Here is yet another post in our series on rewards! This one is on how to start to fade out rewards and common problems which makes the process harder. If you have attempted to stop rewarding your dog for behaviours they already know without much success, or if your dog stops listening as soon as there are no rewards present - this one is for you. If you haven’t read our previous posts, here is a recap from the last one:
“Frequency: How often should you reward? The fancy name for this is reinforcement schedule (RS). 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.”
Problem 1: Presence of food is part of the cue for the behaviour. Solution 1: Be mindful of your posture, where you have your hands, and when you present the reward when you are training. This was discussed in more detail in the post Rewards #3.
Problem 2: Using a fixed RS can make some dogs, predicting your reward pattern, drop in motivation. Solution 2: Keep the RS unpredictable when you start to fade out food rewards (think gambling!). Sometime present one piece of food, sometimes just pet the dog and praise it, and occasionally initiate a reward event, presenting several pieces in a row.
Problem 3: Using “high value” food. Solution 3: If your dog is only performing and accepting food like chicken, cheese or hot dogs, you can quickly fall down a slope of using the food for bribing instead of teaching if you are not very mindful of how it is done. I have discussed this in the previous rewards post as well. We always recommend using the dog’s regular kibble for training. Not only does it make it so much easier to calculate the dog’s nutritional intake, but ensures that you are making an effort to engage your dog in the training. If you want to read more on the topic, have a look at our blog post Food in Training.
For example, our dog Tennessee has no natural food drive. When we got her she would spit out or leave any kind of treat, and barely showed interest in the raw meat that the dogs get for their dinner. If I would just have increased the value of the food without increasing the quality of the interaction, her attitude to training would have stayed the same - little interest, happier to walk away than to engage. Even though she gets fed raw from her bowl (without having to make an effort for it), I insisted on only rewarding her with dry food (lower value) so that we would not take short cuts which would be detrimental later on. I never withheld food or kept her hungry, but instead focused all my attention on HOW the rewards were delivered. She is thus primarily working for the social interaction she gets out of being engaged and responsive, rather than the food in itself. More on this below!
Problem 4: You have not built yourself into the reward events. Your dog is only working for the food, not for you. Solution 4: The focus of all reward deliveries should be the social interaction which take place around an object - regardless of whether that object is a toy, food, or a piece of moss you found in the woods. This can be petting and praising the dog while rewarding it or engaging in a playful social interaction around the food.
Problem 5: You have not been mindful of what behaviours you reinforce in the dog’s daily life, or intentionally used functional rewards. Solution 5: We have discussed this extensively in our previous posts Rewards #1 and Rewards #2. A common example of this missed opportunity is the rewards that follow a recall. Dog comes back, gets food reward, and then runs off before the owner has had an opportunity to pet or put a leash on the dog. One obvious cause of this is only recalling the dog to end the off leash time and put it back on a leash (not reinforcing), but that is outside of the focus of this blog post. A more intentional way would be to recall the dog, reward it with food, block it by stepping on the long line if it runs off, wait for it to seek contact with you again, and then reward it by releasing it to the environment. If you have built in the Premack principle from the start in your training, keeping functional rewards in mind, you rarely run into issues when fading food.
Problem 6: Dog has poor understanding of how to obtain the reward. Solution: This is more common in training situations aimed at a specific activity, such as trick training or play time, where the sessions are short and you require a high level of engagement from the dog throughout. When the reward is removed or the handler becomes passive, the dog gets frustrated or looses interest instead of increasing in engagement. Working on smoothing out switches between work and play can help this. Teach the dog that it has the power to activate the handler, and get the game to start again, by quickly starting to work after the reward is removed. I have some basic videos on the early steps of this, including mistakes, here:
Problem 7: Your dog listens well without rewards in calm places, but does not listen once you move into busy environments. Solution 7: You may be moving forward too quickly. Have a look at our blogpost Disobendient? to help you identify what step you are getting stuck on. For now, go back to rewarding your dog in those places until you have formed a reliable habit.
I hope you found this overview helpful to help you troubleshoot what step of fading rewards you have been struggling with. Let me know if you have any questions and happy training! :)
Cover image by Free Range Dogs.