Horses Learn to Communicate Blanket Preferences with Symbols

Horse Blankets

The debate over whether or not horses should wear blankets will probably never end in equestrian circles, but now, at last, the horses can weigh in on the debate. Recent research provides some insight into blanketing, and more importantly, shows the incredible learning and communication abilities of horses.

In a study published in the journal Applied Animal Behavior Science, researchers at the Norwegian Veterinary Institute taught a group of 23 horses to indicate their blanket preference by selecting one of three boards. One board had a horizontal line to indicate “put blanket on,” another had a vertical line for, “take blanket off,” and a blank board indicated that the horse wanted no change from his current status.

Teaching the Language

There were several steps to the process of teaching the horses the meanings behind the boards. First, the researchers introduced each of the two boards with symbols separately. When horses touched the board with their muzzles, they were rewarded with a piece of carrot as the action that corresponded with the board was performed.

Second, they were shown both of the boards with symbols at once and received a treat only when they touched the board indicating a change from their current status. For example, a blanketed horse would receive a treat when he touched the vertical line to indicate “blanket off” but nothing for touching the horizontal line.

Third, they learned to use the symbols in relevant situations. Horses were tested with the two symbols in cold, wet weather when they were showing signs of being cold, and when they were overblanketed for the weather and obviously hot.

Once they had selected the relevant symbol 12 times, the fourth step was introducing the blank “no change” board to the equation. When a horse touched the blank board, he would receive a treat and no change to his current blanketing status.

Next, all three boards were used in differing combinations and locations, and horses would again receive carrots when they selected a relevant choice.

Finally, the communication system was tested in real-world weather conditions over several months. The results showed that the horses not only understood how to communicate with the symbols, but that they did generally want to be blanketed when the weather was bad:

  • On two sunny, warm days (20-23 ° Celcius, or 68 to 73 ° Fahrenheit) 10 horses started with blankets and 12 without; all indicated the boards that would leave them blanketless.
  • On a rainy, 9 °C (48 °F) day, 10 out of 12 horses who started out blanketless indicated the “blanket on” symbol while two opted to remain unblanketed.
  • On two days where the temperature was -12 °C and 1 °C (10.4 °F and 33.8 °F) and there was sleet falling, all of the horses “asked” for blankets, including the two who had opted to go without on the rainy day.

The researchers explain:

The fact that 22 of 22 horses signaled that they preferred to be without a blanket on summer days without rain and that 20 of the same 22 horses signaled that they wanted the blanket on when it was continuous rain, windy and chilly, strongly supports our prediction that if the horses understood the symbols, their choices would vary with weather.

In their discussion, the researchers acknowledge that the any testing of animals when humans are involved can introduce the human bias into the equation (such as subtle, even unconscious positive cues when a horse selects the “correct” symbol.) However, they believe that by offering treats regardless of which relevant board is selected, they eliminated that bias. To read more discussion of the training method and how the researchers worked to eliminate any potential biases, view the full text here.

Horse Blankets

Teach Your Horse

Could average horse owners apply this method at home to get their own horses to provide some input into blanketing decisions or other management questions? It’s certainly possible. The researchers taught all 23 of the horses to respond to the boards within just 14 days of training (only 22 were ultimately tested as one horse, named Katug, died shortly after the training phase of the study.) The horses used in the study were mares and geldings of various breeds and ages, but all were considered average riding horses. The one prerequisite was that all horses had been blanketed prior to the study; a horse unfamiliar with blanketing might react differently.

Another consideration is the temperature conditioning phase (step three). Ideally, you’d need access to conditions that are both fairly warm and unpleasantly cold in order to teach the association between blankets and comfort/discomfort.

As with any study in animal behavior, it’s important to remember that all horses are individuals, and there will always be some quirks to consider. The article mentions one young test subject in the group named Blue, who seemed to simply enjoy putting the humans to work; he always selected the symbol to have his blanketing status changed and needed some extra temperature conditioning so that he would understand that this form of entertainment could lead to consequences for his own thermal comfort.

Horses can learn to use symbols to communicate their preferences
Mejdell, Cecilie M. et al.
Applied Animal Behaviour Science , DOI:

Leslie Potter is Managing Editor of


  1. First of all I know nothing about horses. My nephew has a horse and I was wondering about making a horse blanket out of a old comforter for his horse to be worn only while in the stall during the cold winter. Would it be to hot or to thick for a horse blanket? Please be honest. Thank you Betty


Please enter your comment!
Please enter your name here