A Pocket Full of Pony Crack

It's Saturday, and all my plans for the day have been pushed aside because horsy Facebook has delivered me a miraculous gift.

Two trainers who I follow have posted on a similar topic.

Now, while these two trainers don't agree on everything, they do mostly move in similar circles and promote similar ideas. And by complete coincidence, this very week I have been studying the online course of yet another trainer who promotes similar ideas, and the topic I most recently explored there neatly links into what the first two trainers were talking about.

Put it all together and the perfect storm in my brain leads us here - to an exploration of the seeking drive in horses, and how it relates to positive reinforcement (R+) training.

Put very simply, seeking in horses is a very necessary survival mechanism. The brain motivates the seeking of resources required for survival by releasing dopamine to reward successful seeking. Once the resource is acquired, the dopamine reward is released, the consumption of the resource is a formality, and then the seeking cycle starts again.

In the context of food rewards for horses, the seeking behaviour is a quest for dopamine. The horse will experiment and and explore based on previous experience until it hits on the behaviour that gains it the resource - the edible treat. The dopamine is released at the moment the treat is awarded, not while it is actually eating the treat. The more times the cycle is repeated in a session, the less dopamine is released when the resource is gained.

Switching from seeking to receiving over and over moves the horse in and out of the seeking state, which reduces their ability to concentrate on a task and limits their ability to consolidate learning.

If the horse performs a behaviour that it expects will lead to a treat and does not get the reward, it will be motivated to resort to escalating resource-seeking behaviours like pushing, grabbing or pawing, which humans will probably object to. This is the horse attempting to satisfy its confirmation bias and prove to itself that the behaviour does indeed lead to the the treat.

Also, a horse whose lifestyle limits the normal seeking behaviours it can perform in its time away from people may take up compulsive seeking behaviours. This can lead to lead to obsessively hunting for food rewards.

So as you can see, there is definitely potential for food rewards to create problems in training if they are delivered on a schedule that causes the horse to focus more on seeking food than on responding to your cues.

The other thing about seeking is that it is not social. It is linked to the horse's relationship with its environment. It does not serve to strengthen social bonds as it is not correlated with any of the neurotransmitters linked to social bonding.

While seeking can be used to facilitate learning, all the seeking in the world is not going to lead to your horse liking you or strengthen your social bond. Liking comes from social emotions, like care and lust. If your goal is consent-based training where your horse is free to say no and interacts with you because it likes you and being with you makes it feel good, training with constant food rewards is going to hijack this process.

We all do things for money that we probably wouldn't do for free. And horses will do things for food rewards that in that moment they would not do of their own volition. The transaction turns consent into a contract. If the consequence of saying 'no' is that you are denied access to a resource required for survival (food, money), you are not free to say 'no' and there is no potential for consent.

There are those who assert that R+ training is the only non-aversive training method because the horse is not punished or subjected to 'pressure'. But withholding food rewards IS aversive and not getting the food causes stress. You're not going to give the treat if the horse doesn't perform the desired behaviour, because how is the horse going to learn what the 'right' behaviour is?

Shaping behaviour by leveraging seeking may be faster, but there is no way of knowing how the horse actually feels about performing that behaviour or interacting with you if it doesn't have the option to refuse without consequence. All you know is that it would rather do the thing than miss out on the reward.

Optimal use of food rewards comes when the reward is not sought, but is more of a nice surprise. This can be difficult to achieve when the horse can smell the treats in your pocket, or has observed you moving the treat to your hand. Some horses are better able to regulate themselves around food rewards than others, and things like the availability of adequate food (mine are ALWAYS hangry in these dry conditions) or previous experience with treats can influence this.

Forming social bonds with horses is complex. Training through trust and care and allowing the horse to try out different behaviours in response to your cues can be difficult for our results-motivated human brains to grasp. But it is the path to the safest, kindest and least stressful way of training horses that we can aspire to.

Training with R+ definitely has its place, and treats are fine if used appropriately. They can be a great tool for quickly shaping a required behaviour or clarifying an action that a horse has become confused about. It can motivate a reluctant horse to explore contact with humans. A lot of kind, understanding horse people integrate food rewards into their training effectively, every single day.

But if you want your horse to like you and to feel safe and regulated in your presence, relying on activating seeking behaviours in your training is not the way. Prioritising your social bond, encouraging two-way communication and giving your horse room to safely get it wrong are what you need to explore.

All your treats are belong to me.