This past spring I retired my old dressage horse, handed over the reins to a hunter/jumper professional at our stable to retrain him as a hunter, and brought in my new dressage horse to the stall next door. Although I expected my old horse to feel somewhat miffed now that he was one on a list of many instead of my number one guy, I was convinced that he was insufficiently cognitively sophisticated to experience real jealousy. That was the stuff of anthropomorphic mumbo-jumbo, the dangers of which I have cautioned against in many previous articles…right?

But as I watched my formerly interactive and personable horse lunging at any horse that passed by, aggressively kicking the walls of his stall at the new arrival, and/or standing with his head pressed into back wall of his stall in the most heart-breaking apparent depression, I began to wonder. When he finally took an aggressive lunge at me in what certainly seemed to be a desperate effort to gain my attention, and later used his head and chin to draw me into his chest and hold me there, this cynical scientist decided to rethink her assumptions.

Secondary Emotions in Animals

Up until recently, most psychological researchers believed, as did I, that animals and humans may share primary emotions such as happiness, sadness, anger and fear, but that secondary emotions, such as pride, guilt, shame and jealousy, require self-consciousness, self-reflection, and an understanding of the conscious intentions of others, and that these qualities reside solely with humans and possibly some primates and cetaceans. Cognitive ethologists (those who study the mental capacities, emotions and motivations of animals) challenge the assumption that secondary emotions suddenly appeared in humans without precursors in other animals (e.g. Bekoff, 2002). And, recent research is indicating that this may be so.

Jealousy has been predominantly studied in romantic couples where cognitive appraisal of the threatening situation is key (Is she more desirable than I am? Is she going to be alone with him? Will he leave me?) But social psychologists, Christine Harris and Caroline Prouvost, from the University of California, reasoned that jealousy may well have a primordial form that could be triggered without complex cognitive understanding about the self, or appraisal about the meaning of the interaction. They note several studies demonstrating that infants as young as six months will fuss when their mothers pay attention to a realistic-looking doll, but not when mothers turn their attention to a book.

The researchers replicated these infant studies with dogs and discovered that dogs react similarly. When owners ignored their own dogs to play with a realistic looking stuffed dog, most of the dogs showed behavioural responses that even hard-core scientists would have to call jealousy. They pushed and nosed and positioned themselves between the toy dog and the owner. Some even tried to attack the offending stuffed interloper. This behaviour was significantly reduced when the owners ignored their dogs to play with a set of plastic pails or to read a book. Harris and Prouvost argue that this primordial jealousy may have evolved in social species (such as dogs) with multiple dependent young who compete for protection, food and affection from their mothers. Similarly, jealousy might arise in animals that require cooperation from other group members for survival and in which alliances are formed, and can be threatened by rivals. This last argument speaks to the possibility of jealousy existing in horses that form enduring bonds within a herd.

Jealousy or Resource Guarding?

The next step – whether horses feel jealous of a rival who may threaten the bond they share with their human owner – is a sizable leap. In a survey of 907 pet owners, Morris and colleagues (2008) found that owners of all types of pets reported their animals exhibiting primary emotions (interest, curiosity, fear, joy, affection, surprise, anxiety, sadness and anger), but only dog and horse owners reported the occurrence of secondary emotions such as jealousy, guilt and pride. Of 69 horse owners, 79 per cent of them reported that horses felt jealous, although the specific contexts in which this jealousy occurred, or whether a horse or human relationship was being threatened, was not explored.

Resource guarding – guarding valued possessions from humans or other animals – is common behaviour for dogs, ranging from relatively benign behaviour such as running away with a desired toy, to more aggressive behaviours such as growling at an approaching person, chasing or biting. Resource guarding makes good evolutionary sense; hanging on to valued items such as food, mates and desirable living spaces makes for good reproductive success and the maximizing of one’s gene pool. However, it tends not be particularly desirable in our domestic companion animals, particularly when the guarding is directed at people. Although, resource guarding is generally conceived of as protecting a valued resource from a human, it is plausible that the human, as the provider of food and attention, may become the valued resource worth guarding. In this case, we might see horses guarding “their human” from other horses.

If you think your horse is guarding you as a valued resource, a solution would be to desensitize and counter-condition the approach of a rival horse or human by providing something desirable any time the rival is present. As identified in the Merck Veterinary Manual (, punishment should be avoided. You want only wonderful things to be associated with the presence of the rival.

Potentially, when horses have limited or no opportunity to form attachments with other horses, jealous feelings might arise in relation to a human owner if a horse feels those bonds are being threatened by a usurper. In order for a horse to feel jealous, however, there needs to be an attachment relationship that he is trying to protect. And when it comes to horses feeling this enduring bond for one particular and specific human…well, this is where it gets tricky.

Attachment Defined

Attachment, as coined by John Bowlby (1980, 1982), evolved as a way of maintaining bonds to a particular other for protection that was seen as critical to an infant’s survival. Attachment behaviours are elicited in times of threat in an effort to maintain and/or regain proximity to the attachment figure. So, infants who stick close to their caregivers do better than those who do not and are more likely to survive to pass on that sticking behaviour to their offspring. Attachment relationships are proposed to exist not only between children and their caregivers, but also in peer relationships with friends, siblings or romantic partners.

Attachment researchers argue that jealousy is simply a manifestation of attachment-related anxiety. When attachment bonds are insecure and anxiety about the threat of losing an attachment figure motivates behaviours (even destructive and counterproductive behaviours) that might regain that connection, jealousy may well raise its ugly head.

The key features of Bowlby’s attachment system are:

  • Proximity seeking: We want to be close to our attachment figures.
  • Safe haven: We seek out our attachment figures for protection in times of threat.
  • Secure base: The attachment figure can be used as source of support that makes us brave about exploring the world.
  • Separation anxiety: We feel anxious and upset when separated from our attachment figure.
  • Irreplaceable: Attachments are to specific others. They are not easily replaced with a substitute.
  • Built over time: Bonds take time to develop and persist over time.

Do Horses Attach to Us?

If we look at these key attachment characteristics, we could make a convincing argument that horses probably attach to each other. And, there is robust literature suggesting that people attach to their companion animals (wanting to be close, using them as a safe haven and secure base and experiencing anxiety when separated). Although there is scant research looking at how humans may attach to horses, presumably horses too could fulfill this safe haven role for their human owners.

Now, revisit that list of key characteristics to argue that a horse might be “attached” to us. Not so straightforward is it? Does my horse (even my jealous horse) always want to be close to me? Does he use me as a secure base from which to explore his environment? Is he anxious when I leave the barn? Or go on holiday? Does he see me as a safe haven in times of threat? Am I irreplaceable? That is a harder sell to this science-dedicated curmudgeon.

In the early years of Bowlby’s formulation of attachment theory he was heavily influenced by the work of Harry Harlow, most famous for his torturous studies on infant Rhesus monkeys. Harlow removed infant monkeys from their mothers and offered them a choice of two “surrogate mothers” – one made of cloth and the other of wire. Harlow was able to demonstrate that tactile comfort, rather than food, was the critical ingredient in infant bonding. Even when their cloth mothers provided no nourishment, infant monkeys would spend almost all of their time clinging to them, only going to their food-providing wire mothers for short feedings. When frightened, infant monkeys always returned to the cloth rather than wire mothers for comfort, even when the wire mothers were their only food source. This tactile or “contact comfort”, the comfortable feeling that infants gain by clinging to a soft attachment figure, was considered by Bowlby to be a critical ingredient in the formation of attachment.

Stroking or wither scratching horses as a way to create this contact comfort has been gaining traction in the recent equine science literature (e.g. DeAraugo et al., 2014). Vigorous wither scratching, that mimics the mutual grooming that horses do with one another, has inherent positive and calming properties: it lowers heart rate, decreases cortisol levels and increases endorphin production. Equine scientist Andrew MacLean has proposed that the soothing contact of wither scratching may be comparable to Harlow’s contact comfort, and thus might just form the basis of an attachment relationship between horse and human (International Society of Equine Science conference proceedings, 2013).

Equine researcher Carol Sankey investigated this notion by comparing food and grooming rewards and their impact upon horse-to-human bonding and learning. Horses were trained using positive reinforcement – either with a food-reward or a grooming-reward (short vigorous, wither scratching) – to stand still for up to 60 seconds without being held or tied. Horses then completed “the motionless person test”, a common measure of animal-to-human connection, where an animal is free to interact with a person standing in the centre of an enclosure. Researchers record the length of time it takes for the animal to approach the person, and the total time the animal remains close.

Not only did Sankey’s food-rewarded horses learn the task quickly and easily (in contrast, only four of the 10 struggling learners from the grooming-reward group managed to achieve mastery), they also seemed to like their trainers better. In the motionless person test, they approached the human more quickly than grooming-rewarded horses and spent more time near her after training than before.

Though not a measure of “attachment”, Sankey’s motionless person test does indicate that whatever horse-to-human bonding may be going on, horses, unlike Rhesus infant monkeys, seem to be motivated more by food than by cuddles. Sankey noted that her grooming-rewarded horses were scratched only briefly, which may have been of insufficient duration to bring about any horse-to-human connecting. Nevertheless, this initial study suggests that if you want your horse to hang with you (and to learn things quickly), be prepared to bring out the carrots.

The Scientist Inches Outside the Box

So, was my horse jealous? I think so. We have been together for 10 years and I have been the main human in his life. He is definitely connected to me if not “attached” by all six of Bowlby’s key features. I believe he understood that the new horse was threatening that connection. Once I started riding him again, the “jealous” behaviours ceased. He is headed to his new home across the country in a couple of weeks to be doted on by his new owner and do an easier job. Will I experience grief upon separation? Undoubtedly! But will he? Although a slight blow to my ego, I’m happy for his sake that I will probably be fairly easily replaced.

How to Boost Your Bonding

Train your horse with positive reinforcement: We train our horses primarily with negative reinforcement. Don’t get hung up by the word negative. Think about negative as in taking something away, rather than negative as in bad. An example would be pressure and release. We apply pressure, when we get the response we want, we release the pressure, and the release tells the horse that this is the response we are looking for.

There are, however, probably more opportunities than we make use of to train horses with positive reinforcement. Simply put, behaviours that are rewarded are more likely to reoccur in the future, so if you see something you like (and initially, even the baby steps of what you eventually want), reward it. (For a more detailed discussion of positive reinforcement training see Horse Canada May/June 2012, or Nearly all good ground manners can be trained and maintained with positive reinforcement: standing quietly for shoeing, injections, oral medications, clipping and mounting, turning and backing up on command, loading into a trailer and so on. The result is not only a pleasant and safer horse, but one that may have developed a stronger relationship to you.

There is now a considerable research that indicates most horses will learn a required task in the designated time frame whether you train them with positive or negative reinforcement. Much like humans, however, horses trained with positive reinforcement learn more quickly, retain the learned tasks longer, experience less stress, react to humans more positively, and are able to generalize this training across trainers, novel tasks, and over longer periods of time (eg. Sankey, 2010).

Be the best option: Do your training at the most boring time of your horse’s day – when eating and greeting times are at a low ebb. You may have to create opportunities where he connects with you for lack of a better alternative. Horses who have around the clock access to other equine buddies need you less than the isolated show horse whose only other contact is with his owner or groom. Think about your horse’s particular social opportunities when you set up your “bonding time”.

Spend more time hanging out: Attachment relationships do not develop overnight. Even human babies take four to six months before they really “get” who they belong to. Have you ever noticed how newborns can be passed around to anyone who wants to hold them, and how singularly unsuccessful this is when they hit about the four- or five-month mark? Hanging with your horse won’t guarantee he’ll attach to you, but not doing so will most certainly guarantee he won’t.

Mood matters: Recent research suggests that horses may be more emotionally tuned in to us than we think, and that this emotional connection may be an important precursor for an attachment relationship. A study by Bridgeman and colleagues from University of Southern Queensland found that 13 of 17 horse-rider pairs showed significant heart-rate synchronization during competition. In fact, the rider’s anxiety levels could be classified into low vs. high with surprising accuracy by looking only at the heart-rate scores of their horses. Higher rider heart-rate was also related to pre-competition riders’ ratings of more anxious and reactive horses and a higher incidence of horse conflict behaviours (bucking, rearing, shying and bolting) during the competition. Correlations do not determine causality, however, nor do they tell us the directionality of the associations. Do anxious riders make anxious horses? Or do riders become anxious as a result of riding reactive horses? The jury is still out, but depending on what emotions we may be bringing with us to the stable, there may be some days when it is better not to ride (Bridgeman, Pretty, & Terry, 2011).

Subscribe to the Horse Canada newsletter and get an exclusive bonus digital edition!
We'll send you our regular newsletter and include you in our monthly giveaways. PLUS, you'll receive our exclusive guide with 85 Tips to Handle Anything You Encounter on the Trail!