Hidden in Plain Sight

A Forensic Behavioural Science Approach to Equine Welfare

Consider a horse presented for a behavioural or health assessment. It stands quietly in its stable, causes no trouble when tacked up, and performs obediently under saddle. By the most commonly applied measures, it looks fine. Yet something does not feel quite right. The horse does not interact or investigate; it seems switched off. Its compliance is too complete, and for those who look carefully, that absence of curiosity is itself a potential cause for concern.

Equine welfare conversations in equestrian circles have historically centred on visible, acute harm: the lame horse withdrawn from competition, the injury caught on camera, the headline neglect case. These are welfare failures we can point to, document, and act upon. Yet there is a growing body of evidence pointing to a different, more pervasive concern: welfare compromise embedded in routine practice, normalised by tradition, and rendered largely invisible by familiarity.

This article explores how the principles of forensic behavioural science can be applied to equine contexts to identify and evaluate precisely this kind of welfare risk. The aim is not to apportion blame, but to promote analytical rigour in assessing equine welfare, in the same way as any other serious evidential process.

What Is Forensic Behavioural Science?

The word 'forensic' derives from the Latin forensis, meaning 'of the forum' or public court. In its modern application, forensic science supports legal and regulatory processes. Standard veterinary forensic work focuses on physical examination and post-mortem analysis. Forensic applied animal behaviour is a newer field that attends specifically to what an animal's behaviour tells us about its internal experience, and provides that evidence in a form that can be used in regulatory and legal contexts.

The foundational principle is straightforward: an animal's behaviour may be considered the gold standard by which mental suffering can be assessed. Behaviour is not merely a by-product of welfare status; it is evidence of it. Applied to equestrian contexts, this approach asks: what does the horse's behaviour, across time and across situations, tell us about how it is experiencing its training and management environment? And does that experience involve chronic negative affect that falls outside any reasonable threshold of acceptable welfare?

Behavioural Scars

The horse presented for assessment today is not a clean slate. It is an animal shaped by everything that has happened to it: every handling experience, every training session, every aversive stimulus and the context in which it was encountered. The residue of that history is frequently visible in behaviour, provided the assessor knows how to read it.

A behavioural scar is a lasting alteration in a horse's response patterns that reflects the neural encoding of a previous aversive experience. Just as a physical scar is evidence that a wound occurred, a behavioural scar is a durable change in behaviour, reactivity, or emotional baseline that persists long after the precipitating experience has ended.

Horses possess a highly developed capacity for associative learning and long-term memory, and research confirms that aversive experiences are encoded with particular durability. Researchers have demonstrated that training history produces effects on a horse's perception of humans that persist for at least five months after training ends. Horses trained using aversive methods showed sustained avoidance of humans; those trained with positive reinforcement displayed lasting increases in contact-seeking behaviour.

The Problem with 'Normal'

One of the most significant challenges in equine welfare is the normalisation of practices that research has demonstrated are likely to cause negative emotional states. When something has been done routinely, across generations, by respected practitioners, the question of whether it causes suffering rarely gets asked. The welfare harm, if present, is hidden not through deliberate concealment but through normalisation.

A forensic behavioural lens disrupts this by applying consistent evaluative standards regardless of how common a practice is. It asks not 'is this usual?' but 'what is the evidence that this causes or risks causing negative affective states, and does the horse's behaviour support that interpretation?' The word 'normal' carries no evidential weight. The question is always animal-centric.

It’s fascinating how the equestrian world has developed a sophisticated vocabulary for reframing potential welfare concerns as training or temperament problems. The horse that resists saddling is called 'cold backed'. The mare that becomes difficult mid-season is showing 'mareish behaviour'. The horse that stands utterly motionless and disconnected under demanding handling has 'a good temperament'. Each of these framings risks closing enquiry that a forensic behavioural approach would explore from a horse-centric perspective.

Behavioural Baseline

Any assessment of a horse's welfare under training must begin with its baseline, the affective (emotional) state it brings to every interaction, shaped by the management environment in which it lives. Horses are social animals with evolved needs for continuous movement, forage intake, and conspecific contact. Stabling for extended periods, restricted forage, and limited social contact are common features of competition/race horse management and are linked with elevated stress responses, the development/performance of stereotypic behaviours such as weaving and crib-biting, and heightened reactivity consistent with chronic anxiety.

Stereotypic behaviours are particularly instructive. Their presence is widely acknowledged as an indicator of previous and/or current welfare compromise, yet they are often managed by preventing their expression rather than addressing the underlying cause. Fitting anti-weaving grilles or removing crib-biting surfaces does not reduce the underlying problem or the horse's motivation to perform the behaviour; it simply prevents its expression, potentially increasing frustration and negative affect.

Fear is a further baseline dimension that warrants careful monitoring. Where fear is a recurrent feature of a horse's experience without adequate resolution, the cumulative welfare cost may be substantial, even if no single event appears obviously abusive. A horse living in a state of chronic low-level fear, but showing few obvious behavioural signs, is not experiencing good or even adequate welfare, regardless of how well it may perform.

Equipment, Pressure and Discomfort

Among the most extensively researched areas of concern is the use of equipment and training methods that rely on aversive pressure. Control of behaviour via use of bits, nosebands, spurs, and training aids are widespread and normalised. Yet research examining the physiological and behavioural responses of horses to these tools is growing, and with that there are clear indications of welfare risk.

What makes these concerns 'hidden in plain sight' is precisely that the equipment involved is standard and the training techniques commonplace.

Nosebands/Bits/Training equipment is sold in catalogues, on the internet, modelled by professionals, and required or permitted under competition rules. There are no restrictions on sales and no knowledge requirements for purchasing or using them. The welfare cost of the use of such equipment, where it exists, probably accrues quietly, experienced by horses who have learned they cannot express discomfort overtly, and they have have no choice.

The Illusion of Compliance

Perhaps the most important concept for understanding how welfare harm can be normalised and invisible is learned helplessness. First described in the 1960s, learned helplessness describes a psychological state in which an animal ceases to attempt to escape or avoid an aversive situation because prior experience has taught it that no response will alter the outcome.

In equine training contexts, there is growing evidence that something functionally similar may occur in horses subjected to training methods that rely heavily on sustained aversive pressure without reliable release. The resulting behaviour, reduced reactivity, apparent passivity, and lack of resistance, is often described as the horse 'accepting' the equipment, training or work.

From a forensic behavioural perspective, however, the same presentation may indicate something concerning such as chronic negative affect and a profound loss of behavioural agency. Research has suggested that horses exhibiting a behavioural profile described as 'dull' or 'shut down' showed elevated stress markers and were less able to experience positive affective (emotion) states. These horses were not in obvious distress; they were just quiet.

Forensic Behavioural Evidence in Practice

Two high-profile cases illustrate the potential use of behavioural evidence, and the gaps that currently remain. An Australian racehorse trainer was prosecuted after surveillance footage captured him applying battery-powered electric shock devices, known as jiggers, to horses on a treadmill. The trainer pleaded guilty to charges of torturing and terrifying thoroughbred racehorses. What is forensically instructive is that jiggers are a tool for deliberate classical conditioning: the shock, paired with specific equipment cues, means that by race day the horse's accelerated performance is not positively motivated, it is conditioned fear. A forensic behavioural assessment could have included those conditioned fear responses, and evaluated their long term impact on the horse’s mental state as part of the evidence of harm.

A parallel case is that of a US showjumper, who received a ten-year FEI suspension in 2021 for repeated use of self-manufactured electric spurs on horses across multiple events and in training, The devices were used routinely, with wires concealed in modified riding boots and a trigger held in the rider's hand during competition. Although these enhanced the performance outcomes, each performance could be viewed as a behavioural scar of the training method used - horses jumping due to fear of pain leading to over-jumping, exaggerated tucking up of legs and other misread symptoms.

Both cases point to a broader challenge: horses conditioned through aversive techniques may display exaggerated or hyperreactive responses that read as athleticism to an untrained eye. For stewards and ground juries, the challenge is to read such patterns not as evidence of a capable or careful horse, but as a question: what has this horse potentially learned to be afraid of?

What a Forensic Behavioural Approach Looks Like

For equestrian organisations and welfare practitioners, adopting this framework means applying structured, evidence-based assessment to welfare as a matter of routine professional practice. This involves several interconnected elements: systematic behavioural observation across time and contexts rather than snapshot assessments; documentation of behavioural history, including changes in response to training or management modifications; evaluation of equipment use against available research on welfare impact; and explicit attention to pharmacological and management histories that might account for the absence of expected behavioural indicators.

Critically, it requires willingness to interpret apparent compliance or passivity not automatically as positive welfare, but as data points requiring further evaluation. The Five Domains Model provides a practical structure for this kind of assessment, paying particular attention to the mental state domain and asking explicitly what the horse's affective experience is likely to be given the totality of its circumstances. The question is not only whether the horse is physically sound but whether it is thriving.

Equestrian Obligations to Horses

Equestrian sport and the wider horse industry currently operate within a social licence that is under increasing pressure. High-profile cases of apparent welfare abuse, combined with growing public awareness of equine behavioural science, have shifted public expectations. Maintaining social licence requires going beyond reactive responses to incidents. It requires the kind of proactive, systematic scrutiny of normalised practice that forensic behavioural science enables.

But the deeper obligation is not to the industry's reputation. It is to our horses. The quiet horse, the compliant mare, the animal that causes no trouble, are not necessarily horses experiencing good welfare. They may be horses that have learned, through accumulated experience, that they have no choices, no agency and no voice. Understanding the difference between contentment and resignation, between a horse that is living well and one that has shut down or given up, is essential for the future of ethical equestrianism.

Key References

Doherty, O., Casey, V., McGreevy, P. and Arkins, S. (2017) 'Noseband use in equestrian sports: an international study', PLoS ONE, 12(1)

Fureix, C., Jego, P., Henry, S., Lansade, L. and Hausberger, M. (2012) 'Towards an ethological animal model of depression? A study on horses', PLoS One, 7(6)

Hanggi, E.B. and Ingersoll, J.F. (2009) 'Long-term memory for categories and concepts in horses (Equus caballus)', Animal Cognition, 12(3), pp. 451-462.

Ledger, R.A. and Mellor, D.J. (2018) 'Forensic use of the Five Domains Model for assessing suffering in cases of animal cruelty', Animals, 8(7), p. 101.

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

Mellor, D.J. (2017) 'Operational details of the Five Domains Model and its key applications to the assessment and management of animal welfare', Animals, 7(8), p. 60.

Sankey, C., Richard-Yris, M.-A., Leroy, H., Henry, S. and Hausberger, M. (2010) 'Positive interactions lead to lasting positive memories in horses, Equus caballus', Animal Behaviour, 79(4), pp. 869-875.

Sykes, B.W., Hewetson, M., Hepburn, R.J., Luthersson, N. and Tamzali, Y. (2015) 'European College of Equine Internal Medicine consensus statement: equine gastric ulcer syndrome in adult horses', Journal of Veterinary Internal Medicine, 29(5), pp. 1288