“Our aim was . . . to develop a tool to ‘ask’ horses whether or not they prefer to wear a blanket under different weather conditions,” the authors wrote in Applied Animal Behaviour Science. In the end, they added: “Horses chose to stay without a blanket in nice weather, and they chose to have a blanket on when the weather was wet, windy and cold.”
More than a century ago, a German horse nicknamed Clever Hans became a celebrity in Europe for his whiz-like ability to answer complex math, reading and spelling questions by tapping his hoof. Today, he’s an example of some of the pitfalls of scientific studies involving people and animals: Hans, it turned out, was stumped when his human testers didn’t know the answer to the question — which meant he was actually clever at reading the unconscious cues of people.
Horse cognition research has come some way since then, and the animals have shown savvy at recognizing shapes and telling objects apart. But this task added another level.
The subjects were 23 ordinary riding horses with charming names including Poltergeist, Virvelvind and Romano (one horse died shortly after the training, leaving 22 in the end). All had previous experience wearing blankets when their owners deemed them necessary. To help the horses express their own preferences, the researchers created three simple boards: One with a horizontal bar meant “put blanket on”; an unmarked board meant “no change”; and a board with a vertical bar meant “take blanket off.”
Then the horses went through a methodical literacy course of sorts. First, the trainers introduced the symbols one at a time. Horses that touched it with their muzzles would be rewarded with a thin slice of carrot, and then the corresponding action would be carried out — the blanket taken on or off. Once the horses had that down, they were shown both symbols at the same time, but they only got their treats if they touched the “relevant” one. In other words, if a horse wearing a blanket touched the symbol that meant “put a blanket on,” he’d get nothing.
Next, the researchers introduced temperature to the equation. Horses were draped in thick blankets that made them sweaty and “obviously hot,” or they were put outside in what the Norwegians called “challenging weather” — so it really must have been awful — until they tensed, tucked in their tails or showed other signs of being cold. They had to pick the relevant symbol 12 times to move onto the next step, which was meeting the blank “no change” board. Picking that one always earned a carrot, but never a blanket status change.
After that, the horses saw all three boards in various combinations and locations, and only relevant choices led to snacks. Then they were given a “free choice” between two relevant symbols.
But the “ultimate test,” the researchers wrote, was carried out in the elements — several variations of them over the course of nearly a year. And the horses nailed it, each and every one of them.
“In total, these results strongly indicate that the horses had learnt to communicate their preferences using symbols,” the authors wrote.
Here’s a chart that depicts their choices:
The horses learned this all in just 10-15 minutes of daily training over 14 days, although a few class clowns slowed things down.
Poltergeist and Runa, for example, learned so fast that they “began to explore other possibilities and solutions to earn more carrot slices,” such as nibbling on the boards, and “needed time to be convinced that there were none,” the study said.
Blue, for his part, “seemed to enjoy the event of blankets taken on and off, as he always touched the ‘change’ symbol,” the authors wrote. But with a few extra hot and cold tests, Blue, like his peers, figured out that choices have consequences — in this case, sweat, shivers or just the right amount of toastiness.