Can Horses Think Through Problems?

Q. I am writing a paper in my agricultural ethics class on the treatment of horses, and one of my discussions deals with the equality of horses to humans. My roommate and I were debating whether horses have the capability to think through a problem--something like unlatching a gate. I'm having trouble finding research on this specific problem. Do you know of any research that has been done in this area?


A. Your question is at the heart of a field of academic study known as animal cognition. Just how do animals process information from the environment around them? How complex are their mental images or conceptualizations of situations? Over the last few years the systematic study of animal cognition has become one of the hot topics in comparative psychology in general.  While it's not easy to find much scientific research on the horse, there is a little older work and some that's fairly recent.

Early work on horse cognitive and learning abilities included classic studies of perception, simple pattern discrimination (triangles, squares, circles), maze learning, and memory.  There is a great article published in 1990 by Cindy McCall, Ph.D., who is now at Auburn University (Journal of Animal Science, Volume 68, pp. 75-81). That paper reviews the work with horses up until 1990.

Some of these abilities (perception, pattern discrimination, etc.) could be explained as simple stimulus-response, associative learning.  That really requires very little higher cognitive ability that would fit the definition of "thinking." So while both horses and humans use those skills to learn and respond to their environment, the questions remained about how complexly horses think or understand.

Nonetheless, the simple learning and perception research is very interesting. Most people find it fun to know what horses can do and how they compare to people or to pigeons, rats, dogs, or dolphins.

For example, the research suggests that in general, horses are very good at many of the simple associative tasks.  You might make an argument that on certain tasks horses are quicker than people, and they can pick up on subtle cueing in ways that people interpret as "brilliant." Other practical findings have been that horses learn more quickly with positive reinforcement as opposed to negative reinforcement, and much better with reinforcement than with punishment.

One of the best-known researchers working now on cognition specifically in horses is Evelyn Hanggi, Ph.D., president of the Equine Research Foundation in Santa Cruz, Calif. (  Her work goes a bit beyond the simple associative learning abilities to what might be called a somewhat higher level of cognitive function. Specifically, some of her work has focused on concept formation in horses. She has done a number of simple experiments in a few horses trying to determine whether any horse can demonstrate the ability to form and apply concepts. She has looked at the simple concept of open versus filled two-dimensional stimulus objects.  The study designs are beautifully simple.

The horse is exposed to a stimulus, say, for example a panel depicting two images--one of an open circle and one of a filled circle. If each time the horse touches the open circle it gets a food treat and each time it touches the filled circle nothing happens, the horse will soon start going immediately to the open circle and avoiding the filled circle.  The stimulus panel is presented over and over with the images in random left and right order and with all sorts of attempts to control any inadvertent cueing for the "correct" stimulus.

Once the horse is performing very well (always touching the "correct" stimulus), he is shown shapes other than the circle, each with an open and a filled example. So now there might be an open and a filled square or an open and a filled triangle. If the horse immediately responds correctly, there is evidence that the horse understands and has generalized the concept of open vs. filled to the different shapes.  The horse did respond correctly.

In earlier work in California published in 1994 (Journal of Animal Science, Volume 72, pp. 3080-3087), Brenda Sappington and Larry Goldman did a similar experiment in which one of four horses learned to respond generally to triangular patterns, as opposed to patterns with right angles or circular edges, both two or three-dimensional.  This suggested that horses can form and use the concept of triangular shapes.

Still, everything that has been done scientifically in the horse so far addresses tasks, learning, and conceptualization at a fairly simple cognitive level, at least by human standards of thinking.  Almost everyone who knows and works with horses would likely have plenty of anecdotal evidence "demonstrating" that horses readily do these simple mental tasks and perhaps much more.  The challenge for scientists is to set up experiments to demonstrate the abilities in a manner that can stand up to scientific scrutiny.  And that's where we are at the moment with horse thinking.

About the Author

Sue McDonnell, PhD, Certified AAB

Sue M. McDonnell, PhD, is a certified applied animal behaviorist and the founding head of the equine behavior program at the University of Pennsylvania's School of Veterinary Medicine. She is also the author of numerous books and articles about horse behavior and management.

Stay on top of the most recent Horse Health news with FREE weekly newsletters from Learn More