Reward Based Training: Understanding Intermittent Reinforcement

Ok, we are going to get a little technical as we dive into the next piece in our series on Reward Based Training (RBT)! “Intermittent Reinforcement” is a term coined by American psychologist and behaviorist, B. F. Skinner, in his studies on Operant Conditioning and Behaviorism. The shortest definition of intermittent reinforcement is that the reward (reinforcement) is only given part of the time that the subject gives the desired response. Although we will shortly get to how this applies with dog training, let’s give a human example to help explain this concept in layman’s terms.

People find gambling to be an addicting behavior. Why? It’s because of “intermittent reinforcement!” Let’s use slot machines to demonstrate this. If people knew that they would put $1 into a slot machine and receive 75 cents back every time, this game would quickly cause people to lose interest. However, because you put in the $1 knowing that most of the time you will not get anything back, sometimes you might make your $1 back, and rarely you will hit a jackpot, this makes the game exciting, fun, and something that you want to do over and over! This is “intermittent reinforcement” in action.

In terms of dog training, intermittent reinforcement is very important for RBT. Many who have not used and do not fully understand RBT worry that using food will cause a dog to become dependent on the food and that the dog will not work unless food is present. This is true if you reward every behavior and don’t properly move to intermittent rewarding in a timely fashion. With RBT, we use a “continuous reinforcement schedule” when first teaching a behavior, meaning we will reward for every correct response. However, once the dog has learned a behavior and the cue, we move to an “intermittent reinforcement schedule.”

What’s important is making sure you use a “variable ratio reinforcement schedule” when using intermittent reinforcement. The word “ratio” means that you have an average number of times you will reward a behavior – for example, 5 out of 20 correct responses. The term “variable” means that the reinforcement will come at unpredictable times. So, if asking for 20 “sits,” with the goal of rewarding 5 times, you will give at more random intervals. Instead of giving on the 5^th, 10^th, 15^th, and 20^th “sits,” maybe you will give on the 3^rd, 10^th, 16^th, and 19^th “sits.” If you reward on the same 5^th “sit,” every time, your dog will quickly pick up the pattern and the dog will know when to expect a reward, moving back to almost a “continuous reinforcement schedule.”

The key is to keep training like a “game” for your pup…make it as exciting as those slot machines are for humans! Next time, we will discuss how to go from “continuous reinforcement” to “intermittent reinforcement” for your pups, with real examples, common questions, etc!

Reward Based Training: Luring vs. Rewarding – REWARDING

Last time, we dissected the concept of “luring.” Once your pup is showing he is ready to move on from luring, its time to switch the way you are giving treats by “rewarding” the behaviors.  

How do I move from luring to rewarding?

Okay, so your puppy is quick to move into position when you give the verbal cue and use a lure! Great! Your pup is now ready to move to a reward instead of a lure! There are several different ways to move your pup from a lure to rewards.  Please keep the 3Ds in mind when starting to move off a lure. Do not make any distraction, distance, or duration too difficult when first asking your pup for the behaviors without a lure.

• Cold turkey! Simply leave your treat in your pouch and give the verbal AND hand signal cues. When your pup executes the behavior, use the “yes” marker word, reach for the treat in your pouch, give treat, praise, release dog.
• Warm up your pup! Sometimes a pup does best when you use a lure a few times and then switch to a reward only. Ask for the behavior 2-3 times with a lure and then ask for behavior without the lure, right away. Gradually move away from having to do a few “warm up” lures for the behavior.

How do I use rewards for continuing my pup’s training?

Once your pup has moved to rewards and is no longer being lured for a specific behavior, you will want to use rewards quite often until your puppy has started to generalize the behavior. You always want to use praise, even once you have become more intermittent with rewarding a behavior! We will discuss how to move to more intermittent rewards in an upcoming Blue Cape Journal entry. Here are some tips for using rewards for training.

• Keep treats in a pouch or pocket. Treats need to be easily accessible. This means if you are wearing pants that have tight pockets, you may need to consider a treat pouch.
• Do NOT hold treats in your hand! Every dog knows when you are holding a treat in your hand, especially because you have been luring to teach them the behavior. If you have to hold a treat in your hand, your dog is still being lured for the behavior. This is a really important one for when we move to very rare rewarding as the dog gets older.
• Use small treats or take from your pup’s meals. In order to not cause your pup to gain unnecessary weight, pull most of your treats from your pup’s meals. However, you might need to use “high value” rewards in very distracting situations (such as puppy meetings). If high value rewards are needed, try to use something soft and smelly, made in the USA or Canada. Soft treats are easy to break into very small pieces, allowing you to get more use out of the treats you are using.

Here is a video of my pup Walton doing sits and downs for rewards (no lures). Notice that I’m still using the “yes” marker word. Also, Walton isn’t always quick on some of his commands and you’ll notice that I patiently wait for a few seconds instead of rapidly repeating the command.

Rewards from Southeastern Guide Dogs on Vimeo.

By pairing the “yes” marker word techniques, luring and rewarding, and keeping the 3Ds in check, your pup is on his way to a great start!

Reward Based Training: Luring vs. Rewarding – LURING

Reward Based Training (RBT) is very effective when the user has a good understanding of the steps to building a solid foundation for commands and behaviors. One of the basic concepts of RBT is knowing the difference between luring and rewarding and when to use each one. Many times, raisers easily get these two training concepts mixed up and/or use the wrong one at the wrong time. This Blue Cape Journal entry, we will be focusing on LURING.

What is the difference between luring and rewarding?

“Luring” is the act of attracting the dog with something like a food reward into the desired position and then giving the dog the food. An appropriate time to use luring in your pup’s training is when your dog is just beginning to learn a specific behavior. By incorporating the “yes” marker word with the act of luring, your puppy will quickly pick up new behaviors, leaving you to fade out the lure and continue to just use the hand and verbal signals.

“Rewarding” is when you give the food reward AFTER the pup has performed the learned behavior when using just your hand and verbal signals. When at the “reward” stage of training, you will not hold treats in your hand as you ask your dog to perform behaviors. Instead, your treats will be in a treat pouch, pocket, or elsewhere, and you will reach for the treat after giving the “yes” marker word.

How do I effectively use luring to begin training a behavior?

Something to keep in mind when using a treat as a lure is that your hand/treat will act as a magnet that will draw your puppy to wherever you need to lead him. You can try this by putting a treat in your hand, showing it at pup’s nose level, and walking around a chair. Your pup will enthusiastically follow your hand. This is fun for your puppy, especially if you release the treat into their mouth! By using this natural drive to manipulate your pup’s body, you can begin to train ANY behavior needed for our program. Here are some tips to using luring for training.

1. Make sure you have a clear idea of how to move your lure to achieve the desired behavior. If you are unsure how to lure a “sit,” for example, then you should not try figuring it out while the puppy is there and trying to follow your hand as you make different movements to get comfortable with the lure movements. Instead, practice without pup, then bring pup out. If you are unsure how to do a specific movement, please ask your AC at your next meeting.
2. Your pup needs to know that a lure is present. If you have a small pup and are just beginning to teach a behavior, make sure you put the treat directly in front of pup’s nose and go slow with your movements. If you go too fast, the pup is likely to lose the lure and you will not be successful. Keep your hand at the pup’s nose level and don’t go too high with your hand. If you go too high, you will find that your pup starts jumping up to try and follow the lure.
3. Use the “YES” marker word protocol. With lure in hand, ask for behavior, move lure to get pup to move into position, say “yes” as pup hits the position, treat, praise, release.
4. While holding the lure, keep your hand as close to the prescribed hand signal as possible. If a flat palm up is for “sit,” then don’t hold the lure with a fist as this does not look like what will be your hand signal later. Instead hold your palm flat and facing up, but hold the lure between your thumb and index finger. This will help your pup transition off of the lure more quickly and will keep your body language more clear for the pup.
5. Only lure for a very short time. The point of a lure is to teach the dog how to move his body in the direction you want and to gain some muscle memory when hearing your verbal signal and seeing your hand signal. There is not set amount of times you should lure a behavior, but as soon as your pup is easily moving into position, you should consider getting rid of the lure. For some behaviors this might only be a handful of times a lure is needed. For other behaviors, you might have to lure for a good week.
6. Lure the behavior in many locations. Dogs need our help to generalize behaviors. If I ask for a “sit” at home, I still want to easily ask my pup for a “sit” while on a walk or at a store. While in the luring phase of a behavior, you should make a real effort to lure the behavior a few times with every location change for your pup. This will teach the pup that the hand and verbal signals mean the same thing, despite your environment.

Luring Puppy from Southeastern Guide Dogs on Vimeo.

Luring Sits and Downs with Young Puppy from Puppy Kennel

Our next Blue Cape Journal entry will discuss how to move from lure to rewards and tips for rewarding your pup. Stay tuned!

Reward Based Training: Using the "Yes" Marker Word

This month launches our series on Reward Based Training! We will be addressing the importance of using Reward Based Training (RBT), how to use it correctly, breaking down different aspects of the training, and troubleshooting.

To have a good basis for RBT, raisers need to have an understanding on how to use the “yes” marker word with their pups. A “marker word” is a simple word meant to mark the precise moment of a dog’s behavior or choice. At Southeastern Guide Dogs, we use the word “yes” as our marker word. The marker word indicates that a reward is coming. If you are familiar with clicker training with animals, the marker word is identical except we are using a word in place of the clicker sound. Think of the word “yes” as though you are taking a camera snapshot of the moment the behavior occurs!

When using the marker word, there are very specific rules for training your puppy. If you fail to follow these rules, this can break down the meaning and effectiveness of the word “yes.”

Rules for Training with “YES” Marker Word

1. If you say, “yes” as a marker word, you MUST give a treat reward – even if your timing was off.  In order to keep meaning to the word, a treat must follow every single time the word is used. If you use the word, but fail to give a treat reward, the dog will start to not trust that the word equals a reward and it will water down the meaning. This is almost a “Boy Who Cried Wolf” type of scenario. If you ask your dog for a “sit,” but he does a “down” and you make a mistake and say “yes,” you still need to give a treat reward…and be more careful of your timing the next go around!
2. “Yes” is NEVER used as a praise or reward word. This is a very easy habit to get into and difficult to change the way you use it, once you are in this habit. Because of this, try not to be too hard on yourself if you accidentally use it as a praise word. However, don’t forget that every time you use the word “yes,” a treat MUST follow!
3. ALWAYS use the marker word in this order: “Yes!” --> Treat --> Praise. Do not praise BEFORE giving the treat reward. Again, this helps the word “yes” keep its meaning. It also prevents any praise words (good dog, nice job, etc.) from becoming the marker word.

Using “Yes” to Advance Skills

Once your pup is solid with using “yes” for basic behaviors, then you can use the marker word to your advantage to help advance your pup’s skills. First, once a pup has learned a skill, the raiser can begin to fade the use of the “yes” marker word. To continue to help your pup generalize, you will still be able to use treat rewards and praise, despite no longer using the marker word.  Giving the treat reward intermittently or fading it altogether can happen gradually, in this phase. However you want to make sure that you always continue giving praise! As a side note, we will be discussing this more in-depth during our Blue Cape Journal series on Reward Based Training.

We can also help our pups advance their skills by using “yes” to mark duration. Once a pup has learned a basic behavior or command, extend the amount of time before saying “yes!” As the pup gets better with a short time/duration, extend further and further before saying “yes.” You can use quiet praise during the duration time to let your pup know he is doing the correct thing, but do NOT use the word “yes” during the duration time. Don’t forget to tell your pup to “break” at the end of the duration.

As an example, here are the steps to teaching a pup to “sit,” without and with duration.

First, we have our pup learning to “sit” and with no duration.

Next, we have our pup who knows how to “sit,” but is advanced and learning to build duration into the “sit” command.

We are aware that many raisers have been using the “yes” marker word for some time, but that some may have been using it differently than explained in this post. It does take time to retrain our brains, so please know that your AC, group leaders, and PRS staff look forward to assisting you with building new habits through coaching, group activities, reminders, etc. There are some steps you can take to help set yourself up for success – making sure to carry treats on you (preferably in an easy-to-use treat pouch), beginning to consciously practice the “yes” marker word with your puppy now despite his age, and practicing any exercises that your ACs teach in regards to the marker word during meetings. By helping your pup understand the meaning of the marker word through adequate timing and consistent use, you can help set your pup up for success for continuing his life long learning, especially in the beginning phases of guide dog training!