Before we define the limits of horse cognition, we should look at the environments in which we expect horses to demonstrate it.
What can horses understand? How well can they solve problems? Can they plan, learn concepts, make deliberate choices or communicate with us?
Questions about horse intelligence and cognition fascinate me. But I realised that before asking what horses are capable of, there is something else we need to consider:
What opportunities have we actually given horses to develop and show these abilities?
There is still a great deal we don’t know about horse cognition. Research can tell us what horses have demonstrated under particular experimental conditions. That is valuable. But failing to demonstrate an ability is not necessarily evidence that the ability is absent.
A review of research on horse cognition and learning also highlights how performance in learning studies can be influenced by external factors, and how much remains to be investigated.
The environment in which learning takes place matters too.
The world we create for horses
Most domestic horses live in a world largely organised by humans.
We decide where they live, who they live with, when and what they eat, when gates open, which objects they can interact with and where they can go. We decide when training begins and ends. And during those interactions, we largely determine when the horse is expected to act, what he is expected to do, and which behaviours are allowed or discouraged.
As a result, many important aspects of a domestic horse’s daily life are determined by humans, with limited opportunities for the horse to influence them.
That matters when we think about cognition.
Skills do not develop in a vacuum. An animal needs opportunities to explore, gather information, try things, make mistakes, discover relationships between actions and outcomes, and use what he has learned in new situations.
I am able to sit here and write this blog after more than 18 years of dedicated parenting and formal education — followed by decades of experience, practice and continued learning. Take away those opportunities from a human, and we would demonstrate far fewer of the abilities we consider typically human.
Yet when we look at horses, we often try to define what they are capable of based on animals who have had comparatively few opportunities to learn, practise and build cognitive skills.
If those opportunities are limited, we need to be careful about the conclusions we draw from the abilities we subsequently observe.
Welfare is more than meeting physical needs
This is also where cognition and horse welfare begin to overlap.
The Five Domains Model describes welfare through Nutrition, Physical Environment, Health, Behavioural Interactions and Mental State. Domain 4 — Behavioural Interactions — considers how animals interact with their environment, other animals and humans. It specifically incorporates opportunities to exercise agency.
Agency means that an animal can engage in voluntary, self-generated and goal-directed behaviour.
More recent work within the Five Domains framework describes choice, control and challenge, together with developing competence, as important qualities of agency and potential contributors to positive welfare.
And this is now being discussed specifically for horses. In 2026, Lindsay Goodale brought research on agency together in an equine context, examining how opportunities for choice, control and challenge could be applied to improve the welfare of domestic horses.
That changes the way we can think about enrichment.
Enrichment creates opportunities to learn
Every interaction with the environment can become part of a horse’s learning history.
When a horse can investigate objects, manipulate them, solve problems and discover that his actions produce predictable outcomes, he gains experience he can use again. A movement learned while solving one problem may become useful in another. Understanding cause and effect creates possibilities for more complex interactions. With experience, the horse develops a growing repertoire of skills and knowledge to draw on.
This is where enrichment becomes particularly interesting to me: what a horse can do tomorrow depends partly on the opportunities we give him today.
At Equithink, we deliberately build on that principle. We create opportunities that allow horses to develop skills and understanding over time, and then use what they have learned in increasingly complex ways.
How we put this into practice
Learning builds from simple to more complex skills. So at Equithink, we are creating a learning path that makes learning fun and gives horses opportunities to build on what they already know.
The Spinner is the first step. Horses learn to gently manipulate an object with their muzzle, explore different movements and discover that their actions can produce predictable results.
From there, we gradually add new puzzles, movements and challenges. Skills and understanding developed along the way can then become useful in completely different interactions. You can explore the learning path and the different puzzles we use to build these skills in our guide (click here).
A button can operate a light. Another can control music. The horse can increasingly interact with parts of his environment and influence what happens around him.
And we want to take that much further.
We are beginning to explore whether horses can learn consistent symbols that refer to their environment, activities, experiences and other concepts that are meaningful to them.
That raises fascinating questions about horse cognition and communication.
But before asking what horses can ultimately learn to communicate, I think we need to build the foundations that make such communication meaningful.
What are we actually measuring?
When we ask a horse to solve a cognitive task, we are not testing a brain that developed independently of its environment. We are testing an individual with a learning history.
The opportunities he has had to explore, solve problems, make choices, manipulate objects and influence his surroundings have helped shape the skills and strategies he brings to that task.
So when a horse does not demonstrate a particular ability, the result may tell us something about what he can do under those conditions, with that particular learning history. It does not automatically tell us where the limits of horse cognition lie.
That distinction matters. Especially when the horses we study have spent much of their lives in environments where humans determine what happens, when interaction takes place and what responses are expected from them.
And there is another part of that learning history we should consider: has the horse learned that his actions matter? Has he experienced that making a choice, initiating an interaction or giving an answer can actually change what happens next?
The abilities a horse brings to a cognitive task may therefore reflect not only what he has learned, but also what he has learned is worth doing.
I explore this question further in [Why would a horse communicate if his answer doesn’t matter?]
So, what could horses become capable of?
The most interesting discoveries are still ahead of us. We are only beginning to create the kinds of learning histories that can show us what horses become capable of when they are given years of opportunities to learn.
If we give horses more opportunities to explore, learn, solve problems, make choices and influence their environment, we are creating learning histories that have rarely existed for domestic horses.
A horse who has spent years building skills, discovering relationships, making choices and learning that his actions matter will approach new problems with a very different learning history from a horse without those experiences.
That is what I want to explore with Equithink.
We are already taking the first steps into communication, while continuing to develop increasingly complex puzzles and responsive environments. From there, I want to explore educational games, touchscreens and computerised puzzles that allow horses to keep learning — and allow us to study how that learning develops over time.
And ultimately, I would love to see horses around the world following that learning journey. Not all doing exactly the same things, or progressing at the same pace, but giving us something we currently have very little of: the opportunity to see what happens when horses are given years of progressively richer cognitive experiences.
If a horse gets the opportunity to go through kindergarten, what skills will he have developed by the time he reaches Horse University? And what might he be capable of then?
I don’t know.
That’s exactly why I want to find out.
And you can be part of that journey.
Give your horse opportunities to explore, solve problems and build new skills. Start with the Spinner and discover where learning can take you both.
[Explore the Equithink learning path →]
And follow Equithink as we take the next steps towards communication, educational games, computerised puzzles and Horse University.
References & further reading
Brubaker, L. & Udell, M.A.R. (2016). Cognition and learning in horses (Equus caballus): What we know and why we should ask more. Behavioural Processes, 126, 121–131. The review covers human–horse interactions, social learning and independent learning, and specifically notes the influence of external factors on horses’ performance in learning studies.
Read the PubMed record
Mellor, D.J., Beausoleil, N.J., Littlewood, K.E., McLean, A.N., McGreevy, P.D., Jones, B. & Wilkins, C. (2020). The 2020 Five Domains Model: Including Human–Animal Interactions in Assessments of Animal Welfare. Animals, 10(10), 1870. The paper explains the revised Behavioural Interactions domain and its relationship to animals’ opportunities to exercise agency.
Read the open-access paper
Littlewood, K.E., Heslop, M.V. & Cobb, M.L. (2023). The agency domain and behavioral interactions: assessing positive animal welfare using the Five Domains Model. Frontiers in Veterinary Science, 10, 1284869. This review develops the role of competence-building agency and the qualities of choice, control and challenge within positive welfare assessment.
Read the open-access paper
Goodale, L. (2026). Opportunities for agency in domestic horses: Applying the behavioural domain to increase equine welfare. Animal Welfare, 35, e26. This recent equine-specific paper brings together research on horse management and agency and considers applications of choice, control and challenge to domestic horse welfare.
Read the open-access paper
