One of the concerns that people have is that using food (or other added reinforcers) somehow changes the motivation of the horse to do behavior, and that this is detrimental to the relationship between the horse and their trainer. Similar questions are asked of anyone who uses positive reinforcement training, whether it’s teachers giving out stickers or zookeepers working with elephants.
This is often called the debate between using intrinsic and extrinsic reinforcers. Intrinsic reinforcers are reinforcers that come from doing the activity itself. I ride because I enjoy riding. Extrinsic reinforcers are reinforcers that are provided as an additional incentive to do the activity. If I rode only to compete and get ribbons, then I would be riding for an extrinsic reinforcer. Whether or not there is any intrinsic reinforcement to an activity can vary and is not fixed. It can depend upon a lot of factors and there will be some days when it is sufficient to motivate you to do it, and others when some extrinsic motivation helps. If you want to get me to clean my house, adding an extrinsic reinforcer would work better than relying on any intrinsic reinforcement. But if my house is REALLY dirty, I might be motivated to clean it without additional reinforcement.
The question of intrinsic vs. extrinsic motivation is a hot topic in business because employers want to understand how to motivate people to work harder, be more responsible, be more creative, and perform better in lots of other ways. There are a lot of questions about it. Is it a function of how much they are paid? Is it more about how the type of job? Is it about the people they work with? What makes one person a super employee who eagerly takes on new challenges, compared to another employee who does the bare minimum?
These same questions are relevant to horse trainers. What is the best way to motivate our horses? At the Equine Clicker Conference in 2014, there were some video presentations and one was about intrinsic and extrinsic reinforcers. The presentation was given by Inge Teblick, who is a horse behavior consultant from the Netherlands. She has published several books and is active in teaching and promoting clicker training. Her website is http://www.ingeteblick.be/en/welkom-2/.
This article is based on notes I took while watching her video lecture which was about how to apply the principles of game design (computer games) to horse training. This is not the first time I have seen a comparison between animal training and designing computer games. In a private discussion at the Art and Science of Animal Training Conference (ORCA), Joe Layng talked about how people who write games are experts at shaping behavior. And Inge makes the same point. In order to keep someone playing a game, you have to know what they need at any given time in order to keep them playing long term. Here are some highlights from her presentation in which she took a close look at motivation and how extrinsic and intrinsic motivation work together.
Inge lists the 4 x-factors of “fun” as:
- Autonomy: being able to make choices
- Competence: having the skills to be successful
- Purpose: is there some sense in what you are doing?
- Relatedness: social learning, is there some interaction with other individuals?
If I think of the activities that my horse Rosie enjoys most, they all have these four elements. If I think of some where she seems less enthusiastic, I can see where they are lacking in one or more areas.
The first time I heard that extrinsic rewards could decrease intrinsic motivation, I didn’t understand how this could happen. It seems like getting additional reinforcement for something you already enjoy could only add to the value of the activity. But it turns out that how we are reinforced for something does change our motivation for doing it, and in some (but not all) cases, extrinsic motivation can decrease intrinsic motivation and lead to an overall decline in performance.
Daniel Pink talks about this in his YouTube video on TED, where he compares intrinsic and extrinsic motivation (http://www.youtube.com/watch?v=rrkrvAUbU9Y). Even though he says intrinsic motivation is “better,” he does say that there are certain situations where extrinsic motivation does seem to work. Inge expands his list to include other conditions under which extrinsic rewards can work in the long run:
Extrinsic Reinforcement Works Well Under These Conditions:
- When there is a simple set of rules and a clear destination.
- When there is no intrinsic motivation to start with.
- When there’s an engagement gap towards autonomous action and intrinsic motivation is already high. This refers to the case where the horse wants to do something but needs a little external encouragement to tip the balance toward doing it.
- When there’s plenty of room for autonomy, purpose, and relatendess. In this case there would be a mix on intrinsic and extrinsic motivators.
- In individually tailored variable ratios ( which means the reinforcement schedule is appropriate for that learner for that behavior in that situation).
- If the extrinsic motivation provides positive feedback rather than tangible rewards. We often say that praise is not a good reinforcer, but it can be effective if it is reinforcement for a task that has been completed (think of the audience clapping for a good performance).
I always like to think about how to make horse training more interesting for the horse, but there are times when it seems difficult to figure out how to present a training exercise as “fun.” This lecture gave me some new ideas for how to think about training, even when we are working on challenging tasks. If you enjoyed this article, you may be interested in Inge’s free on-line course titled “The Theory of Fun.” You can find her course at http://itrainmyhorsesbrain.com/course/theory-of-fun-en/.