When The Horse Says No: Rethinking Problem Behaviour In Sport Horses

This article addresses the shifting paradigm in modern equestrianism regarding unwanted horse behaviours, traditionally labelled as “naughtiness”, “evasion”, or “disobedience”. These labels place the source of the problem in the horse, and they have shaped practice accordingly, as the conventional response to a horse deemed naughty or difficult has usually been correction, restraint, or escalating pressure. Evidence from the Ridden Horse Pain Ethogram, developed by Dr Sue Dyson and colleagues across a multi-phase study of over 400 horses, demonstrates that many behaviours commonly attributed to attitude or character are, in fact, expressions of musculoskeletal pain — and that this connection is causal, not merely coincidental.

Moving away from historic, coercive “putting them in their place” methodologies, this piece explores how a scientifically grounded approach to listening to the horse can be implemented within high-stakes environments such as elite show jumping and racing. By reframing misbehaviour as conflict behaviour driven by pain, confusion, or stress, the article argues that integrating equine ethology and learning theory is the definitive path to safeguarding the welfare of the equine athlete while concurrently supporting performance.

By Debora Cinganelli, Equine Neuroethologist & Welfare Consultant

Published June 5, 2026

The modern crossroads of tradition and science

The vocabulary of equestrian sport has never been short of words for horses who do not comply: words like “evasion”, “resistance”, “disobedience”, and more colloquially, “naughtiness” have been stable features of equestrian language for generations, appearing in training manuals, competition rule books, and everyday yard conversation alike. These are terms that locate the problem firmly in the horse: they imply wilfulness, attitude, a character defect to be overcome. And because language shapes perception, they have also shaped practice. The conventional response to a horse labelled difficult has been correction, restraint, or escalating pressure, on the assumption that if the horse is refusing, the answer is to insist more firmly.

That framework sits increasingly at odds with what the scientific literature now tells us about equine behaviour, cognition, and pain. A growing body of peer-reviewed research points in a different direction as many behaviours long interpreted as defiance or disrespect are, in fact, a form of communication, and quite often the communication is about discomfort or pain.

Acknowledging this does not require dismantling equestrian tradition; it requires the willingness to expand what that tradition recognises as expertise.

Deconstructing the “naughty” myth

In a 2025 commentary published in Animals, O’Connell, Dyson, McLean, and McGreevy argued directly against the continued use of terms such as evasion, resistance, and disobedience in human-horse interactions. These labels, the authors contend, fail to recognise current knowledge of the horse’s nature and mental abilities, inadvertently placing the horse at fault without acknowledging the many mechanisms that may be affecting or motivating the behaviour shown. Their proposed redefinition of “conflict behaviour” reflects a genuinely multidisciplinary understanding: unwanted responses exist on a continuum from subtle to overt, driven by competing motivations that may be biological, environmental, or anthropogenic in origin.

This is not just semantics: within the equestrian industry, there has been a tendency to normalise coercive corrections of problematic behaviour and to ignore or overlook the motivations behind it (O’Connell et al., 2025). When a horse’s unwillingness is attributed to character, the underlying cause goes uninvestigated. The horse continues to work, the discomfort continues, and the behaviour, rather than resolving, often escalates or becomes learned.

In 2010 McLean and McGreevy identified a related problem in training contexts: techniques that defy the principles of learning theory frequently generate conflict behaviours precisely because they create confusion or pain rather than clear communication. When a horse cannot predict the consequence of its actions, or when pressure is applied without logical release, the resulting anxiety can manifest as exactly the kind of resistance that is then punished further. The cycle is self-reinforcing and, without a change in perspective, difficult to interrupt.

The high-stakes paradox

The equestrian disciplines in which unwanted behaviours carry the highest immediate risk, show jumping and racing, are also those where the pressure to suppress, manage, or simply push through such behaviours has historically been greatest. The competitive calendar in both disciplines is demanding, the financial stakes are significant, and the culture of toughness is deeply embedded. A horse that loses form across a season, that becomes difficult to load, that backs off at a particular type of fence, or that grows progressively harder to handle on the gallops may be managed with equipment changes, altered regimes, or stronger riding before anyone has asked what the horse itself might be trying to communicate.

This environment creates what might be described as a high-stakes paradox: the contexts in which accurate reading of equine signals is most consequential for welfare and arguably for performance are the same ones in which the time, culture, and tools for doing so have been least available. The horse that is uncomfortable but still winning may receive less scrutiny than one that is comfortable but underperforming. Yet the evidence reviewed in this article suggests that these categories are not as separate as they appear.

The science of listening: the Ridden Horse Pain Ethogram

The most substantial scientific contribution to this area comes from the work of veterinary researcher Dr Sue Dyson, whose Ridden Horse Pain Ethogram (RHpE) has transformed how musculoskeletal pain can be identified in the ridden horse. Developed through a multi-phase study spanning three years and involving over 400 horses, the RHpE catalogues 24 behaviours that research has consistently associated with pain during ridden work (Dyson et al., 2018). The majority of these behaviours are at least ten times more likely to be observed in lame horses compared with non-lame horses, and the observation of eight or more of the 24 behaviours is considered likely to reflect the presence of musculoskeletal pain (Dyson, 2022). Crucially, the causal relationship between pain and behaviour was confirmed through diagnostic anaesthesia: a marked reduction in RHpE scores following resolution of lameness proves that the connection is causal, not coincidental.

The 24 behaviours catalogued in the RHpE are not dramatic: they are not bucking or rearing, they include ears held back for more than five seconds, a glazed or intense stare, repeated tail swishing, hindlimb toe drag, and the front of the head carried behind the vertical. These are signals that most riders will have encountered and, in many cases, either disregarded or attributed to temperament. The science suggests they warrant considerably more attention.

When the RHpE was applied to horses competing at British Eventing one-day events across levels from 90 cm to Novice, a consistent association emerged between pain ethogram scores and competition results: horses placing in the top three showed markedly lower scores than other finishers, and a positive correlation was found between dressage penalty scores and RHpE scores at all levels (Dyson & Pollard, 2022). The horse that is struggling physically, it turns out, is frequently also the horse that is underperforming. Performance and comfort are not separate questions.

A parallel tool, the Horse Grimace Scale (HGS), developed by Dalla Costa and colleagues (2014), extends this evidence base to facial expression. Validated initially in horses undergoing surgical castration, the HGS identified six facial action units that reliably indicate pain state. Taken together with the RHpE, these validated instruments represent a shift away from intuition-dependent assessment towards repeatable, evidence-based evaluation that can be applied consistently across professional contexts.

It is important to note a current limitation: as of the most recent published review of the RHpE (Dyson, 2022), the ethogram had been validated specifically for horses performing dressage-type movements, trained to carry the head in a vertical position. Its formal application to horses while jumping, or to racehorses, had not yet been established. This is a research gap worth naming directly, but one that does not diminish the underlying principle. The biology of pain and its behavioural expression does not differ between disciplines; what differs is the availability of validated, discipline-specific tools to assess it.

The conditioning of conflict

One of the most practically significant findings in this body of research concerns the continuum between subtle, easy-to-miss pain signals and the dramatic, dangerous behaviours that eventually prompt intervention. Hockenhull and Creighton (2015) documented associations between unwanted ridden behaviours and aspects of management and exercise, highlighting how problems that appear behavioural in origin often have physical or environmental roots. Research on horses with vertebral problems found that those most severely affected were significantly more likely to react aggressively towards humans, while this was not the case for unaffected or mildly affected animals; crucially, most of these horses continued to work because the discomfort was systematically underestimated through lack of objective assessment and through interpretation of behaviour as attitude (Fureix et al., 2010).

The mechanism by which pain-driven behaviour becomes a conditioned response is now well described. When a horse successfully avoids a painful stimulus by unseating a rider, or refusing a fence, or making a handler retreat, the behaviour is negatively reinforced: it worked, so it is repeated. Over time, what began as a direct response to pain can become a learned response that persists even after the original source of pain has been treated (Hockenhull & Creighton, 2015). This means that waiting for overt, unambiguous behaviours before investigating a physical cause is, in many cases, waiting too long. By the time a horse is regularly refusing or displaying dangerous behaviour, the pattern may have been building through subtler signals for months.

Sykes and Lovett (2025) reinforced this point in a recent review examining the range of differential diagnoses warranted when horses are presented with behavioural problems, noting that chronic pain can produce a negative emotional state encompassing fear, anxiety, and distress, and that behavioural traits commonly labelled as grumpy, spooky, or stressy may reflect underlying physical discomfort rather than innate personality, particularly when the change in behaviour is observed over time rather than being a consistent feature from the outset.

Implementing learning theory in elite sport

If the science asks equestrians to rethink the language of naughtiness, it also offers something in return: a coherent, evidence-based framework for a different kind of practice. Equitation science, as articulated by McGreevy and McLean (2010) and developed further through the work of the International Society for Equitation Science, applies the established principles of learning theory (operant and classical conditioning, reinforcement, habituation, and sensitisation) to the training and management of horses. The goal is both a more ethical approach but also a more precise and effective one.

Within this framework, behavioural auditing becomes a professional skill rather than an optional refinement. This means treating any change in a horse’s rideability, responsiveness, or demeanour as a clinical sign first and a training problem second. It means understanding that a horse which was previously willing and is now resistant is offering information, and that the first question should not be “how do we fix this” but “why is this happening”. It means equipping trainers, riders, and handlers with the observational vocabulary to recognise the early, subtle signs catalogued in tools such as the RHpE before they progress to something harder to address.

This does not require abandoning the demands of competitive sport: elite performance and equine welfare are not opposing values, and also the evidence reviewed here suggests they are, in practice, aligned. A horse competing without musculoskeletal pain and without the confusion and anxiety that poorly applied training can generate is a horse more likely to perform consistently and to remain sound. The investment in listening is a practical advantage.

What changes at the level of daily practice is, in many respects, a question of culture as much as method. It requires creating environments in which expressing concern about a horse’s signals is normalised rather than perceived as weakness or inexperience. It requires trainers and riders who are willing to pause before escalating pressure, to call a veterinarian before calling a trainer when a horse’s behaviour changes, and to understand that a horse that says no is almost always saying something specific.

Fostering innovation by respecting tradition

The history of equestrian sport contains a great deal of genuine expertise: accumulated knowledge of how horses move, how they learn, and how they respond to pressure and release. That tradition is worth respecting. What the science of the past two decades has made possible is not its replacement but its refinement, so the addition of validated, repeatable tools and an evidence-based understanding of equine pain and communication that traditional horsemanship, for all its virtues, could not systematically provide.

The horse that pins its ears during tacking up, that drifts at a particular fence, that loses its characteristic forward willingness over the course of a season, is communicating.

Now we have tools available that allow us to take this communication into account and using these tools to integrate “listening to the horse” in day-to-day practice is paramount to equine welfare and it is both the scientifically sound and the ethically necessary direction of travel for elite equestrian sport.

---

Debora Cinganelli is an equine neuroethologist and welfare consultant with a multidisciplinary background spanning ethology, animal physiology, and equine nutrition. She holds an MSc in Ethology, specialising in neuroethology.

She is a member of the International Society for Equitation Science (ISES), the International Society for Applied Ethology (ISAE), and the International Society for Neuroethology (ISN).

Her work centres on the integration of behavioural neuroscience, physical health, and nutrition into a coherent, evidence-based approach to equine welfare in management and training.

Through her practice, Path to Bloom, she offers behavioural and management consultations, educational programmes, and science communication aimed at translating peer-reviewed research into accessible, practical knowledge for horse owners and practitioners.

Learn more about her work here or via her Instagram account @pathtobloom.equine.

---

References

Dalla Costa, E., Minero, M., Lebelt, D., Stucke, D., Canali, E., & Leach, M. C. (2014). Development of the Horse Grimace Scale (HGS) as a pain assessment tool in horses undergoing routine castration. PLoS ONE, 9(3), e92281. https://doi.org/10.1371/journal.pone.0092281

Dyson, S., Berger, J. M., Ellis, A. D., & 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 Behavior, 23, 47–57. https://doi.org/10.1016/j.jveb.2017.10.008

Dyson, S., & Pollard, D. (2020). Application of a Ridden Horse Pain Ethogram and its relationship with gait in a convenience sample of 60 riding horses. Animals, 10(6), 1044. https://doi.org/10.3390/ani10061044

Dyson, S., & Pollard, D. (2022). Application of the Ridden Horse Pain Ethogram to horses competing in British Eventing 90, 100 and Novice one-day events and comparison with performance. Animals, 12(5), 590. https://doi.org/10.3390/ani12050590

Dyson, S. (2022). The Ridden Horse Pain Ethogram. Equine Veterinary Education, 34(7), e385–e396. https://doi.org/10.1111/eve.13468

Fureix, C., Pagès, M., Bon, R., Lassalle, J. M., Kuntz, P., & Gonzalez, G. (2010). Partners with bad temper: Reject or cure? A study of chronic pain and aggression in horses. PLoS ONE, 5(8), e12434. https://doi.org/10.1371/journal.pone.0012434

Hockenhull, J., & Creighton, E. (2015). Unwanted oral investigative behaviours in horses and associations with reported management and ridden exercise. Journal of Veterinary Behavior, 10(1), 62–67. https://doi.org/10.1016/j.jveb.2014.10.004

McLean, A. N., & McGreevy, P. D. (2010). Horse-training techniques that may defy the principles of learning theory and compromise welfare. Journal of Veterinary Behavior: Clinical Applications and Research, 5(4), 187–195. https://doi.org/10.1016/j.jveb.2010.04.002

McGreevy, P. D., & McLean, A. N. (2010). Equitation Science. Wiley-Blackwell.

O’Connell, E., Dyson, S., McLean, A., & McGreevy, P. (2025). No more evasion: Redefining conflict behaviour in human–horse interactions. Animals, 15(3), 399. https://doi.org/10.3390/ani15030399

Sykes, B., & Lovett, A. (2025). Can all behavioral problems be blamed on equine gastric ulcer syndrome? Animals, 15(3), 306. https://doi.org/10.3390/ani15030306