Conditioning

I put together this chart for my book, which should be out in a few months (with the kind guidance from Dr. Jenifer Zeligs and reference from her amazing book: https://www.amazon.com/Animal-Training-101-Pr…/…/ref=sr_1_1…)

I wanted to share it because we've had a lot of discussions about these topics lately! I thought it would help to have some clarification on some of these topics...

Classical Conditioning is simple pairing, X=Y. We take a stimulus that elicits a response from the horse (simply put, an object that has meaning to the horse) then we pair it repeatedly with a stimulus with no meaning. The neutral stimulus (the one with no meaning) gains the meaning of the other stimulus. So simply put, when we pair an aversive with a neutral a few times, the neutral will take on the aversive meaning. At the same time if we pair an appetitive stimulus with a neutral stimulus a few times, the neutral will become appetitive.

We take advatange of this conditioning with our clicker and a whip. A clicker we classically condition as a predictor of a wonderful appetitive. The horse learns when they hear the click something wonderful will happen, soon the click alone elicits the same emotional response as the reinforcer it's been paired with! The whip works the same way. When the sight of a whip is paired with tapping, smacking, flicking or any aversiveness, the sight of the whip will carry the same aversive meaning. This is why often we only need to carry a whip for the horse to become more forward. Conditioning is always happening, intentionally or not.

Habituation is like the opposite of conditioning, it's when a stimulus that elicited an emotion, gradually looses it's meaning, through lack of pairing. So for example if you take an appetitively conditioned clicker, and click it 100 times without pairing the reinforcer, the click will lose meaning. Over time the clicker will fade to an almost neutral meaning - not entirely loosing it's meaning, but enough to be an ineffective training tool. This is why some trainers are very avid about reinforcing with every click. The same thing happens with aversives. If you take you frequently ride with a whip, to encourage forwardness, but never pair it with the actual aversive, the whip will gradually lose meaning. We can take advantage of this with objects we'd like our horses to be comfortable around, objects we'd like to desensitize it can help to allow the horse to habituate to it first, at their own time. While the object may never reach neutral on it's own, the meaning does reduce over time.

We can take the extra step and use systematic desensitization to bring the stimulus the rest of the way to neutral. This can speed up the process of habituation and bring the stimulus much closer to neutral. Which can open our door to begin some counter conditioning.

Counter conditioning is really what it sounds like, conditioning the opposite. We do this the same way as regular conditioning, with pairing. If we take a stimulus that held an appetitive meaning and pair it with a stimulus that holds an aversive meaning, the appetitive soon predicts the aversive, and soon becomes aversive in itself. This can lead to poisoned cues when we aren't careful about mixing learning quadrants. It can also happen in reverse, where we take an aversive stimuli and pair it with an appetitive, soon the aversive will predict an appetitive and gradually gain it's own appetitive meaning. However this is always a precarious job, we want to ensure the stimulus is moving in the direction we want. To ensure we're moving the correct direction you need to make sure you keep your scale heavily tipped in your favor. If you're trying to counter condition an aversive, you'll want to ensure the appetitive heavily outweighs the aversive, by reducing the aversiveness of the stimulus (with habituation, SDS and by making it as small as possible). The same vice versa.

These can be complicated topics to wrap our minds around but it's important we do, because Conditioning is always happening! Classical, habituation, DS, and CC all happen in the background of all the operant training we're doing. While we may be teaching a "back up", with a target and a clicker, what else are we conditioning into the scenario? Have we conditioned our relationship a little more appetitive?

<- Back to Training Articles

Skip to toolbar