, I like that behaviour. The lesson here is to pay attention to what specific rewards do to your dog’s state of mind. Are they helping your dog learn or sending your dog’s excitement levels through the roof? Stay aware, and choose wisely!
TIMING
The tricky part of positive reinforcement (and any) training is ensuring you are rewarding the right behaviour. Whatever you reward with, you have to do it fast—really fast. The only way the behaviour will become associated with the reward is if it is almost simultaneous—most trainers say you only have about 1 second to reward a dog effectively! Even in dog years, that’s not very long.
If you take too long to reward, your dog might associate the reward with a different behaviour entirely. A good example of this would be during a recall - you call your pup, ask it to sit in front of you and then fish around in your pocket for a treat - what are you rewarding for? Not the recall, not even the sit - but the wait while you have your hand in your pocket. This will also poison your recall - why would your pupster come rushing back if they will have just to sit and be bored?
Another is house training. Many people reward dogs who are house training when they come inside after they p*e or p**p, but that rewards the dog for coming inside, not for p*eing or p**ping outside. You think you’re house training, but your puppy might still think the world is their toilet—as long as they come when they’re called! To reward your potty-training dog for p*eing outside, you should go out with them and offer a treat immediately after they p*e. Once your dog is consistently going outside and not having accidents, you can begin to phase out the treats.
LURING
Dogs are visual learners and understand body language better than our human verbal one. So it’s important to time our language and physical cues carefully. When a dog is first learning a behaviour, you will find it easiest to start by using treats, your hand or a toy/ball to lure them into the position or place you want and using hand signals. Once your dog understands the verbal cue, ask for the behaviour before you offer any body language, such as a hand signal or a treat. Once they do the behaviour, then you can offer the treat.
We want to offer the verbal cue first before reaching for the treat because we don’t want the dog to associate the behaviour (like sitting or heeling) with seeing a human reaching for the treat pouch rather than the verbal cue. You want them to understand the verbal cue to replicate it if you don’t have treats available!
Remember that offering a verbal cue without a hand signal is only possible with commands your dog already knows. Set your foundation with physical cues, then phase them out as needed. Always offer a verbal cue at every stage of training since it’s the hardest part for your dog to learn.
Also, try not to lure your dog away from an unwanted behaviour, as you may inadvertently reward it!
If your dog isn’t demonstrating the behaviors you would like to see, even though you are trying to reward them—don’t worry, you don’t have a defective dog! No need to contact your doggy manufacturer, just practice your timing.
The foundation of operant conditioning is the association of behaviours with certain results, but that depends on the results (like a tasty treat or a game of tug) coming immediately after the behaviour. We also want to be mindful of the timing of our verbal cues to be sure that our dogs understand what we are trying to get them to do, and that we reward the right behaviours! Timing in dog training can be tricky, but with practice, your dog can learn whatever you want.
