Without understanding the science behind clicker training, it can be hard to see how a click means anything to a horse in clicker training. The click is what is called a marker or bridge signal. It bridges the gap between the behavior you are reinforcing and the time it takes to deliver the reinforcement. Positive reinforcement can be used without a bridge signal, however a bridge signal has proven to be extremely effective at clearly communicating to the animal the exact moment/behavior that is being reinforced. For example, if you are teaching your horse to pick up the correct lead on the lunge line and he gets it right, by the time you get him stopped and reward him, he may think he is being rewarded for stopping, not picking up the correct lead. With a bridge signal, the trainer can mark the correct lead and the horse will know that is what he is receiving reinforcement for. So how does the horse know the bridge signal is marking a certain behavior?
In the 1890s, Ivan Pavlov studied dogs’ response to food by salivating. With the dogs, Pavlov observed that they salivated before feeding them. He rang a bell before feeding them and eventually the dogs started drooling at the sound of the bell, because of its association with the food. What he discovered was that by pairing a neutral stimulus (bell ringing), something that could not itself illicit a response, with an unconditioned stimulus (food), something that produces an unconditioned or natural response (drooling) that eventually the neutral stimulus (bell) alone would illicit the same response as the unconditioned stimulus (food) and become a conditioned stimulus.
With clicker training, we use a distinct sound like a click and pair it with food. In the initial stages when we condition the clicker, we click/feed, click/feed, click/feed. This is how we give the clicker value. It is how the horse knows that the bridge signal is rewarding him. When the horse hears the bridge signal, dopamine spikes. It’s part of the involuntary response produced by the clicker through its association with the food.
Classical conditioning deals with involuntary responses where operant conditioning is made up of conscious decisions made by the learner. However, in training the two always overlap. The animal is always forming associations. In the above example involving the dogs, we the unconditioned stimulus (food) is an appetitive that creates a positive association, but classical conditioning works both ways, meaning it can involve aversives too and create negative associations. For example, if your saddle does not fit your horse properly and causes him pan, eventually the horse will shy away when you approach with the saddle. Many people don’t understand why their horse has developed a sudden fear of the saddle seemingly out of the blue. The saddle is a neutral stimulus, it does not naturally cause the horse to shy. However, pain is an unconditioned stimulus, it does naturally cause the response of fear and shying. Through pairing the two together the saddle and pain, eventually the sight of the saddle produces the response that the pain does. The horse has learned that the saddle predicts pain.
In riding, we are taught to ask, tell, demand. This is classical conditioning as well. When we squeeze with our legs or point with the lead rope it is a neutral stimulus for the horse- it doesn’t produce a natural response. But when we pair it with a conditioned stimulus- pain/fear of pain by kicking the horse, using a crop, or swinging a stick and string that produce a natural response- move or run, eventually the horse forms the association that the squeezing or pointing means pain/fear and he will move/run when the neutral, now conditioned stimulus is presented.
Whether you choose to utilize clicker training or not, it is important to understand classical conditioning because it is always happening whether you intend it to or even realize it. Without understanding how and why our horses are forming good or bad associations with various things, we cannot create the ones we want.