Reinforcers for Agility Candi Moon
Why use different kinds of reinforcement?
When training your dog in agility, or in any discipline for that matter, it helps to have a wide variety of reinforcers at your disposal.
What exactly is a reinforcer? Quite simply anything that your dog likes and will work for, from a piece of roast chicken to a walk, game of tug or chance to go and play or sniff around. The emphasis here is on 'that your dog likes'. If your dog doesn't like it, its not a reinforcer.
Its important to keep in mind too that what your dog really wants can vary depending on your dog's mood, how hungry or tired she is, where she happens to be, the time of day and her emotional state.
When your dog is always rewarded with the same treat he can get bored. No matter how much you love say, bacon and eggs, you'd soon get tired of it if you got it every day and for every meal.
The more unpredictable you are, the more interesting you are to your dog. Many different ways of reinforcing your dog ensures that she stays excited about playing the training game with you, improves your relationship with your dog and, as Susan Garrett says, helps you 'become your dog's cookie'.
Different rewards also help you to differentiate, for your dog, how much you liked the last performance he gave you. For example, if you run your dog through the tunnel twice and the first time he strolls through and the second time he runs through with speed and enthusiasm, you know which if those two you'd rather have, right? But if he gets the same piece of polony both times, how is he going to tell?
One way of getting around this problem would be to withhold the treat for slower work and reward for faster. You could also jackpot, give several treats, for a really great performance. The other alternative would be to have different kinds of rewards. One piece of kibble for the stroll through the tunnel and a piece of biltong for the enthusiastic run.
Using food is usually a highly effective reinforcer for which most dogs will work. Food is also great for training precision, but its not always so great for training speed, and can lead to sniffing around and loss of focus on the handler when used as lures on the ground. A dog who is full, or nervous, may not want to work for treats, and treats can run out.
Using toys is great for getting drive and speed, not always so great for precision. Interactive games, such as tug, build great focus on the handler and can help relax a nervous dog, but may not work so well if the dog is low on energy.
It makes sense then to train your dog to work for a wide variety of reinforcers that will keep her guessing and will mean you can use the best reinforcer for the task, day and your dog's mood each time.
Categories of reinforcers
Reinforcers can be roughly divided into three main categories:
Food reinforcers
Food reinforcers could be any collection of food treats that your dog likes, which you could divided into categories depending on your dog's preferences. For my dogs a general food reinforcement hierarchy would be:
Lowest level reinforcers: dog kibble, cat kibble, dog biscuits
Medium level reinforcers: polony, Beeno Flatties treats, viennas
Highest level reinforcers: biltong, chicken, liver cake, cheese
My whippet mix Freya will stand on her head for cheese, so for her, that's her top reward. My Irish terrier mix Xandria is not too crazy about cheese, for her, its only a medium level reinforcer, but she will do cartwheels for biltong. My border collie x basenji Ozzy loves viennas so much that for him they're a top level reinforcer.
One of my clients has a Rotweiler who is crazy about carrots and many dogs enjoy little pieces of apple. Some dogs have a sweet tooth and others don't.
Knowing which food rewards to use means you need to experiment and try different things with your dog, judge by her reactions what's not so exciting and what really floats her boat and create a hierarchy for your dog.
Interaction reinforcers
These are interactions with you that you can use to reward your dog.
Some interaction reinforcers I use with my dogs are belly rubs, TTouch ear slides, butt scritches (scratching the spot on the back just above the tail) and being allowed to jump up on me. Most dogs also love it if you run away and let them chase you.
My dogs have learnt that the cue 'hug' means that they are allowed to jump up. If they jump up without receiving this cue they are ignored.
Again, interaction rewards can be divided into categories, depending on what your dog likes best. For my Labrador Hagan a belly rub would be his lowest level interaction reward, ear slides are better, butt scritches better still and being allowed to jump up is just the best, especially if I also make silly (embarrassing) noises! For my border collie Faolan, the hierarchy is the same, except butt scritches don't feature, for him, those are an aversive, he doesn't like them at all.
A little experimentation with your dog will help you figure out what works for them.
Game / toy reinforcers
These are rewards that involve playing with your dog with a toy. Examples would be fetch with a tennis ball or frisbee, fetch or tug with a soft squeaky toy or tug with a rope toy.
You can also be creative in terms of 'toys'. Clients of mine have a bulldog who enjoys to play with a mop. Hagan likes to play with brooms, empty food dishes and road cones.
With Hagan fetch with a frisbee is fine, a tennis ball is better, squeaky toys are great but nothing beats tug with his giant rope toy.
My other dogs far prefer squeaky toys to tug ropes.
Here you need to figure out what kind of toys your dog likes best and also how they like to play with them best.
For agility, having a dog who loves to play tug is invaluable as your dog needs you to play too in order for the game to be any fun. That way you build value for your dog for paying attention to you and interacting with you. Its also more interactive with you than fetch.
What if your dog is not interested in toys, or is not always interested in toys? For a food motivated dog, teaching them to tug on a tug toy that has food inside it works well. A sock with food inside it is one option or you can buy a specially designed tug toy with a compartment for food. If you're using a sock be careful to keep a good hold on it, if your dog is e.g. a voracious Labrador, they may ingest the food, sock and all!
Susan Garrett has a great article on her website 'How to create a motivating toy'. You can find her website here: http://www.clickerdogs.com
Mixing reinforcement hierarchies
You'll also want to consider how your different types of reinforcers compare to each other across the board. For example, for Hagan again, all food treats are better than interaction rewards most of the time, although jumping up is definitely as good as lower level food rewards. Tug, on most days, is just as much fun as food.
Lures in agility
For most of the obstacles we teach in agility, reinforcers will be used as lures at some point e.g. to entice the dog over a jump or onto the table for the first time, or put out at the end of a sequence, e.g. at the end of the weave poles, to encourage the dog to run straight through.
Food is usually our first option, especially when we are getting the dog started on a new obstacle. However, overusing food as a lure can become a problem e.g. if I leave a food reward out at the end of the weaves to encourage my dog to run through them. If I do this too often or for too long I train my dog to stop at the end of the weaves and start sniffing around for the food. Now I have a dog who has come to a stop, is sniffing around, tuning me out and is definitely not ready for my directions or to do the next obstacle.
While food at the end of the obstacle may work great to help your dog get over his initial concerns about an obstacle, and is also great for teaching the dog precisely the correct movement or position, you will want to phase it out as soon as possible.
If you have a dog who loves to tug, and you use the tug toy as a lure in the same way at the end of an obstacle, you now have a dog who comes blasting through at high speed, anxious for her game of tug, grabbing her tug toy and immediately turning to you for the game of tug. Now, instead of your dog slowing and sniffing around at the end of the obstacle she is maintaining her speed and focusing on you. A much better option for this kind of training.
Its not the end of the world though if you're still working on your tug drive or you have a dog whose tug drive is never going to be huge, there are other options for luring that are almost as good.
Target objects
One option would be to train your dog to run one or several obstacles in order to nose touch a target at the end of the sequence and then be rewarded. After touching the target you will reward your dog with a food treat or another reinforcer that he likes. The target is essentially a neutral object and your dog learns that he will earn reinforcement for touching it.
For a dog who is already familiar with a target stick, you can use a target stick standing upright on a base.
Another option would be to teach your dog to nose touch a soft object. Examples would be a glove, a sock or a facecloth, something soft that you can toss without it bouncing or blowing away. First teach your dog to nose touch the article while you hold it in your hand at your dog's nose height. Slowly, working over several short training sessions, hold the target lower and lower until you can hold it on the ground. Then try having your dog touch the target on the ground without your hand on it. Finally start tossing the target out further and further so that your dog has to go out to it to touch it. If you get stuck at any stage, you've probably pushed too far too soon, go back and make it easier and work through the steps again more slowly.
The target object might not get quite the enthusiasm a tug toy will, but the more targeting you do and reinforce, the more enthusiastic your dog will be about it. The target object also has the advantage that your dog will re-orient to you for her reinforcer after touching it, rather than sniffing around.
Food-like objects
For highly food motivated dogs a container with food in it makes a better lure than loose food on the ground. Your dog knows that there is food in the container but when he reaches it he will re-orient to you as you need to open it and give him his reward.
Most pet shops carry toys that are designed to be opened and have food put inside. You can also make your own. Some people use toilet rolls sealed with tape at either end, but I personally don't like these as you can only use them once. Also, a dog like my Labrador Hagan will simply rip these open and reward himself, and I want my dog to focus on me to receive the reward.
Depending on the size (and jaw strength!) of your dog there are a number of things you can use. For my smaller dogs I like empty plastic vitamin bottles. For larger dogs an empty plastic peanut butter jar works well.
First, you need to teach your dog that the container has treats in it. Simply place treats inside your container, then open it up and give one to your dog. Rinse and repeat. You can also start tossing the container on the floor and opening it and rewarding your dog when she brings it back to you.
This article may be freely distributed provided it is left intact and the source is credited.
Download the pdf of this article
|
||||
 |
||||
