The use of treats in horse training can have mixed reviews depending on who you talk to. I’ve met horse trainers that have a strict no hand-feeding rule, stating that it will create a biting horse. To some extent, they are correct, and by staying clear of feeding treats via their hands, they are certainly stopping this from happening. But by excluding this valuable asset in the training of a horse, they are also missing out.
A big moment for me was starting to understand the science behind the WHY it works. We need to understand the operant conditioning quadrant to help us here.
Behaviour can either be reinforced (meaning it’s more likely to happen again) or be punished (meaning it’s less likely to happen again). We then also see the words positive and negative. Positive means to add something, and negative means to remove something. None of these quadrants function as a mutually exclusive paradigm, and understanding the model helps us see this interconnectedness.
In traditional horse training, most behaviour modification sits within the realm of negative reinforcement. This is typically known as pressure and release. The pressure is added, the horse finds this uncomfortable, and so does something to seek out their own comfort. If the horse has done what the trainer desires or a step in the direction of what they are trying to teach, then the pressure is removed, thus giving the horse the relief and comfort. The horse then learns if they do X behaviour, then there is comfort and relief in this, and they are more likely to perform that behaviour again. The horse learns to perform the behaviour the trainer requires of them because they are trying to avoid the consequences of what comes next—most of the time, this is an escalation in pressure. A good trainer will try and avoid too much escalation and will remove the pressure early to help the horse out—way before any panic or worry sets in. I’ve also seen trainers who I look up to quit the session when the horse has given a really great try—the ultimate use of pressure and release. The horse is put back to their stall or pasture and allowed to soak on everything they have learned. They often then come back to the next session having a good understanding of the skill they were learning at the end of the last session.
This is the bit that got me, however: when the ultimate reward in using negative reinforcement as your primary reinforcement paradigm is to quit the session.
I understand the premise here, and I believe it’s very valuable. Too many times, we keep asking for ‘just one more,’ and often this is our downfall. We overcook things. I’m sure we can all put up our hands and admit to being guilty of that. But my overarching intention in training my horses is to build and enhance a relationship with them. I want them to WANT to come to a training session. I want them to not want it to finish, to be eagerly waiting at the gate for the next one—not really to have their ultimate reward be for it to finish.
Before we get lots of people bashing this, I do understand there are different circumstances when one might find the above to be more beneficial than the relationship-first approach. For horses that are assigned a trainer for a short amount of time—one that has been employed to teach them X, Y, Z—with not a huge amount of knowledge as to what kind of home the horse is going to end up in, I can completely see why trainers might choose the above type of approach.
I also see and 100% agree there is a real need for a horse to have a thorough understanding of pressure and release. Pressure is everywhere, and teaching a horse to expand their tolerance and understanding of it, in my opinion, is a necessary skill set to teach.
Positive reinforcement is the addition of something the horse values in order to make a behaviour more likely to happen again. The key element here is ‘something the horse values.’ I see people all the time say that they use positive reinforcement when their horse does something well and proceed to give the horse a big slap on the neck. Generally, this is paired with the stopping of whatever they were doing. The horse learns to tolerate the big slap on the neck and also learns that it is usually the end of what they were doing. It is not that the pat was reinforcing in itself but more being paired with the stopping of the activity. The alternative here might be to use a scratch as opposed to a big slap—a scratch that elicits the nose curl from the horse; one that they really enjoy. This can also be paired with the stopping of what they are doing, and the rider may find that it has an even bigger impact. The horse is getting something out of it that THEY enjoy—not just the thing that the rider thinks is a reward for the horse.
Another reinforcer that normally holds high value to a horse is food. Food can come with its own set of challenges because it is so highly valued by some horses. I like to think of food as an accelerant for behaviour. It can go in either direction—good or bad—really quickly. It’s important to understand that this can happen and to have a baseline of rules to follow if introducing food rewards, like teaching your children good table manners; I’ll save more about this for another time.
The thing to note here is that the horse is wanting to earn the reward and so seeks out what it is that has earned them it. The horse is a voluntary participant in their learning.
Both positive and negative reinforcement have their place. Both, done well, create well-rounded horses.
For me, a blended approach works well. My horses have a good understanding of real-world pressure; if they were ever not with me, then I trust they could cope out in the world wherever they may end up. They also look forward to their training sessions, generally don’t want their sessions to end, and also want to come back in the next day.