I use food a lot in training, especially in early training. Food is for many dogs and situations a great reinforcer – it is a biological imperative that dogs eat, it is easy to use, allows for rapid reinforcement, and is a powerful and effective teaching tool. However, there are a lot of times when I don’t use food. Knowing when and how to use food is important, but equally important is knowing when food may not be the right or best choice in training. Reinforcement can not just be used to create behaviors, but, when thoughtfully implemented, can be used to create a state of mind. You can use reinforcement to “rev” dogs up and calm them down, depending on your goals within a given session. Here are a few examples of training in which reinforcement other than food may be indicated:
Wait to get out of the crate: I generally don’t use food at all to teach this behavior. When the dog is in the crate and the door is shut, I simply wait for a relaxation behavior, usually sit or down. As soon as the dog assumes the position, I begin cracking the door open. If the dog’s butt lifts off the floor, the door closes. The door begins opening again when the dog’s butt hits the floor. Repeat until the door is all the way open and the dog is maintaining the position for duration. In these situations, I will often reward the release from the crate, (“Free!”) with food or another reinforcer. Sometimes, the opportunity to get out of the crate and engage with the handler is reinforcement enough! I do a variation of this exercise to teach dogs to wait at boundaries and not bolt through doors as well.
Loose leash walking: I usually do train this behavior with food. However, in some dogs where the behavior is not firmly entrenched with a strong reinforcement history, I use manipulation of distance as the reinforcer. Among other reasons, dogs pull because it works – it gets them where they want to go. As soon as the leash goes tight, I begin moving in the opposite direction, slowly and deliberately. No jerking on the leash, but I do apply slow, steady, and consistent pressure. I do find that clicking and treating the dog when he chooses to be in the appropriate position speeds this behavior along, but is certainly not necessary. You may look like a yo-yo at first, five steps in one direction, ten in another, etc., and may not make it past your block at first, but with consistency this method can be very effective.
Relaxation behaviors: While I do use a clicker and treats to teach a dog to go on the mat and settle, both of these reinforcers may be counter-productive when it comes to building relaxation on the mat. I want the mat to be a cue for relaxation. If I’m clicking and treating frequently, the clicks and treats may actually get the dog more aroused. In these situations, depending on the dog, massage, speaking in a soothing voice, T-Touch exercises, tandem breathing, etc., may actually work better in teaching the dog to pair the mat with the relaxation experience I’m going for. If I do use food, I would likely choose a low-value food reinforcer and instead of pairing it with the click, would pair it with low, slow, and soft verbal reinforcement (praise). Sometimes food amps a dog up when we need him to calm down.
Dealing with prey drive: If Mokie is fixated on a squirrel, deer, or porcupine (she finds porcupines fascinating, much to my dismay), she doesn’t care what kind of food I have. I could have a perfectly cooked steak and she wouldn’t even twitch her nose in the direction of the tantalizing aroma. In that situation, the squirrel chase is the most potent reinforcer I have. Trainers can get themselves in a trap here – they begin offering more and more tantalizing food when the dog doesn’t actually want food, this can significantly reduce the power of the food reinforcer in other situations and can also train a dog to actively ignore even high-octane treats. Sometimes there are reinforcers in the environment which are much more salient in that exact moment – real life rewards that trump your food. These reinforcers masquerade as distractions. They may include scenting, chasing, biting, digging, running, etc.
Proximity sensitivity: For a dog that is fearful, nervous, or otherwise reactive, sometimes distance can be a powerful motivator. If a dog does not want to be near another person or dog, using distance from that dog or person as a consequence for desirable behavior can be a powerful “functional reward.” This concept is the foundation for Grisha Stewart’s wonderful behavior modification protocol Behavior Adjustment Training. Distance is the primary reward for the dog – food, if used, is only a “bonus.”
Chaining: When I’m creating a behavior chain, the cue for the next behavior should be the reinforcement for the previous behavior – the primary reinforcer should come at the end of the chain.
Crazy Cuba: My adolescent Saint Bernard, Cuba, was very well-socialized as a puppy. He loves other dogs. In fact, he loves them so much that he loses his mind and acts a total fool when the leash prevents him from engaging with them at his whim. At a glimpse, his response looks much like that of my previous rescued reactive Saint, Monte. If you look closely, there are subtle nuances which indicate the situation is very different. Monte wanted other dogs to “go away, now!” His body posture was low and stiff. His barking and growling was low in pitch. Contrast this with Cuba, who is also lunging on the leash and barking. However, Cuba’s body is bouncy, front paws po-going off the ground. His bark is higher in pitch, interspersed with an occasional whine. His eyes are not hard, but bright and shiney. On walks, when he displays this behavior, if he cannot get to the strange dog, he starts play bowing at Mokie, my Chow mix. This is not a dog that is threatened by other dogs (although if I handled this in the wrong way it could turn into that), this is a dog that wants to play and has no mechanisms in place yet for controlling those desires and channeling them into appropriate outlets.
With Monte, I would have used visual barriers and going farther away from the scary thing as a reward. With Cuba, visual barriers would increase the frustration, as would moving away. He doesn’t care about food in these situations, he wants to play! For this, I’m experimenting with using tug as opposed to food – playing reduces stress and frustration (is in fact incompatible with stress). This also gives Cuba what he wants in that moment – a way to channel that excitement into play. The benefit is, it channels that focus back to me as well. We also use manipulation of distance – calm dogs get closer to that friendly dog and maybe get to play!
Because I believe his behavior is not exclusively barrier frustration but, at least as it relates to being in the classroom, contains elements of territoriality as well, sometimes I use access to the classroom as the reinforcer. He wants to be in the classroom with the other dogs, but that is a reward for quiet dogs and not a place for demanding barkiness. Barky? Quiet, prompt, unemotional removal from the classroom followed by a brief (30 – 60 second) time out. Quiet? The reward is coming back into the classroom and resuming work.
Jumping: Mokie LOVES to jump. Loves it. New clients at the classroom think it’s cute because she’s small and adorable. Because they are untrained, they tend to reinforce this behavior. Some people actually like having their dogs jump on them but are understandably frustrated when the dog jumps on guests. For Mokie, the opportunity to jump is a powerful reward. There are times when I’ve used the opportunity for “legal jumping” (jumping up to touch a hand target) as a reward for polite greetings. If you sit, then you can legally jump.
In other, socially gregarious dogs, the opportunity to greet [another dog, person, or both] is often the sole reward you need – jumping makes people go away, “four on the floor” means friends and scratches and greeting opportunities!
These are but a few of the examples when I may not choose to use food in training, or even a clicker. Yes, I’m a clicker trainer – but clicker training is not about the clicker, or the food necessarily. It’s about creative use of reinforcement, understanding what the dog wants, and finding ways to use those things to your advantage.