Rewards #3
What tools do you have to reward your dog? Do you have the ability to suddenly bring out a playful interaction? To get the dog to relax? To get it to engage and tune in with you?
Verbal Signals
- Ready? / All Done: The school bells that tells the dog when the training exercise or play time starts and ends. This added clarity prepares your dog for what’s coming up, and lets it know when it is all over. Knowledge and understanding of these cues keeps the dog tuned in with you until you give a release cue (e.g. dog stays with you after it has been recalled and received a reward until you say “go play”). It also creates clear expectations which help maintain motivation by avoiding disappointment.
- Yes: Marks the moment the dog offered the behaviour. Can mean “reward is coming” or ”Go get the reward”, often followed by a toy or food reward.
- Good: Praise to keep the dog going. A calm “good” for extra feedback during a long down stay, or “good” + praise after the dog responds to Look to encourage the dog to keep giving eye contact.
- OK: Release cue towards a reward at a distance or the environment, like greeting a dog or person, smelling a tree, continuing exploring on a walk...
Commands: Fading cues and Generalisation
Our signals for a certain behaviour is often much more than a verbal cue. Your signal for the dog to sit might be having the dog in front of you, being in your kitchen, crouching down towards the dog, looking it in the eyes, holding food in front of you, and saying the word “sit”, maybe even more than once. If you want the dog to perform a behaviour in any situation and first time you say the word, you need to gradually phase out these other cues (your body posture, where you are in relation to the dog, what the environment looks like).
Another unfortunate combination of cues is a recall where the owner calling the dog, while reaching out a hand full of food towards the dog, or while shaking a toy they are holding. This often result in the dog pausing, looking at the owner to consider what the reward might be that time, evaluating the situation, and then deciding that the food presented that day wasn’t interesting enough. To avoid this problem, only ever present the reward after your dog has responded to the command. What do you want to reinforce? For the recall, it would be the dog running towards you with good speed (or if working on smaller steps, the moment the dog disengages with a distraction, or checks in with you). Work on delaying the presentation of the reward, and building yourself into the reward event by recalling the dog, having it come in close, pet the dog and praise it, then bring out the food, let it nibble on it while you continue petting. Stop petting as you remove the rest of the food (don’t pull back an empty hand), make sure the dog stays engaged with you for a bit (expecting more!). Surprise the dog! Offer another reward for staying with you, start a game of play, or release the dog with “OK” or “Go Play” to end the interaction. More about social interactions and rewards in the next blog post (sign up at the bottom of the page to get an email when its out)!
Reward deliveries
What do you want to achieve? What influence do you want it to have on the dog?
Do you want the dog’s activity level to increase or decrease with the reward? Do you want your dog to change its focus away or towards something? Do you want the dog to rest or explore independently, or become more oriented towards you?
- Placement: Where is the reward delivered? If you are teaching a dog to settle on a dog bed or blanket, it is more desirable to drop the treat on the bed than delivering from your hand. This changes the dog’s focus from you to the place where it is resting. If you are teaching the dog to ignore a distraction, deliver the reward away from the distraction to create a bias in the dog’s focus and expectation over time. If you are trying to get the dog more comfortable around a certain trigger, deliver it towards the trigger if the dog is looking in a relaxed way, or away from the trigger if you want to emphasise the dog’s escape route. Reward placement creates an expectation which determines where the dog will direct its energy in the future. If you want to practice fast response to a command (like Down or Sit), toss the food away from the dog to get it to leave the position so that you get more repetitions. If you want to add duration to a command (a long down or Sit stay), bring the food to the dog.
- Delivery: How do you deliver the reward? Do you toss it in the air for the dog to catch? Or do you casually drop it on the ground close to the dog? Try to transfer your emotion (calmness and relaxation or playfulness and engagement) to the dog through the way you deliver the reward. What are your movements like, how are you breathing, what does your voice sound like?
For example, when rewarding a reactive dog with increased distance to the trigger, adding a food reward can help you encourage a specific emotional response in the dog. For a nervous dog that has a tendency to become withdrawn when stressed, you can throw a treat to encourage it to become more playful and outgoing through the chase. For an overexcited or anxious dog, you can drop a few treats on the ground for a mini-scent work to help the dog settle and centre itself. For a dog that quickly becomes absorbed by the environment, let the dog get the food directly from the owner, chasing it in the hand if it is a little withdrawn, or sitting down together and nibbling on the food if it has a tendency to become overstimulated.
- Timing: When do you deliver the reward? Anyone knows that timing is important in dog training. You have to mark the exact moment the dog offers the behaviour, produce a reward relatively quickly and yada yada… This is pretty straight forward when you are teaching the dog basic commands like sit and down, but requires a more thought through plan when working on more complex behaviours or emotional problems. If you are teaching the dog to settle, reward when the dog lies down, flip its hip to the side, puts his/her head down, sighs… (When you reward determines -> what behaviour you reward.) If you have a very engaged dog and you want it to be more comfortable just resting outside, reward at a low frequency, and when the dog is paying attention to the surrounding rather than staring at you.
- The reward: What reward do you deliver? I am a big believer in getting the dog to work for its regular kibble. Using high value rewards can act as a distraction to the dog, and give a false sense of security when working on reactivity issues (if the reward is suddenly absent, the dog is over threshold). The food reward should be as a gold star sticker on the homework – a quiet nice thing to get, but with no real value. It is just an add-on to the social interaction or game which should always remain the primary motivator and focus. If your dog is not motivated by its regular food, there are many ways to build it up by controlling when and where the dog is fed, adding a scent to the food (like crumbled dehydrated tripe or lung) and slowly phasing it out, as well as how you deliver the reward (the value of a food reward can increase by for example simply adding movement if this is an element your dog finds reinforcing). I only use toy rewards inside the house and isolated training situations on walks, once the dog has mastered emotional regulation, and is able to quickly and fluidly switch between high and low arousal states. If introduced earlier, I find that the added arousal can either cause the dog to be more impulsive towards the environment (when lacking engagement and obedience), or give a false sense of security (dog is engaged but simply distracted and forgets about the triggers until the reward is removed).
- Frequency: How often should you reward? The fancy name for this is reinforcement schedule (rs). I’ll only touch on this briefly here as I’m planning to make another blog post on the subject. When teaching a dog a new behaviour, you first use a continuous rs that is, the dog is rewarded for every response, and with time you move onto a fixed rs, where the dog is rewarded after a certain number of responses, or variable/random rs, where a variable number of responses are required for the dog to be rewarded.
I have a bunch of videos on most things covered in this post, send me a message if you want the list to get some more ideas. :) Hope you enjoyed the post, have fun training!
