Is the Clicker a Conditioned Reinforcer?

Melissa Alexander's picture
Filed in - Dogs - Clicker Digest

A primary reinforcer is a reinforcer that an animal is born needing such as food, water, shelter. Secondary, or conditioned, reinforcers are stimuli, objects, or events that become reinforcing based on their association with a primary reinforcer. A dog isn't born wanting to play with a squeaky toy, but when that toy is paired with primary reinforcers such as fun and social interaction, it becomes a conditioned reinforcer. The toy, then, can be used to reinforce behaviors you like, much as you would use a food treat.

The clicker is commonly called a conditioned reinforcer. It is a neutral stimulus paired consistently with a primary reinforcer. Is the association, however, strong enough that you can use the clicker without a primary reinforcer? If so, are there any negative repercussions, and how long can you use the clicker before you needed to "recharge" it? These were the questions debated on the Click-L mailing list this week.

Students at the University of North Texas are currently researching this topic. They've found, according to fellow student Kellie Snider, that "treatless clicks" resulted in frustration behaviors and other learning problems. However that research was done using dogs new to clicker training. Helix Fairweather, a Click-L member, decided to try the experiment on her clicker-savvy Havanese, BJ. Although the "power" of the clicker lasted longer than it had in the UNT experiment, she found that it extinguished fairly quickly, especially once her dog figured out the pattern of reinforcement. When there was nothing in it for him, there was no reason to play the game.

And that, I think, is the crux of the matter. People who use treatless clicks have simply missed the point. We don't use the clicker in order to get by with as little additional reinforcement as possible. We use the clicker to tell our dogs, "That right there is what I want!" The power of the click, to the trainer, is as an event marker.

To the dog, however, the power of the click is as a signal that he gets something he wants. If he can get something he wants by listening to the click and then repeating that behavior you want, then we've found a wonderful win-win tool. But when you take away his motivation for listening, then the clicker is nothing but a tin noisemaker.

About the author Melissa Alexander is a writer and clicker trainer in Seattle, WA. She owns the highly popular ClickerSolutions mailing list and is the author of Click for Joy, the award-winning, essential guide to clicker training.
Jenny Ruth Yasi's picture

How a dog thinks of a click

Hi Melissa!
I just wanted to add that newcomers to clicker training often worry that their pockets will always be full of cheese. But the wonderful thing is, a click not only draws attention to behavior, it also draws attention to reinforcement. If we are consistent about ALWAYS delivering reinforcement after a click, the dog starts to identify things that happen after the click, that he might otherwise have taken for granted, as reinforcements.

Example: We can ask for a behavior, then click and release dogs to go outside. Or click and go RUNNING down the street with our dog. Or click and release to chase a ball or frisbee. Or, click and let Bowzer get up on the couch.

I start with food reinforcement, and I remember the first few times when I gave what might look like a "subtle" reinforcement. I'd click and toss the frisbee and she'd have a slightly surprised expression, but she loves frisbee, so she went for it! Or for example, I asked my dog "leash," and when she brought it to me, I clicked and put the leash on her, and took her for a walk! She might have glanced around for a second there, "click and what? no food? no frisbee?" But then she seemed to get it. "OH. The WALK must be my treat this time." Just as children get more responsibilities as they grow up, clicker training helps puppies learn about all the privileges they can earn.

So if we're walking and she surges ahead of me, I stop. She bounces back to heel position, I click and we get to start walking again. That's a pretty subtle reinforcement, but her history with the clicker is a big part of what helps her to understand that reinforcement, and appreciate it for what it is. Sometimes I've been able to reinforce her calm behavior around other dogs by simply clicking and then picking her up, holding her in my arms, where she feels safer.

So, as my dog matures, I don't need to carry food all the time. After she does a really awesome down stay, I can just click and let her out in the garden!

Formerly a contributing writer for Maine Times and editor/publisher of Harbor Voices: Fact, Fiction and Rumor Jenny Ruth Yasi's writing is found nationally in Mothering Magazine, Pass it On!, Country Dance Magazine, and elsewhere. Her dogs are registered working therapy dogs, and train in agility, freestyle, watersport, and service behaviors.

LIKE TO SEE A COMMENT ON THIS

I am not at trainer. I love the clicker concept but have not used one yet.

I had a dog (dingo, collie X) that I trained to a reasonably impressive level. He died at 15 yrs.
I never used treats as a reward for following a direction. Just praise and patting. This suggests that praise and patting can be sufficient reward.
I have a dog now (dingo, cattle dog X ) that has a stubborn streak he follows directions (“sit”, “down”, “sit”, -comes back up to sitting position, “leave it” that is ‘don’t touch the food yet’ “lie down”, “roll over”) perfectly most of the time time but if he dose not feel like it ti is no go. I trained him with treats and then phased them out. Now, if he is uncooperative I may take out a treat and if I signal “down” he drops like a brick. A little disappointing that i have to go back to treats when is streak clicks in.My question is: with or without the clicker and I suggest with is best, ideally, should not praise be enough without a reward like frizbee, walk or whatever?
kind reagrds,
Ralph

Aidan Bindoff's picture

Dogs Who Work for Praise Alone

Hi Ralph, without reinforcement, operant (think "voluntary") behaviour will eventually cease. To increase or maintain behaviour you must reinforce it. Like gravity is a law of physics, this is a law of behaviour. Similarly, if you are seeing these behaviours maintained or increased, they ARE being reinforced.

Your first dog was probably one of those highly socially motivated dogs who found praise and patting reinforcing. These dogs are few and far between. (I should add that there are many traditionally trained dogs who seem to respond well to praise and patting alone. The truth for some of these dogs is that praise and patting merely signals that a correction will not be forthcoming.)

Even amongst those dogs who find praise and patting genuinely reinforcing, you can get to a point where it ceases to be reinforcing.

A service dog who must perform the same behaviour 20x in a row, 5x each day will probably start to suffer some break-down in the behaviour if praise and petting were the only reinforcer used. These dogs have big hearts and even bigger stomachs!

There are a few reasons why a dog won't perform a behaviour unless you have food in your hands.
1. You aren't reinforcing often enough to maintain the behaviour
2. Food may have become part of the cue to perform the behaviour
3. There may be too many distractions for the dog to perform the behaviour without enticement.

Getting a dog to perform behaviours reliably without food isn't difficult, but you can't take short-cuts. My dogs regularly perform cued or uncued behaviours without reinforcement, and more often than not I don't have food on me or near me. This was certainly not always the case!

Regards,

Aidan

http://www.PositivePetzine.com 

Jenny Ruth Yasi's picture

For Ralph

My teacher always told me to "get the treats out of your hand" and so I got pretty tricky, left treats on windowsills and in the crotch of a tree. My dogs didn't need to see a treat in my hand to suspect maybe I had something up my sleeve.

To get over this hump, go ahead and train with food in your hand, but don't ever feed *that* food to the dog. Reinforce like it's a coin trick, you have to figure out how to pull reinforcement out of thin air! Slowly reduce the amount of food in your hand until the dog learns that food or no food in your hand is irrelevent, because you've always got some kind of reinforcement "up your sleeve."

Jenny Ruth Yasi's writing is found nationally in Mothering Magazine, Pass it On!, Country Dance Magazine, and elsewhere. She trains and teaches canine fun and games in Portland Maine.