Most of the communication that horses receive is from human to horse, we always are telling them what to do and when. However, with newer research being put into teaching animals to communicate one barn has taught horses to communicate when they want to wear a blanket or not. Learning theory has been increasingly becoming more popular in the equitation world, giving us more of an insight into the cognitive abilities and preferences of horses.
One barn used reward based operant conditioning to teach horses to touch a board, and to understand the meaning of three different symbols (Mejdell, 2016).
Operant conditioning is teaching animals to form associations between behaviors and resulting events. In this case the behavior was touching a board with a symbol, and the resulting event was removing or putting on a blanket. The researchers use positive reinforcers to shape the behavior they wanted. Shaping is reinforcing behaviors to get closer and closer to a desired behavior (Mejdell, 2016). So, in this experiment they started with reinforcing just the horse walking up and touching the boards, then they continued reinforcing the behavior and added the resulting consequence until the horse understood desired behavior.
The reinforcer in this experiment was positive reinforcement, this is when you add a desired stimulus, and in this case, it was horse treats. The treats were a primary reinforcer, meaning that they were something that the animal finds inherently reinforcing. The researchers also used clickers as a secondary reinforcer as a bridge between the behavior and the primary reinforcer. This was so that there was no lag between desired behavior and positive reinforcer (Mejdell, 2016).
The experiment was done because of a commonly disputed management technique in the winter. To blanket a horse or not to blanket a horse. They hoped to be able to “ask” the horse whether or not they wanted a blanket, and then record the temperature that they chose a blanket to better understand when horses truly felt cold (Mejdell, 2016). The experiment used 23 horses, 13 cold-blooded, 10 warm-blooded, 3 Arabians, and 1 Thoroughbred. All ages from 3 to 16. The positive reinforcement training was done by an animal trainer, and they used this training to teach horses to choose whether they wanted a blanket on, off, or not changed. The horses, once the symbols were learned, was given a single choice and given a consequence (Mejdell, 2016).
Though this experiment they found that horses were able to discriminate between simple visual symbols (the boards) and extend that to associations of different outcomes (the blanket on or off). As could be predicted, horses preferred to wear a blanket during bad weather and stay without a blanket in nice weather. When 10 horses were tested on sunny days with a relatively high temperature (about 68 degrees Fahrenheit) all 10 choose to have their blankets off, and when they were tested on another day with and temperatures about 41 degrees Fahrenheit these same horses choose to keep their blankets on (i.e. no change) (Mejdell, 2016). The understanding of choice showed the horses were able to understand the effects that the blanket change would have on their thermal well-being (Mejdell, 2016).
This experiment paved more paths for operant conditioning to be seen as a successful way of training horses. Within the past two years positive reinforcement, and clicker training have been becoming more popular. Many people are now realizing that horses are very motivated learners, and can easily distinguish and learn what is right and wrong. From my own personal experience with operant conditioning with horses, I have found it very easy to teach a horse different tasks using positive reinforcement. Such as lifting their feet, coming in from the pasture, heeling, ground manners, and overcoming fears (because horses are naturally flight animals). They pick up on cues and body language quickly, and will do anything for a reward.
Citation:
Mejdell, Cecilie M., et al. “Horses Can Learn to Use Symbols to Communicate Their Preferences.” Applied Animal Behaviour Science, Elsevier, 29 July 2016, www.sciencedirect.com/science/article/pii/S0168159116302192.
