Summary: The round-pen training technique, popular among proponents of “natural horsemanship” is said to be effective and ethical because it uses equine ethology to induce desired behaviors. This review looks at what we know about ethology and illustrates the ways this technique is aversive to horses, and how it is based on a flawed understanding of horse behavior.

Traditional views express the need for “horsemanship” during human-horse interactions. Horsemanship is defined as the art of training horses through the understanding of the horse’s ethology and language, as having extensive knowledge and practical experience, and as promoting the development of a mutual relationship between horse and human.

Although domestic horses may be tamer, faster-growing and stronger due to breeding selection, their natural behaviours and instincts most likely remain.¹ In the past 40 years, approximately, training methods based on understanding the “natural” behaviour of horses has increased in popularity.² Common factors understood in “natural horsemanship” training include ensuring the trainer is able to establish “leadership” or “dominance” and “respect” from the horse.³ This is based on the belief of the importance of incorporating intraspecific interactions into the human-horse relationship to support the cues and techniques used in natural horsemanship training. The idea was formulated by natural horsemanship trainers Pat Parelli and Monty Roberts after assessing wild horse and domesticated horse interactivity. A popular method used by natural horsemanship trainers, such as Roberts, Lyons and Brannaman, is the round pen training (RPT) technique. RPT will be carried out in a round structure of a sufficient height to ensure the horse cannot escape⁴ and the trainer will mainly be in the centre of the round pen, predominantly using visual and auditory cues, rather than equipment, generally as negative reinforcers to cue acceleration, deceleration and changes of direction either through adding or removing primary or secondary reinforcers.^5,6

Physical actions

Initial stages of round pen training

Initially, an RPT trainer wants to introduce the concept of what they require from the horse within the round pen through operant conditioning and negative reinforcement, and eventually habituation to the activity.³ The trainer will use fear as a motivator for the horse to move at an accelerated gait around the edge of the pen away from the trainer.⁶ The RPT trainer will use exaggerated cues, such as smacking of the leg, shouting, or flailing of the arms to send the horse out to the outside of the pen and follow the horse behind the driving line by using bodily placement as pressure to the horse;⁷ once the horse is moving around the edge of the pen, the cues are removed. The exaggerated and aversive cues initiate the sympathetic nervous system in the horse, causing behaviours such as increased gait and high head position, which correlates with an increased heart rate,⁸ vocalisation and potentially escape attempts.⁹

RPT trainers correctly state they are initiating their flight response. However, they state they are “allowing” the horse to present this response, suggesting the horse is willingly offering the desired response (discussed further in the section on Partnership, below). Pressure from the RPT trainer’s bodily placement at either side of the driving line (behind or in front of the shoulder) is using the horse’s ethology to train, since horses are always displaced/driven from behind by conspecifics,¹⁰ and when the trainer intercepts the horse’s path, the horse will avoid colliding with them and change their direction.¹¹

Negative reinforcement

The forward movement of the horse is the horse’s attempt to increase distance between them and the trainer.¹² If the horse is unable to increase their distance from the trainer, this could lead to flooding, which is highly stressful. Situations where an animal is highly stressed and unable to act to decrease their stress can induce learned helplessness. The attempt to use negative reinforcement can, therefore, inhibit learning.¹³

Negative reinforcement is the most common form of horse training^5,14 and allows the horse to attempt to trial behaviours when pressure is applied.⁶ However, negative reinforcement training further decreases the likelihood of investigative behaviours and decreases the likelihood of habituation. This suggests the inhibition of learning in comparison to positive reinforcement training.¹⁵RPT trainers do state the significance of rewarding the horse by rubbing the forehead once the horse has “submitted” (discussed further in the section on Leadership), therefore positively reinforcing them. However, this has not shown to be an area often groomed by horses; a study has suggested they prefer to be groomed at the withers.¹⁶

RPT trainers state the importance of removing pressure, and this correlates with the theory of associative learning, as the removal of the pressure is the reinforcer, and the timing of the removal is fundamentally important for learning to occur.⁵

However, these trainers remove their aversive cues, but still maintain pressure through bodily placement by following the horses around the pen at the start of RPT. This is essentially positive punishment, as the pressure is not released after the desired behaviour is offered,¹⁷ which further decreases the likelihood of the horse offering the desired responses.⁵

RPT trainers state that once the cues are removed and the horse reduces their gait or stops to look at the trainer, the horse is “submitting” to them and wants to be accepted by their now established “leader,” mimicking intraspecific interactions (further discussed in the section on Leadership). However, Henshall et al. executed this RPT scenario with a remote-control car and found that horses were not only more stressed when carrying out negative reinforcement training in comparison to the addition of positive reinforcement training, but that the remote-control car acquired the same responses from the horse as an RPT trainer would.¹² This suggests that horses are learning simply through negative reinforcement rather than the trainer mimicking horse behaviour.

When a horse “submits,” they are expected to approach the trainer once the pressure has been removed. However, this can confuse the horse as the horse has been conditioned through escape/avoidance learning to increase distance from the aversive stimulus, the trainer, in the round pen.^6,18 If the horse does not approach the trainer, it is seen as a sign of disrespect and results in further chasing. Through repeated trials, the horse learns through negative reinforcement and other forms of operant conditioning to approach the trainer once cues have been removed to avoid being chased, rather than it being the horse presenting “submission” to the trainer.¹⁹

Arousal levels

Due to the miscommunication arising from maintained bodily pressure from RPT trainers, watchers following the advice of these trainers could find their horse to be increasingly aroused by the pressure maintained, without understanding the role of body placement. Therefore, inconsistent and high pressures will continue to be exerted onto the horse,⁷ which can cause confusion, induce learned helplessness,²⁰ or cause the horse to attempt to remove the pressure either by attempting to escape the round pen or potentially injuring the trainer.^1,6

Wilk and Janczarek state the importance of ensuring the horse’s arousal levels are properly assessed in the first 10 minutes of RPT.⁹ This may be because of the amount of pressure initially exerted onto the horse, and after 10 minutes, the horse is either showing fatigue,⁴ learned helplessness,²⁰ or has quickly learned through simple learning theory.²¹

Horses in a negative affective state experience increased arousal levels²² and simply anticipating such an activity has shown to increase arousal levels in other species.²³ This is apparent in RPT when inducing fight or flight responses. However, there can also be considerable differences in the level of arousal in horses depending on breed, sex, temperament, and genetics.^9,24–26Differences in arousal levels between individual horses can also affect the successfulness of technique repeatability, as results of beginners taking on the RPT techniques may not be considering individuality. This was shown in Krueger’s study on round pen training.¹⁹ In this study, 26 domestic horses were subjected to round pen training so that their behaviour and interactions with the trainer could be analysed. During the training, three of the horses needed to be pulled from the study, “Those horses showed untypical, immense sweating and did not respond to the experimenter anymore. Therefore the test was stopped for animal welfare reasons.”

Classical conditioning

At the beginning of RPT, cues are paired with other cues to make the horse look obedient by being under stimulus control.⁶ For example, initially, when changing the direction of a horse the trainer will quickly approach in the horse’s path and the horse will change direction to avoid collision. Through classical conditioning, an experienced RPT horse will change direction as soon as the trainer positions themselves in front of the driving line, without the use of a fast approach and raised arms. These details are apparent visible in an RPT trainer’s training seminars if the viewer is knowledgeable, but it is very important to understand learning theory prior to carrying out RPT for effective results and a less confused horse.¹⁸

Terminology

Natural horsemanship ethology

Intraspecific interactions RPT trainers say are the most salient in human-horse interactions are dominance hierarchies, agonistic behaviour, and herd social organisation.⁴ RPT trainers state that horse herds are structured based along a dominance hierarchy, where a leading horse, usually the mare, is in charge of the herd and all other associates follow. They state that the order of the herd is decided through aggressive behaviours and the horses of a lower rank present submission to the dominant horse. They also suggest that competition for the higher rank is often presented through agonistic behaviours and disrespect from the lower ranked horses.

Dominance

Leaders or dominant horses are generally seen during intraspecific and context-specific interactions when conflict over resources is present — horses do not form strict dominance hierarchies in their day-to-day life in herds.²¹ Dominance has been described by ethologists as repeated learned relations that minimise the need for aggression or threats and submission.^27,28 However, even in scientific contexts dominance has been frequently misunderstood and described to be shown through one-off forms of aggression over food, which simply signifies rank.²⁹

Some RPT trainers believe asserting dominance occurs through invading personal space and some do not define the word when using it – so there is a slight disregard for the ethological definition of dominance, which is then further broken down by the public.²⁸

Aggressive interactions are generally seen when resource guarding²¹ and other feral intraspecific interactions, i.e., stallions fighting over mating with mares, and resource competition is also rare, as evidence by the reported low aggression rates while grazing .²⁸

In addition to the definition of dominance being flawed, the relevance of dominance within a human-horse relationship in RPT is flawed. The need for social cohesion through collective organisation amongst conspecifics is needed to decrease the vulnerability to predators, and it must be kept in mind that horses are prey animals that rely on cohesion, avoiding collision with one another, communication, and assembled flight through self-organisation rather than establishing a dominance hierarchy to survive.^30,31

Leadership

RPT trainers frequently comment on a horse’s desired body language and the need to notice subtle cues, which is essential for communication with the horse.³² These trainers want their horse to ‘search for a leader,’ specifically to view the handler as their leader, and to “bow” and present “submission” to them. This is seen when a horse lowers their head, attempts to come closer to the trainer, relaxes, turns their inner ears toward the trainer, and engages in licking and chewing.

RPT trainers believe this derives from submissive behaviours shown by youngsters when the “lead mare” keep them away from the group for a certain period of time. “Bowing” is a human social gesture to express thanks or gratitude and is not part of the horse’s natural ethogram used in the context of a higher cognitive ability to “respect” the RPT trainer.¹⁸ Lowering of the head in horses has been described as a displacement behaviour in a similar situation where an aggressor is present.³³ The inner ear fixed on the trainer shows attentiveness,³⁴ and licking and chewing has been suggested by Krueger to be a horse returning to a parasympathetic state after a state of arousal.¹⁹

Finally, Warren-Smith et al., found that a mare and a youngster in the same setting did not present the behaviours associated with “seeking a leader” in the way the RPT trainers describe.³⁵ This suggests that RPT trainers could be misinterpreting the behaviours their horses present, and mistaking fear for respect.

This suggestion of submission has also been called into doubt in the aforementioned study by Krueger,¹⁹ who observed whether RPT trainers could achieve the same results with their RPT-trained horse in a field as they did in a round pen. The results showed the horses did not respond in the same way in a natural setting as in a round pen, despite RPT trainers’ claims to be using “natural” techniques.

Partnership

RPT trainers state they must create a partnership with the horse. This assumes that horses are able to willingly cooperate with trainers to accomplish shared goals.³⁶ This requires a more complex social cognitive ability,³⁷ which horses lack.³⁸ Assuming a horse can willingly participate in this kind of activity is anthropomorphism and compromises welfare, as any horse that does not respond as the trainer desires may be at risk of punishment,, which increases arousal levels.⁶ This is all due to the misunderstanding of cognitive ability,³⁹ as lack of desired response is viewed by the trainer as a sign of disrespect.

Conclusion

RPT must be carried out correctly and with caution. Neither horse nor human has evolved to partake in RPT, so both need to learn and adapt to each other’s cues and responses. To achieve this, an understanding of equine body language, ethological behaviours, arousal levels/emotional state, the four quadrants of operant conditioning, classical conditioning, and how to efficiently apply and remove cues is vital to minimise stress.

Horse social status is unlikely to be related to human-horse interactions. The belief that the two are related is based on inferences made by RPT trainers about the ethological importance of these cues and responses from horses in RPT. There is also no evidence that horses view humans as conspecifics. Therefore, the suggestion that the horse’s social status can be transferred over to human-horse interactions to attempt to achieve compliance through dominance is not supported. The context of how and when social status of horses is established and the variety of factors involved, such as visual signalling which humans are unable to mimic, is more complicated than RPT trainers acknowledge.

Horses’ responses during RPT are likely to be based on simple learning theory: responding to reinforcers and punishers rather than attempting to achieve a higher social status than the trainer. Teaching and communicating this belief endangers equine welfare, as handlers are likely to perceive fight or flight behaviours as disrespect and attempt to assert their dominance through punishment or increased pressure. RPT should incorporate scientific evidence based on the horse’s ethology into the training process. It should involve more positive reinforcement, rely less on fight or flight responses, and use exertions of pressure humanely and correctly to increase motivation and maximise horse welfare. RPT trainers need to efficiently communicate their actions to their audience using evidence and learning theory to support them.

References

1. Waran, N., McGreevy, P., Casey, R.A. (2007). Training Methods and Horse Welfare. In: Waran N. (eds) The Welfare of Horses. Animal Welfare, vol 1. Springer, Dordrecht.
2. Nave, J.E., Miller, R.M., Rollin, B.E., Madigan, J.E., Whittemore, J., Freeman, D.A., Mundy, G.D., Schonholtz, C.M., DeHaven, W.R., Reece, V.P., Friend, T. (2000). Veterinary Medical Association’s Tenth Annual Animal Welfare Forum. Journal of the American Veterinary Medical Association. 4:8, 1231-1261.
3. Goodwin, D., McGreevy, P., Waran, N., McLean, A. (2009). How equitation science can elucidate and refine horsemanship techniques. The Veterinary Journal. 181:1, 5-11.
4. Henshall, C., McGreevy, P.D. (2014). The role of ethology in round pen horse training—A review. Applied Animal Behaviour Science. 155, 1-11.
5. McLean, A.N., Christensen, J.W. (2017). The application of learning theory in horse training. Applied Animal Behaviour Science. 190, 18-27.
6. Fenner, K., Mclean, A.N., McGreevy, P.D. (2019) Cutting to the chase: How round-pen, lunging, and high-speed liberty work may compromise horse welfare. Journal of Veterinary Behavior. 29, 88-94.
7. McGreevy, P.D. (2007). The advent of equitation science. The Veterinary Journal. 174:3, 492-500.
8. Smiet, E., Van Dierendonck, M.C., Sleutjens, J., Menheere, P.P.C.A., van Breda, E., de Boer, D., Back, W., Wijnberg, I.D., van der Kolk, J.H. (2014). Effect of different head and neck positions on behaviour, heart rate variability and cortisol levels in lunged Royal Dutch Sport horses. The Veterinary Journal. 202:1,. 26-32.
9. Wilk, I., Janczarek, I. (2015). Relationship between behavior and cardiac response to round pen training. Journal of Veterinary Behavior. 10:3, 231-236.
10. Ransom, J.I., Cade, B.S. (2009). Quantifying Equid Behavior— A Research Ethogram for Free-Roaming Feral Horses. [pdf] Last accessed 28^th December 2020
11. McGreevy, P., McLean, A., Buckley, P., McConaghy, F. and McLean, C. (2011). How riding may affect welfare: What the equine veterinarian needs to know. Equine Veterinary Education. 23:10, 531-539.
12. Henshall, C., Padalino, B., McGreevy, P. (2012). The radio-controlled car as a herd leader? A preliminary study of escape and avoidance learning in the roundpen. In: Proceedings of the 8th International Equitation Science Conference 18^th– 20^thJuly, Eds; H. Randle, N. Waran, J. Williams, British Society for Animal Science, Edinburgh, 157. At: Edinburgh
13. Maier, S.F., Seligman, M.E. (1976). Learned helplessness: Theory and evidence. Journal of Experimental Psychology. 105:1, 3-46.
14. Ahrendt, L.P., Rodrigo, L., Malmkvist, J., Nicol, C.J., Christensen, J.W., (2015). Development of a standard test to assess negative reinforcement learning in horses. Applied Animal Behaviour Science. 169, 38-42.
15. Innes, L., McBride, S. (2008). Negative versus positive reinforcement: An evaluation of training strategies for rehabilitated horses. Applied Animal Behaviour Science. 112:3–4, 357-368.
16. Feh, C., de Mazieres, J. (1993). Grooming at a preferred grooming site reduces heart rate in horses. Animal Behaviour 46, 1191-1194.
17. McGreevy, McLean, A.N. (2009) Punishment in horse-training and the concept of ethical equitation. Journal of Veterinary Behavior. 4:5, 193-197.
18. McGreevy, P.D., Oddie, C., Burton, F.L., McLean, A.N. (2009). The horse–human dyad: Can we align horse training and handling activities with the equid social ethogram? The Veterinary Journal. 181:1, 12-18.
19. Krueger, K. (2007). Behaviour of horses in the “round pen technique”. Applied Animal Behaviour Science. 104:1–2, 162-170.
20. Hall, C., Goodwin, D., Heleski, C., Randle, H., Waran, N. (2008). Is there evidence of learned helplessness in horses? Journal of Applied Animal Welfare Science. 11:3, 249-66.
21. Hartmann, E., Christensen, J.W., McGreevy, P.D. (2017). Dominance and Leadership: Useful Concepts in Human–Horse Interactions? Journal of Equine Veterinary Science. 52, 1-9.
22. Starling, M.J., Branson, N., Denis, C., McGreevy, P.D., (2013). Conceptualising the impact of arousal and affective state on training outcomes of operant conditioning. Journal of Equine Veterinary Science. 52, 1-9.
23. Gillete, R.L., Angle C.T., Sanders, J.S., DeGraves, F.J. (2011). An evaluation of the physiological affects of anticipation, activity arousal and recovery in sprinting Greyhounds. Applied Animal Behaviour Science. 130:3–4, 101-106.
24. Wolff, A., Hausberger, M., Scolan, N.L (1997). Experimental tests to assess emotionality in horses. Behavioural Processes. 40:3, 209-221.
25. von Borstel, U.K., Euent, S., Graf, P., König, S., Gauly, M. (2011). Equine behaviour and heart rate in temperament tests with or without rider or handler. Physiology & Behavior. 104:3, 454-463.
26. Schrimpf, A., Single, M-S., Nawroth, C. (2020). Social referencing in the domestic horse. 10:1, 164.
27. Drews, C. (1993). The concept and definition of dominance in animal behaviour. Behaviour. 125:3-4, 283-313.
28. Rees, L. (2017). Horses in company. Marlborough: J.A. Allen.
29. Houpt, K.A., Law, K., Martinisi, V. (1978). Dominance hierarchies in domestic horses. Applied Animal Ethology. 4:3, 273-283.
30. Rees, L. (2011). Social organization in horses, and its implications for the horse-human relationship. Last accessed 1/5/2020
31. Rees, L. (2019). Dominance in horse social order and training (video). Last accessed 1/5/2020
32. Ladewig, J. (2019). Body language: Its importance for communication with horses. Journal of Veterinary Behavior. 29, 108-110.
33. Warren-Smith, A.K., McGreevy, P.D. (2008). Preliminary investigations into the ethological relevance of round-pen (round-yard) training of horses. Journal of Applied Animal Welfare Science. 11:3, 285-298.
34. Gleerup, K.B., Forkman, B., Lindegaard, C., Anderson, P.H. (2015). An equine pain face. Veterinary Anaesthesia and Analgesia. 42:1, 103-114.
35. Warren-Smith, A.K., Greetham, L., McGreevy, P.D. (2007). Behavioral and physiological responses of horses (Equus caballus) to head lowering. Journal of Veterinary Behavior. 2:3, 59-67.
36. Goodwin, D. (2005). ‘Horse behaviour: Evolution, domestication and feralisation. In: Mills, D., and McDonnell, S.’ (eds.) The Domestic Horse: The Origins, Development and Management of its Behaviour. Cambridge: Cambridge University Press 1-17.
37. Hecht, E. E., Patterson, R., & Barbey, A. K. (2012). What can other animals tell us about human social cognition? An evolutionary perspective on reflective and reflexive processing. Frontiers in Human Neuroscience.
38. Rørvang, M.V., Christensen, J.W., Ladewig, J., McLean, A.N. (2018). Social learning in horses—fact or fiction? Frontiers in Veterinary Science. 5.
39. Smith, A.V., Wilson, C., McComb, K., Proops, L. (2017). Domestic horses (Equus caballus) prefer to approach humans displaying a submissive body posture rather than a dominant body posture. Animal Cognition. 21:4.

A recent graduate of Writtle University’s Equine Behavioral Science program, Chloe is passionate about spreading information and education on horse behaviour, management and training. During her studies, Chloe won the IAABC award, Best Overall Academic Performance award across all equine and canine studies at Writtle University, and the Equine Behaviour Forum award. She has also carried out successful behavioral consultations with clients, lectured at talks set up by Justine Harrison, and is currently working towards her accreditation to become a certified animal behaviour consultant.