Is It Better To Reward Before or After the Release Cue?
The most common obedience mistakes I see revolve not around the commands but around the release. Many clients have never heard of a release. ”Huh? Release? What?” Other times there is just a lack of clarity or a lack of appreciation for how important it is. So, what is a release and what’s the big deal?
In a nutshell, any command that has a duration must also have a release. For example: “Sit” means not only that the dog should put his butt on the ground but that he should keep it there until released. This is one of those precious gems of the dog training world, one of the very few things that has no debate among professional dog trainers….at least not that I know of. But, as any good dog training researcher can guess, the warm fuzzy amongst dog trainers stops pretty much right about there. Why? Because there is debate over whether to reward before or after the release. Have no fear friends, by the end of this article you will have no need to join the debate because you will have such a clear understanding of the topic that debate will feel like a waste of time. You will learn that much of the debate is actually due to confusion. You will find that both techniques are valid and the best approach depends upon what you are training for and what you are rewarding with. Are you training a Therapy Dog, a Protection Dog or a Pet Dog? Are you rewarding with affection, treats or a tennis ball? These are all crucial questions to ask in order to determine the best approach.
This article is a bit lengthy and gets into some slightly advanced learning theory. There is some redundancy for the sake of clarity but I think it is well worth the study for those interested in accurate dog training. The key to success is having communication that is so crystal clear that your dog will truly understand you and hence pay more attention to you.
A dog that pays attention learns better, faster and has more fun. Enjoy!
Reward and Release Primer: The Basics
Below is an introductory comparison of two reward based training styles and the relationship between Reward and Release.
Thriving Canine Basic Obedience Program. My beginning program uses what I’ve dubbed the CPR protocol which is an acronym for Command-Praise-Release. For example “Sit…Good Dog…Okay (release word).” In this program any rewards such as treats or petting go along with the Praise portion, hence coming BEFORE the Release. Pretty simple right? So that must mean that it’s best to reward before the Release, right? Problem solved, article done, let’s move on. Not so fast there, smarty pants! It just means that’s the way I start my basic training process because it’s very comprehensive for beginners. It’s also great for stabilizing behaviors should they decide to continue to advanced obedience where the dog is expected to Down-Stay for long durations at an outdoor cafe.
Other programs (like Clicker Training). There are also programs that give the reward AFTER being released via a Reward Marker (see step 1 below). Some of these are very intentional, like Clicker Training or Drive Building Programs but some are simply not ultra aware of the difference. This should not be ambiguous, there should be a definite intention behind your choice in a training process and the dog should have a clear sense that they must be released. In these programs it should go something like “Sit….Yes!” and then have the dog come and get the treat or toy. This coming-to-get-the-reward works great for building enthusiasm because it taps into the dog's natural prey drive. The potential down side is a dog that’s “poppy” or that tends to jump out of position or from position to position “throwing behaviors” at you trying to get a reward. Once the enthusiasm is built you will need to start stabilizing the behaviors. I prefer to introduce this style later on rather than in my initial basic obedience program, choosing to stabilize first and build drive later. Building enthusiasm first is good if you know you are training for high-drive purposes but most of my clients just want a well-behaved pet so we are often trying to keep the dog as mellow as possible, not get them all amped up. Agility training or controlled retrieves and tug games would be good times to use this style.
Details, Details, Details.
While dog training debates will probably go on as long and as passionately as political ones do, I believe the mystery of whether to reward before or after releasing can be solved with two important bits of knowledge:
- A basic understanding of Marker Training
- Defining, very specifically, the term Release
Once these two concepts are understood you will see that there is no right or wrong, per se. There are merely different ways of using reward-based training, both of which have benefits when used at the right time. Not only that, you will find that they actually compliment each other and work well in tandem! Isn’t that good news?
Understanding the terms and techniques outlined below allows for a ton of versatility in training a wide variety of dogs.
Step #1: Basic Understanding of Marker Training
A Marker is any word, sound or signal which "marks" or indicates the moment the dog has performed the behavior. There are many different markers, including those for unwanted behaviors, but for this article we will only focus on two of them- the Reward Marker and the Continuation Marker.
- Reward Marker: A Reward Marker (popularized as Clicker Training) is a word, sound or visible cue that “marks” the exact moment of a wanted behavior and releases the dog from the command to receive a reward. It says “Well done, now come get your reward.” It is a definite promise of a reward. Clicker Trainers use a “Click” from a small device. “Yes!” is also commonly used as a verbal Reward Marker. (I prefer to use my voice because it keeps my hands free, is always with me and is more personal.) The dog is still engaged with the handler but the particular exercise or sequence of exercises has been marked, ended and rewarded. He is released to get his reward but the session has not ended. The dog is not free to walk away, go pee on a bush, etc. Hence this is not a true Release Cue. This concept is generally introduced with food but really works great when rewarding with toys since the dog must be released from position to get the toy. Example: You say “Sit” and the dog sits. Next you “Click” or say “Yes!” present a toy and the dog jumps out of position to engage in a game of tug. Note: In my CPR (Command, Praise, Release) model a Reward Marker falls under the heading of Release but still requires a true Release to end the session. Example: “Sit”….”Yes!”…..give the dog a treat and as he’s looking for the next cue say “Okay” and end the session.
- Continuation Marker: (aka Praise) a Continuation Marker such as "Good Dog" says "I like what you are doing, don't stop, keep doing it." This can be used in tandem with treats or petting but requires a separate Release Cue. “Good Dog” does not mean it’s over. Therefore this is used when rewarding BEFORE the Release. It can still be used to "mark" the behavior by timing it with the exact moment the behavior was performed. Saying “Good Dog” precisely as the dog sits for example. While Praise is often combined with treats and petting it is not always or necessarily so and it is not a promise of reward as seen above with a Reward Marker.
Step #2: Defining the Term Release
There are actually three potential ways to release a dog from a particular command or position but only one true Release Cue. Let’s have a look.
- Release Cue: The Release is a cue that truly ends the exercise or series of exercises. "Okay" is commonly used but any word, sound or visual cue is fine. A true release signals that the dog is punched out, off duty and free to do whatever he likes….within reason of course. All normal rules of conduct and good behavior still apply. There is a natural or environmental reward that comes automatically in the form of freedom but handler-based-rewards and engagement are over at this point, until the dog's attention is called upon again. This could be a matter of seconds, hours or the rest of the day. My basic training program starts out with CPR or Command, Praise, Release…in that order. Example: “Sit….Good Dog….Okay” In this case any rewards coming from the handler, such as treats or affection, come during the Praise, which is BEFORE the Release.
- Chaining or Sequencing: This is when you cue the dog from one command to another. Going from Sit to Come for example. He is released from the original command but only to go to the next command, so it is not a true Release Cue.
- Reward Marker: (Please see Step #1 above for details) The Reward Marker is a form of release but it is a release with engagement, not a release to freedom. It is also a promise that the reward is coming. Never use a reward marker and not deliver the goods. Reward Markers are great when using toy rewards or when rewarding fleeting behaviors that really don’t have a duration.
So Which is Better, Rewarding Before or After?
So, here we are back to the original question. Is it better to reward before or after the Release? Before getting into which is best, let’s recap and make sure we understand that we now have two different cues for releasing the dog, the Reward Marker and the Release. One releases to engagement, the other to freedom.
So, the answer is….it depends. I actually find huge value in both, depending on what I am training and what I am using as a reward. I find that rewarding a dog BEFORE releasing him is best for stabilizing behaviors such as Stay. On the other hand, I find that rewarding AFTER releasing him and is better for building enthusiasm and excitement in the dog and playing with the dogs prey drive by rewarding with toys, rough housing or excited praise.
The only rewards that should come after the actual Release Cue are environmental rewards such as going to play with another dog, checking pee-mail, etc. Rewards that the dog must be released to get but require engagement, such as tennis balls and tug toys come after the Reward Marker. Technically, sure you could use the Release Cue before throwing the ball but technically the dog would also be within his rights to go pee on a bush instead of getting the ball……get it? This is why we have a Reward Marker which allows us to use active, high-drive rewards while still maintaining engagement and without clouding the meaning of our Release Cue.
This may seem like splitting hairs but it is this hair splitting clarity that will make your training the best it can be. I find that humans have a difficult time grasping these concepts at first but dogs working with a trained handler pick it up in a flash.
To make this as clear as I can, without actually being able to demonstrate, let’s look at some real world training examples.
Example 1: Best Rewarded BEFORE releasing. The Long Down.
Let's say I am working on my dogs Down-Stay. I do not want him to anticipate his release, I want him to grow roots in the ground because he will be there for a while. I will reward him BEFORE the Release because staying put is the hard part and the Release is pretty much a reward in itself since he can go sniff, run around, etc. Also, as this exercise advances he may be doing hour long Down-Stays and I want to be able to let him know that I have not forgotten about him and appreciate what he is doing. I may do this with a verbal "Good Dog", petting or treats but none of those signify the Release. When the exercise is over I release him with "Okay" and he is free to do what he likes.
Example 2: Best Rewarded AFTER releasing. The Fast Down.
Let's say I want to build my dog's speed and enthusiasm on the Down command. I am going to work with his prey drive by rewarding AFTER releasing and making him chase his reward. As soon as his belly hits the ground I say "Yes!" and have him come take the treat from my hand which is luring him to chase it. Now he is getting excited and starts dropping into the down position quicker and more enthusiastically. I then transfer this to a retrieve or tug as a reward.
Example 3: Combos. Rewarding Before AND After the Release.
Ok, so in the above training examples I used "Good Dog" as a Continuation Marker to basically say to the dog "I like what you are doing, keep doing it" and I used "Yes!" as a Reward Marker to essentially say "Job well done, now come and get your reward." I also used “Okay” as a Release which means "all done, go ahead and do what you like". So "Yes" and "Okay" both release the dog from his command but the important distinction is that "Yes" is a release with engagement and "Okay" is a release to freedom. If I said "Yes" and he went to go pee on a tree, that would be a mistake. If I said "Okay" then he is free to pee on the tree. Once the difference between a Reward Marker and a Release Cue is understood there is no need to choose one technique over the other since they can actually be used in combo. For instance: I have my dog Sit. I say "Good Dog" and give a little pet on the head. He is still sitting. I walk away a few yards and turn back to him. "Good Dog," I say again and return to my dog that is still sitting. I then pull out a piece of rope with a "Yes!" and we play tug. I have him let go of the toy and Sit again. "Good Dog" I say and give him a small piece of meat, he is still sitting. I release him with "Okay" and we stroll leisurely around the park.
In Closing
So as you can see, there is no cut and dried answer whether it is better to reward before or after the release. The real question is which is better for any particular training scenario. The key is having an ultra-clear, super-specific communication system in place and understanding that there is a separate cue for rewarding while still under command, another for releasing to get a reward and yet another for releasing to freedom.
Chad Culp–Certified Dog Trainer, Behavior Consultant, Certified Holistic Chef for Animals
Copyright 2005-2013 Chad Culp, Thriving Canine. All rights reserved. Chad@ThrivingCanine.com
