This post is part of a series of several posts on cues.
I recently posted about putting cues under stimulus control. (Read that post here.) However, I realize now that perhaps I should have started with a discussion of what exactly constitutes a cue.
A cue (sometimes called a discriminative stimulus, or SD) is an environmental signal that triggers a behavior. The animal has learned through conditioning that she can earn reinforcement by performing the behavior in response to the cue. Clicker training uses cues as green lights for the animal–the animal knows that by responding to the cue, she earns a treat, belly rub, chance to play with a favorite toy, or some other sort of reward. (What makes a good reinforcer?)
We generally think of cues as sometime we purposefully teach, such as training a dog to come when we say the word “come.” However, a cue is anything the animal learns to associate with the potential for reinforcement (or punishment). Get out a can opener and a can of cat food, and most cats will come running. Get out the cat carrier and the cat is no where to be found. The first cue signals to the cat that he might earn a reward if he comes and the second indicates a punishing trip to the awful vet’s office is in store. The cat quickly learns both cues, even if the owner had no intention of teaching them.
What can be used as a cue?
Anything that the animal can perceive can be used as a cue. Common types of cues include:
- Visual Cues
- Auditory Cues
- Tactile/touch Cues
Visual cues include anything the animal can see. Many dogs learn hand signals much easier than they do words. (Partially, perhaps, because they’ve learned to tune out our constant chatter.) We can also teach the animal that the presence of an object means a certain behavior should occur. For example, in the presence of a jump a horse learns that he should jump over it, rather than run through it or stop. The jump becomes a visual signal for the behavior of jumping.
Auditory cues include the words and sounds we use to ask for behavior. Humans often teach verbal cues to animals. I’ve taught my dog words such as sit, down, stand, stay and so forth. Many sheep herding dogs are taught to respond to whistle signals. One whistle means go left, another means go right, one means come, one means go away, and so on.
Tactile and touch cues occur whenever we come use physical contact to give a signal. For example, I rub the inside of my horse’s leg to ask her to pick up her foot or I place a hand behind her ears to ask her to drop her head. In the horse world, these types of cues are usually taught with lots of pressure and negative reinforcement. (What is negative reinforcement?) However, they don’t have to be taught with negative reinforcement. (I’ll talk more about this later in the week.)
Cues should be Clear and Consistent
Any cue we use should be easy for the animal to understand and distinct from other cues. Pick words and hand signals that can’t easily be confused with other already established cues, especially at the beginning of training. Any animal (but especially an animal new to training) is liable to get confused and frustrated if they can’t differentiate between two cues to figure out what you’re asking. Cues should be sharp and clear, especially when you first teach them. Later, once the animal knows the behavior, you can gradually fade out the cue until the animal responds to a tiny hand signal or a mere whisper.
You might have covered this elsewhere but I find that combining cues–verbal with a hand signal often is a better combination–you can then fade them or use one or the other depending on the circumstances.
You do a very clear summary of some technical aspects of training and I enjoy coming in to read your posts!
i have to agree with the above comment…verbal plus hand signal is definitely the best
I have just come across your site and I really wanted to congratulate you! I have a background in psychology but I now work in the horse industry in the UK. Many, many times I come away from conversations/lessons with owners/riders with a heavy heart. This is because the horse world is really very unscientific in it’s approach to equine behaviour and the way horses learn. It appears to me that unlike dog training (which does appear to use scientific methods more) the horse world is full of anecdotes and myths about riding and horse management that ultimately adversly affects the welfare of the horse. Well done! I have bookmarked your site and will continue to read with interest.
Hi Twilko,
Thanks for the positive feedback! I’m so glad that you are enjoying my site.
I think the horse community is making progress, however as you noted, horse training is still years behind dog training.
Mary
Hi Twilko,
Thanks for the positive feedback! I'm so glad that you are enjoying my site.
I think the horse community is making progress, however as you noted, horse training is still years behind dog training.
Mary