A number of months ago when I was bringing my horse in from the field, I saw to my horror that she was hobbling on three legs as she approached. As she limped into the stable, someone remarked that it was probably not serious as we all knew that my horse was prone to theatrics. Another comment referred to the overly-dramatic nature of chestnut mares.

I was curious about this line of thinking. Is it true that some horses make much ado about nothing? Are some horses wimps, and others soldiers? Or do they simply feel pain differently?

I think it is unlikely that horses are feigning their injuries to shirk their responsibilities. The literature on humans describes pain as a multifaceted event with biological, psychological, social, cultural, and contextual elements (Myers & DeWall, 2020). The phenomenon of phantom limb pain, where a person feels pain in an amputated limb (phantom limb sensations have even occurred in those born without that limb (Melzack, 1998)) demonstrates that pain is much more complex than a neural impulse sent to our brain. The same pain-inducing stimulus will not be experienced by all people at the same intensity, or even by the same person in different situations. Think of athletes who compete through excruciating pain without feeling compromised until the game is over, or a severely injured victim of a disaster who charges ahead to save others and feels little pain until they reach safety.

Horses have been accused of ‘faking’ their lameness as they set out on a trail ride because the limp miraculously resolves itself once the animal is turned back to the stable.

Horses’ experiences of pain are also mediated by context. Horses have been accused of ‘faking’ their lameness as they set out on a trail ride because the limp miraculously resolves itself once the animal is turned back to the stable. However, when the physiological source of the lameness has been diagnosed and treated, the faking behaviour disappears, suggesting that horses probably do not hatch devious plans about how to call in sick for work. The more probable explanation for this sudden recovery is that the extra adrenaline boost associated with a return to the stable and rest is sufficient to temporarily override the protective function of favouring an injury (McDonnell, 2005).

Determining how much pain our horses are suffering is challenging even for professionals. Lameness, our typical indicator of pain severity, is a crude instrument for measuring the extent or intensity of a horse’s pain. Many of us know of extremely successful sport horses who should not be standing upright based on their imaging results, and science too has recorded this discrepancy. In a study where seasoned equine practitioners assessed the degree of lameness in a sample of horses with varying issues, and then blindly assessed actual tissue damage in the same (but unidentified) horses’ ultrasounds and radiographs, there was no relationship between the horses’ observed lameness score and actual tissue damage (Ijichi, 2014). Since lameness is routinely used to assess the progress of a condition, potential treatment, evaluation of interventions, and to inform decisions about euthanasia, Ijichi’s findings (and lack of findings!) are noteworthy.

The advent of the “Lameness Locator”™ objectively quantifies a horse’s body movement asymmetry using small body-mounted sensors and a hand-held tablet. Undoubtedly the seller’s nightmare in a pre-purchase exam, it has been found to be helpful in diagnosing subtle, multiple-limb, or compensatory lameness provided the practitioner has the necessary skill to interpret the output.

Recently, researchers have created pain scales to more accurately assess the degree and intensity of equine pain (e.g. Post Abdominal Surgery Pain Assessment Scale (PASPAS) Gaubner et al, 2011; Composite Pain Scale, (CPS) Bussieres et al., 2008; Equine Utrecht University Scale for Composite Pain Assessment (EQUUS-COMPASS) and Facial Pain Assessment (EQUUS-FAP) van Loon & Van Dierendonck, 2015). Equine researchers de Grauw and van Loon comment that a pain scoring system will only work in practice if it is easy to use, with relevant and well-defined descriptors of expressions or behaviours, and where assessments can be made with different observers with consistent results. The scale needs to be at least sufficiently sensitive to distinguish mild, moderate, or severe pain, and not so complex that it dissuades potential adopters or where extensive training is required for reliable scoring (de Grauw & van Loon, 2016).

The Horse Grimace Scale

One such scale that has been gaining traction is The Horse Grimace Scale (HGS, Dalla Costa et al., 2014) which objectively measures equine pain solely from facial expression. Costa’s group analyzed horses’ facial expressions before and after undergoing a castration and rated horses on six facial expressions (e.g. positioning of the ears, and muscle tension around eyes, mouth, face and nostrils, etc. see figure 1). Horse people were able to correctly categorize post-surgery horses from horses who had not undergone surgery through photographs of only the horse’s heads.

Figure 1 – The Horse Grimace Scale.


Similar studies have also proved the HGS effectiveness in other pain situations, where horses suffering from laminitis and acute and chronic orthopaedic pain could be identified from controls based solely on photographs of facial expression.

In a sample of 64 sport and riding school horses in regular work and presented by their owners to be sound and working comfortably, Dyson found that 73% of horses were lame in one or more legs…

Ridden Horse Pain Ethogram

In a three-year, six-phase study assessing over 400 horses, equine practitioner and pain specialist Dr. Sue Dyson and her colleagues developed the Ridden Horse Pain Ethogram (RHpE), also known as the Ridden Horse Performance Checklist, to reliably predict musculoskeletal (MSK) pain that may not be obvious to even an experienced eye. The RHpE identifies 24 behaviours of a horse under saddle, which are often labeled as disobedient, disrespectful, or dishonest (such as reluctance to go forward, ears back, tail swishing, head tossing, bucking and so on).

Dyson maintains that many behaviours that we have come to see as normal, or as a character defect, are more likely manifestations of pain. She notes that even behaviours exhibited while being tacked up (teeth grinding, tail swishing, pawing, etc.) could be a reaction to the anticipation of the pain that is to come when under saddle. As Dyson comments for this article, “the equine fraternity has become ‘normalized’ to many of these adaptations … How many times have I heard ‘my horse does not like dressage, but he jumps well so he must be OK’ without asking why the horse ‘does not like’ dressage.”

Dyson has found that almost all of these behaviours (see chart below) are at least 10 times more likely to occur in lame horses than in sound ones (2018). She notes that if the score is greater than 8/24 the horse is experiencing MSK pain, but that some lame horses have a score less than 8/24. She adds that “the total score does not necessarily reflect the degree of discomfort experienced by a horse, which is extraordinarily difficult to quantify, especially if there is more than one source of pain [which is] very common).”

Dyson, S., Berger, J., Ellis, A., Mullard, J. (2018) Development of an ethogram for a pain scoring system in ridden horses and its application to determine the presence of musculoskeletal pain. Journal of Veterinary Behaviour: Clinical Applications Research, 23, 47-57.


In a sample of 64 sport and riding school horses in regular work and presented by their owners to be sound and working comfortably, Dyson found that 73% of horses were lame in one or more legs (four were excluded as their lameness was sufficiently severe as to be a concern for further testing), 47% had gait abnormalities in the canter, 37% were both lame and showed canter gait abnormalities, 58% had evidence of back pain, and the RHpE median score was 9/24, where a score of 8 or higher is indicative of a horse suffering from MSK pain. The most convincing evidence for the RHpE is that when Dyson is able to temporarily terminate the MSK pain through diagnostic anaesthesia, or nerve blocks, and the horse is ridden again, the behaviours disappear.

Given that these horses were presented as sound by their owners suggests that owners are not particularly good at detecting lameness in their own horses, especially when they are lame in more than one limb. Indeed, when horses are experiencing multi-limb lameness, the American Association of Equine Practitioners’ lameness grading system typically used by equine veterinarians is notoriously inaccurate (Dyson, 2022). However, the behavioural indicators of the RHpE may be more easily recognized by an owner or trainer than a low grade, or multi-limbed lameness, and could therefore help with an early detection of an underlying pain-related problem.

Dyson comments that the display of discomfort depends on multiple factors including, but not limited to, the horse’s individual pain threshold, horses’ variations in their behavourial expression of pain, rider skill level, athletic demand placed on the horse, saddle and tack fit, horse fitness and coordination, riding environment, and horse management (turnout, access to forage, and social interaction).

Wimps or soldiers?

Being more sensitive to pain than another horse is not a character flaw, but a real and lived experience. Indeed, there is no logical reason for a horse to exaggerate their pain. In fact, quite the opposite is true. Horses, as prey animals, are evolutionarily designed to minimize their suffering and make accommodations to mask their lameness. Horses who showed their weakness were more likely to be tonight’s dinner for the next predator, and thus have no opportunity to pass those wimpy genes on to their offspring and to future generations.

Horses have an array of adaptations to mask obvious lameness and reduce discomfort: for example, a shortened length of step, lower height in the arc of the step, a shortened proportion of stride time that the limb is bearing weight (or in multi-limbed lameness, a sharing of loads among limbs for longer periods of the stride), reduced range of motion in the back and tail, and altered head and neck positioning (Dyson, 2022)

If we have a horse who seems to be louder than another about their pain, we should consider ourselves lucky, since pain detection in horses is such a tricky business. Dyson notes the factors that contribute to a horse’s display of discomfort is what her team has endeavoured to identify in their research, but that “it remains a difficult subject to address objectively.” She comments, “It is clear from all our published competition data that the presence of pain does adversely influence performance. However, some supremely talented horses that are well-ridden and have huge scope and trust will continue to jump well despite underlying pain.”

I would argue that the horse who does not do so is hardly a wimp, but experiences this pain differently; their pain is legitimate, and we may have to do some digging to get to the source. Rather than a wimp, this horse should be lauded as the top performer of the year in “Equine-to-Human Communications”!


Sue Dyson has turned her groundbreaking research into a film entitled “24 Behaviours of the Ridden Horse in Pain” accessible for free here. The film documents the story of Lauren McMahon and her beloved show horse Galina, who though not obviously lame, was showing a deterioration in performance in the show ring. I highly recommend this informative and moving documentary!