Journal of Veterinary Behavior (2010) 5, 180-186 RESEARCH Over-flexing the horse’s neck: A modern equestrian obsession? Paul D. McGreevya, Alison Harmanb, Andrew McLeanc, Lesley Hawsona a Faculty of Veterinary Science, University of Sydney, New South Wales, Australia; Mount Hawthorn ESC, Mount Hawthorn, Western Australia, Australia; and c Australian Equine Behaviour Centre, Broadford, Victoria, Australia. b KEYWORDS: horse; equitation science; dressage; hyperflexion; vision; rollkur; respiratory stress; equitation practices and regulations Abstract We used an opportunistic review of photographs of different adult and juvenile horses walking, trotting, and cantering (n 5 828) to compare the angle of the nasal plane relative to vertical in feral and domestic horses at liberty (n 5 450) with ridden horses advertised in a popular Australian horse magazine (n 5 378). We assumed that horses in advertisements were shown at, what was perceived by the vendors to be, their best. Of the ridden horses, 68% had their nasal plane behind the vertical. The mean angle of the unridden horses at walk, trot, and canter (30.7 6 11.5; 27.3 6 12.0; 25.5 6 11.0) was significantly greater than those of the ridden horses (1.4 6 14.1; 25.1 6 211.1; 3.1 6 15.4, P , 0.001). Surprisingly, unridden domestic horses showed greater angles than feral horses or domestic horses at liberty. We compared adult and juvenile horses in all 3 gaits and found no significant difference. Taken together, these findings demonstrate that the longitudinal neck flexion of the degree desirable by popular opinion in ridden horses is not a common feature of unridden horses moving naturally. Moreover, they suggest that advertised horses in our series are generally being ridden at odds with their natural carriage and contrary to the international rules of dressage (as published by the International Equestrian Federation). These findings are discussed against the backdrop of the established doctrine, which states that carrying a rider necessitates changes in longitudinal flexion, and in the context of the current debate around hyperflexion. Ó 2010 Elsevier Inc. All rights reserved. Introduction The rules of dressage issued by the sport’s peak body, the Fédération Equestre Internationale (FEI), state on 9 occasions that the nasal plane should at all times be in front of the vertical (FEI, 2009). So it is fair to assume that this posture is desirable if horses are being prepared in a way Address for reprint requests and correspondence: Paul D. McGreevy, Faculty of Veterinary Science (B19), University of Sydney, NSW 2006, Australia; Phone: 161-2-9352810; Fax: 161-2-93513957. E-mail: paul.mcgreevy@sydney.edu.au 1558-7878/$ -see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.jveb.2010.03.004 that aligns with the FEI rules. Paradoxically, to most dressage riders these days, ‘‘on the bit’’ means that the horse travels with its neck arched and nose tucked in, on or behind the vertical. This degree of neck flexion is not advocated by leading equestrian manuals, such as the Official Instruction Handbook of the German National Equestrian Federation and the British Horse Society’s Manual of Equitation. However, theoretically, a vertical nose does not necessarily mean that the horse is ‘‘on the bit,’’ especially if it is being held in that position by the rider’s hands (McGreevy et al., 2005). By definition, the horse that carries itself ‘‘on the bit’’ is adjusting to McGreevy et al Over-flexing the horse’s neck the additional weight of the rider by caudally shifting its nose, shortening and arching its neck to some extent, bringing its hocks underneath it, and thereby shifting its centre of gravity caudally (FEI, 2007) with minimal contact or pressure from the reins. This postural change seems to have been predicated on the assumption that the rider cannot align his weight over the horse’s centre of gravity. It is worth considering how the concept of being ‘‘on the bit’’ may have been corrupted by common equestrian parlance. It is a poor translation from the French mise en main which literally means, put in the hand, and refers to the relaxation of the jaw and of the poll (Decarpentry, 1949) so that the horse ‘‘releases the mouth softly and loyally accompanies the hand wherever it moves’’ (Karl, 2008). The mise en main does not imply that the bit is necessary, or that the nasal plane should be vertical, or that behind the vertical is desirable. The neck and head position of the horse, in what is considered to be correct schooling, where dorsoventral flexion of the proximal cervical vertebrae and atlanto-occipital joint (flexing at the poll, poll flexion, or roundness) results in the nasal plane being approximately 6 in front of the vertical or 12 at walk (McGreevy et al., 2005). In addition, the poll should always be the highest point of the neck, and the horse should be trained for this self-maintained posture by achieving what is known as self-carriage (FEI, 2009). This self-maintained posture is intended to optimize the balance of the ridden horse and its readiness to respond to the signals transmitted by the rider. Alterations to this carriage correlate with the horse’s level of training. As the dressage horse’s capacity to collect develops, his poll raises further and the neck arches more, which brings his nose somewhat closer to his chest than in those of less trained horses. This results in changes in the topline physique of the horse over a period. These alterations are achieved through the anabolic effects of changes in velocity (McLean and McLean, 2009). It is likely that such changes facilitate the constant accelerations, decelerations, and changes of line that characterize the horse in the sport of dressage as opposed to the horse at liberty. Hyperflexion (extreme overbending of the neck; i.e., flexion of the neck toward the chest of the horse, thus narrowing the angle toward the vertical from the normal extended position) is seen in many dressage horses, especially during warm-up, and may be extreme as in the rollkur debate, where the nose sometimes remains adjacent to the horse’s pectoral region for some time (FEI, 2010). Head position can be readily moderated by bit pressure but doing so mixes 2 responses (deceleration and neck-flexing) with 1 signal (bit pressure) and as such is a violation of the principles of learning theory (McGreevy and McLean, 2007). In addition, neck flexion is controversial because it can most easily be achieved by relentless pressure on the mouth and when excessive seems to have the potential to compromise pulmonary ventilation (Petsche et al., 1995) and vision (Harman et al., 1999). Increased airway resistance during exercise could be expected to reduce alveolar ventilation. However, the effect on 181 oxygenation of the arterial blood, heart rates, and blood lactate concentrations during exercise has not been described. The effect on vision is subject to current debate since Bartos et al., (2008) drew attention to the ability of the horse to rotate its eyeball to maintain the pupil in a horizontal position. These authors claimed that the horse maintains the ‘‘horizontal eyeball position regardless of head position,’’ suggesting that this rotation overcomes any deficit in vision that might otherwise arise from the horse having its nose pointed caudally by the rider. We wished to consider this broad claim in the context of extreme neck flexion and extensions. We used an opportunistic review of photographs to audit the nasal angle in feral and domestic horses at liberty and compare them with those of Australian horses advertised in the Australian print media. Materials and methods We examined the first author’s 5-year archive of photographs of domestic and feral horses (in the Guy Fawkes National Park, NSW, Australia, and Tanna Island, Vanuatu) at liberty, along with all photographs appearing in the November 2009 edition of Horse Deals magazine (Agricultural Publishers Pty Ltd, Mount Gambier, South Australia). From a total pool of 7,469 photographs (2,532 of domestic horses at liberty, 177 of feral horses, and 4,760 of advertised horses), 828 (291 domestic, 55 feral, 378 advertised ridden, and 104 advertised unridden) were selected for further scrutiny, if the horses were within the following criteria: a. were moving at the walk, trot, canter, or gallop (the gait in each case was determined from the arrangement of the legs and the number of feet touching the ground), b. had their longitudinal axis of the head and neck perpendicular to the camera, c. had all 4 hooves apparent. These criteria were used to do the following: a. avoid measuring horses that were flexing their necks for a reason that could be expected to compromise forward vision (e.g., during bucking), b. permit an accurate assessment of the vertical inclination of the nasal plane, c. permit an accurate assessment of the horizontal inclination of the weight-bearing surface. For each horse satisfying these criteria, the gait was noted and the following measurements were taken (see Figure 1):  angle from the vertical of the nasal plane (vertical references were taken opportunistically from fence posts and tree trunks),  angle from the horizontal of the ground surface. Nasal angle is the angle formed between the dorsal flat part of the nose (i.e., in imaginary line formed by the frontal and nasal bones of the skull) and the vertical. 182 Journal of Veterinary Behavior, Vol 5, No 4, July/August 2010 Figure 1 Lines used to determine angles of the nasal plane and the weight-bearing surface in a typical photograph from the domestic cohort. Analysis We used a Linear Mixed Model within a residual maximum likelihood (REML) analysis, which took account of unequal replication. We used the Wald statistic for testing interactions and main effects. For the current analysis, this was based on an F distribution. The analysis proceeded in 3 steps. First, we tested for the 3-way interaction. If this was not significant, we then tested for the 3 two-way interactions when each variable was entered last in the analysis. Finally, we tested for a main effect, only if that main effect did not appear in a significant 2-way interaction at step 2. Results The distribution of nasal angles appears in Table 1. In 68% of the ridden horses, nasal planes were behind the vertical (40% at the walk, 71% at the trot, and 58% at the canter). We tested for the effect of age (juvenile, adult), horse type (advertised – ridden; advertised – unridden; domestic – at liberty; feral – at liberty), and gait (canter, trot, walk) in a single analysis (see Figure 2), Clearly, there were no juvenile advertised ridden horses but a REML showed no Table 1 significant interaction between gait and age on nasal angle. However, it did reveal a strongly significant 2-factor interaction between gait and the 4 types of horse (F statistic 5 4.58, d.d.f. 5 815, P , 0.001) reflecting differences in the REML predicted means, as shown in Table 2. In step 1 (of the steps in our analysis described above), 3-way interaction was not significant (F statistic 5 0.76, d.d.f. 5 807, P 5 0.051). In step 2, the pattern of the angles among horse types changed significantly (F statistic 5 3.19, d.d.f. 5 811, P 5 0.004) across gaits. There were no significant interactions that included age. In step 3, the interaction between gait and type was strongly significant (F statistic 5 4.58, d.d. f. 5 815, P ,0.001) but, again, age had no influence. The difference in angle of the nasal plane of ridden versus unridden horses was highly significant in all cases for each gait (canter 5 30.87, P , 0.001; trot 5 36.63, P , 0.001; and walk 5 18.04, P 5 0.001). The angles for advertised unridden horses were all significantly larger than those for domestic and feral horses for all gaits (domestic canter difference from advertised unridden horses 5 27.90, P , 0.001; trot 5 33.31, P , 0.001; walk 5 17.75, P 5 0.001; and feral canter difference from advertised unridden 5 22.00, P , 0.001; trot 5 31.73, P , 0.001; walk 5 12.46, P 5 0.003). Among feral and domestic horses, the angle associated with walking gait was significantly smaller than in other gaits (P , 0.001). Discussion Our results show a significant trend in unridden horses to moderate their nasal angle according to speed or gait, with the nose tipping up as the horse accelerates. This is also repeatedly prescribed by the FEI to occur in the ridden horse with increasing speed through increasing stride length (FEI, 2009). However, this trend is violated in our series of photographs of ridden horses, which shows radical departure from what is normal in unridden horses and what is considered correct in dressage. The significant increase in the nasal angle in unridden advertised horses when compared with domestic and feral horses at liberty is of interest because it most likely reflects the tendency for unridden horses to be chased during presentation for sale. This seems The distribution of ridden and unridden horses studied for nasal angle at 3 gaits Canter Trot Walk Groups n Average S.D. n Average S.D. n Average S.D. Advertised - ridden (n 5 378) Advertised - unridden (n 5 104) Domestic - at liberty (n 5 291) Feral - at liberty (n 5 55) All unridden (n 5 450) 45 16 136 9 161 1.4 32.1 28.8 25.4 30.7 14.1 8.8 12.9 5.9 11.5 318 82 89 29 200 25.1 30.8 26.1 23.5 27.3 211.1 9.8 15.3 10.5 12.0 15 6 66 17 89 3.1 31.3 25.0 19.9 25.5 15.4 10.6 10.6 12.0 11.0 McGreevy et al Over-flexing the horse’s neck Figure 2 183 Angle of nasal planes in photographed horses (n 5 828). to be derived from an attempt to showcase the horse’s paces. This is unfortunate because chasing is contraindicated in any training system designed to make horses safe around humans (McGreevy et al., 2009). The current findings demonstrate that the marked neck flexion of the degree desirable in the popular equestrian domain is not advocated by the FEI (2009), nor is it a common feature of unridden horses moving naturally. Arguably, this latter finding could be because of age in that more ridden horses in advertisements are adults. However, a comparison of all unridden horses clustered according to age showed no significant difference resulting from age. Although in feral and domestic horses, nasal plane angle positively correlates with gait (and/or speed), no such correlation is seen in the ridden horses. In fact, the nasal plane angles are significantly less at trot than at walk. Biomechanical studies show that horses stabilize the 3 segments of their body (i.e., head, neck, and trunk) in all planes during all 3 gaits, but walk is known to show the greatest angular displacements of the head (Dunbar et al., 2008). Thus, at trot, there is less longitudinal movement of Table 2 Table of predicted means for type ! gait interactions Advertised ridden Advertised unridden Domestic Feral Canter Trot Walk 0.95 32.50 29.34 23.72 25.59 31.62 28.19 26.35 2.64 21.00 20.82 15.38 the neck compared with walk and canter, so it may well be that riders are taking advantage of this lowered neck mobility to shorten the neck more than at other gaits and that the horse is held in a shortened posture by the rider’s hands. Regardless, the results show that, if they are ridden as they are advertised in trot (the most common gait in horses advertised under saddle), horses in the current Australian sample are generally being ridden more incorrectly than at other gaits, according to the FEI directives (2009). Although FEI rules prescribe some arching of the neck, this is limited by the directive to maintain the nose in front of the vertical line. It seems that advertisers tend to favor photographs of horses with arched necks and it seems apparent that over time, over-arching of the neck and ignoring the angle of the nasal plane must have become fashionable. Yet many engravings since the renaissance depict highly arched necks in horses trained in dressage. It seems plausible that if trainers and riders followed the principle of self-carriage, where the horse’s head and neck posture is a learned response rather than a forced one, overarching and nasal planes behind the vertical would have been prevented. This supports previous calls for trainers and riders to be educated in the principles that arise from learning theory (McGreevy, 2007; Warren-Smith and McGreevy, 2008). Our results reflect a tendency for humans to mistakenly assume that horses are educated and ‘‘on the bit’’ because they have their noses tucked in (especially at the trot). Many equestrian authorities suggest that increasing amounts of neck flexion are counterproductive and put the horse ‘‘on the forehand’’ (so it is heavier on the forequarters) and that is presumably why flexion behind the vertical is penalized in competition dressage. That said, it 184 is perhaps perplexing that in a sport aimed to exhibit optimal training, riders warm-up their horses in neck postures at odds with optimal performance. This suggests that psychological or biomechanical advantages are to be gained through this contrary practice, which is a widely held belief. However, any increased curvature of the upper respiratory tract increases turbulence and so is likely to compromise athletic performance. As workload reaches maximum in oxygen-deprived horses, lactic acid and beta–endorphin concentrations show a marked increase (Mehl et al., 2000). This may also provide fertile ground for future research that explores the physiological effects of neck-flexing. This trend toward hyperflexion may also have unwelcome implications for rider safety if, in response to rein tension, the horse neck-flexes, rather than decelerates. Safety is further compromised if this confusion generates conflict behaviors that manifest as dangerous hyper-reactive behaviors (McGreevy and McLean, 2007). Although neck flexion can be forced/trained for collection, dressage exponents regard it as a natural response (they point to similar postures in horses at liberty when excited by the presence of a new conspecific or mating possibility). Here the stride shortens and the trot elevates (as in collected trot) and there is a significant concomitant elevation of the poll and shortening of the neck. Dressage is said to mimic these natural postures so that during collection the neck shortens and the poll is raised, whereas in all gaits the neck lengthens and the poll lowers as the stride lengthens. Furthermore, it is also believed that the added weight of the rider necessitates a caudal shift in the horse’s centre of mass to achieve balance. Indeed, in the FEI (2009) guidelines for dressage judging, the stated aim of collection is to improve the balance and equilibrium of the horse, which has been more or less displaced by the additional weight of the athlete (FEI, 2009). However, this may represent a legacy from the long-held belief that a horse, when mounted, regains some mechanical advantage by a caudal shift in its centre of gravity when it is moving forward in collection (FEI, 2007). A recent study by Buchner et al. (2000) shows that the approximate body centre of mass of unmounted horses is on a line approximately 2 cm below the true hip joint, 1 cm left lateral of the centre line and level with T12–13 (i.e., in a position consistent with the anticlinal in the dorsal spinous processes). This is more caudal than previously thought but fortuitously aligns directly under the rider’s body centre of mass, apparently obviating the need for any shift. So, moving the horse’s centre of mass further caudally than the rider’s centre of mass may actually be mechanically inefficient. Instead of bestowing a biomechanical advantage for the horse, true collection may simply be a demonstration of advanced training and provides the additional bounce that makes horses somewhat more comfortable to ride. Holmstr} om and Drevemo (1997) showed that Grand Prix horses have a higher compression of hindlimb joints with Journal of Veterinary Behavior, Vol 5, No 4, July/August 2010 longer stance phase as collection increases. That said, in at least 1 study, (Holmstr}om et al., 1995) the placement of the hindfeet more cranially under the horse’s body did not increase with higher collection, and over-track is known to decrease with collection in trot (Clayton, 1994). In so-called classical training (which nowadays refers to a multinational post-Middle Ages approach to horse training, principally for the purposes of war), the correct position of the nasal plane when riding a horse ‘‘on the bit’’ is always in front of the vertical and with the head more or less raised, according to the level of training (FEI, 2009). In classical texts, the height of the poll (the atlantooccipital junction) is given considerable attention because it changes according to stride length: when the horse makes short strides, the poll rises (and comes a little closer to the withers), and when the stride lengthens, the poll lowers (and the neck lengthens a little). Collection (characterized by the higher poll, more arched neck as well as lowered croup) develops as a result of repeated transitions of lengthening and shortening the stride over time, and the topline (the cervical and thoracolumbar) muscles develop as a result of this (Podhajsky, 1966). Horse trainers who adhere to an older ‘‘classical’’ approach claim that these alterations in poll height (and to a lesser extent neck length) cause the necessary shift in centre of gravity back and forward as strides shorten and lengthen (Karl, 2008). As already noted, the need for this has recently been refuted. It seems likely that this dogma began at the end of the Middle Ages when the weight of armor added significantly to the rider’s weight. In addition, the front end of the horse was also covered by metal armor (e.g., the chanfron, crinet, peytrel, and flanchard were all metal covers for the head, neck, shoulders, and sides of the horse) and this extra weight may well have necessitated the transfer of some weight to the hindquarters of the horse for it to be balanced and locomote optimally. Accordingly, classical training emphasizes that collection is purely the result of the physical development of the horse’s body as a result of transitions that enabled weight to be transferred to the hindquarters. In contrast, contemporary dressage training focuses on bringing the nose behind the vertical, which is believed to make the horse work ‘‘correctly’’ and ‘‘over the back’’ where the lowering of the head lifts the vertebral column (Van Weeren and Schrijer, 2003). Equitation Science (McGreevy et al., 2009) should be able to clarify this issue by showing how a strong lumbar musculature in which the muscles raise the back relates to the softer back the ‘‘classical’’ masters describe. Similarly, future research should explore how and when the trend toward permitting and even promoting carriage behind the vertical emerged. In 2010, the FEI claimed to have resolved the issue by redefining hyperflexion/rollkur as flexion of the horse’s neck that is the product of aggressive force (FEI, 2010). The same statement went on to assert that flexion without undue force, a technique they referred to as low, deep, and round, is acceptable. The group unanimously agreed McGreevy et al Over-flexing the horse’s neck Figure 3 185 The apparent orientation of the pupil in a pony with its neck relaxed, extended, and hyper-flexed. that any form of aggressive riding must be condemned but made no mention of how this might be defined or measured. If the only force that is unacceptable is characterized by the intention of the rider (i.e., the rider’s aggression), then stewards will struggle to be sure that a breach has occurred. The International Society for Equitation Science (ISES) has written to the FEI suggesting that this statement makes the need to collect rein tension data more pressing than ever. Policing action against aggressive force relies on a universally accepted definition of aggressive force, which can be achieved only by ensuring that objective measures are made to confirm or refute the observations made by stewards. The effect of neck-flexing on vision is controversial. Harman et al. (1999) indicated that horses ridden behind the vertical have compromised vision. In contrast, Bartos et al., (2008) noted that neck arching is seen by stallions approaching one another and questioned why horses would ‘‘handicap themselves visually during stressful social interactions that can lead to a fight causing injury or death or in an uncertain situation that might be dangerous for the subject.’’ The reality is that parallel prancing is an important feature of inter-male interactions (i.e., when stallions approach one another, rather than merely advancing directly head-on, they do so with lateral and diagonal displays that include neck-flexing). When horses cavort, their noses occasionally point very much in the direction of the ground they have just covered but this would be maladaptive only if it were maintained. By way of an example, neck-wringing in play may have the same effect on vision but we should not assume that vision is always prioritized by horses. Brief displays may be of sufficient proximate importance to horses that surveillance may be sacrificed, but only very briefly. The Bartos et al. (2008) study reported that the pupil maintained its horizontal position relative to the ground regardless of head position. However, they did not report the head positions they examined, and their photographs seemed to imply that the nose was always somewhat in front of the vertical. They were not concerned with forward binocular vision and did not use any technique such as ophthalmoscopy to determine visual field (Harman et al., 1999) or alignment of the pupil relative to the corpora nigra, so it is difficult to comment on their findings. Our own preliminary data indicate that in extreme head positions the pupil is not parallel to the ground (Figure 3), but further work is needed to confirm this. The ideal study would apply ophthalmoscopy during radical neck flexion and extension to a cohort of horses with different head shapes (Evans and McGreevy, 2006a) because it is known that the distribution of ganglion cells differs with skull length (Evans and McGreevy, 2006b). When horses are grazing, the head is well in front of the vertical and the binocular field is directed toward the ground in front of the nose. However, with their laterally placed eyes, they have excellent, extensive lateral vision, which enables them to scan the environment in this position (Harman et al, 1999). By contrast, when the horse is 186 moving through space, it needs to use the binocular visual field, necessary for seeing in 3 dimensions. This field is located down the nose of the horse, not directly in front of the animal as it is in humans (Harman et al., 1999). This explains why the horses at liberty in the current series of photographs travel with the head well in front of the vertical. The current trend for over-flexing the neck significantly compromises forward vision. This head position produces a horse that, because it cannot see in front, is rendered somewhat powerless, which may be necessary for riders of modern, vigorous competition horses. It is clearly contraindicated in jumping horses in whom forward vision is crucial. Conclusions Inappropriate neck-flexing behind the vertical is surprisingly common in ridden horses advertised in Australia and only very rarely seen in horses at liberty. In contrast, true collection by horses at work (with the poll raised and the neck moderately flexed) may be a demonstration of appropriate training for particular locomotory requirements and may therefore be desirable, even without meeting a perceived need for a shift in the horse’s centre of gravity. However, artificial collection is easily mimicked by simple neck-flexing through increased rein tension, an approach that is clearly contraindicated on training and welfare grounds because it exposes the horse to 1 pressure signal for 2 responses (neck-flexing and deceleration), increased turbulence in the upper respiratory tract, and compromised forward vision. It may also have implications for rider safety if the horse shows less deceleration in response to rein tension. The trend toward a focus on incorrect neckflexing in contemporary elite dressage may account for the popularity of this ill-advised practice in the wider dressage rider population. Acknowledgments Pierre Malou and Dr Mick O’Neill are warmly thanked for their contribution to measurements of nasal angles and statistical analyses, respectively. References Bartos, L., Bartosová, J., Starostová, L., 2008. Position of the head is not associated with changes in horse vision. Equine Vet. J. 40, 599-601. Buchner, H.H.F., Obermüller, S., Scheidl, M., 2000. Body centre of mass movement in the sound horse. Vet. J. 160, 225-234. Journal of Veterinary Behavior, Vol 5, No 4, July/August 2010 Clayton, H.M., 1994. Comparison of the stride kinematics of the collected, working, medium and extended trot in horses. Equine Vet. J. 26, 230-234. Decarpentry, A.E.E., 1949. Academic equitation. J.A. Allen, London, UK. Dunbar, D.C., MacPherson, J.M., Simmons, R.W., Zarcades, A., 2008. Stabilization and mobility of the head, neck and trunk in horses during overground locomotion: comparisons with humans and other primates. J. Exp. Biol. 211, 3889-3907. Evans, K.E., McGreevy, P.D., 2006a. Conformation of the equine skull: a morphometric study. Anat. Histol. Embryol. 35, 221-227. Evans, K.E., McGreevy, P.D., 2006b. The distribution of ganglion cells in the equine retina and its relationship to skull morphology. Anat. Histol. Embryol. 35, 1-6. Fédération Equestre Internationale, 2007. Dressage handbook guidelines for judging. FEI, Lausanne, Switzerland. Fédération Equestre Internationale, 2009. Rules for dressage events, 23rd Ed.. FEI, Lausanne, Switzerland. Fédération Equestre Internationale, 2010. FEI round table conference resolves rollkur controversy. Available at: http://www.fei.org/disciplines/dressage/ press-releases/fei-round-table-conference-resolves-rollkur-controversy. Accessed February 18, 2010. Harman, A.M., Moore, S., Hoskins, R., Keller, P., 1999. Horse vision and the explanation of visual behaviour originally explained by the ‘ramp retina’. Equine Vet. J. 31, 384-390. Holmstr}om, M., Drevemo, S., 1997. Effects of trot quality and collection on the angular velocity in the hindlimbs of riding horses. Equine Vet. J. Suppl. 23, 62-65. Holmstr}om, M., Fredricson, I., Drevemo, S., 1995. Biokinematic effects of collection on the trotting gaits in the elite dressage horse. Equine Vet. J. 27, 281-287. Karl, P., 2008. The twisted truths of modern dressage. Cadmos, Brunsbek, Germany. McGreevy, P.D., 2007. The advent of equitation science. Vet. J. 174, 492-500. McGreevy, P.D., McLean, A.N., 2007. The roles of learning theory and ethology in equitation. J. Vet. Behav.: Clin. Appl. Res. 2, 108-118. McGreevy, P.D., McLean, A.N., Warren-Smith, A.K., Waran, N., Goodwin, D. (Eds.), 2005. Defining the terms and processes associated with equitation. Proceedings of the 1st International Equitation Science Symposium, Broadford, Victoria, Australia, Post-Graduate Foundation in Veterinary Science, Sydney, pp. 10–43. 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? Vet. J. 181, 12-18. McLean, A.N., McLean, M., 2009. Academic horse training: equitation science in practice. McLean Andrew & Manuela. Australian Equine Behaviour Centre, Broadford, Victoria, Australia. Mehl, M.L., Schott, H.C., Sarkar, D.K., Bayly, W.M., 2000. Effects of exercise intensity and duration on beta-endorphin concentrations in horses. Am. J. Vet. Res. 61, 969-973. Petsche, V.M., Derksen, F.J., Berney, C.E., Robinson, N.E., 1995. Effect of head position on upper airway function in exercising horses. Equine Vet. J. Suppl. 18, 18-22. Podhajsky, A., 1966. The complete training of horse and rider in the principles of classical horsemanship (E. Podhajsky and V.D.S. Williams, Trans.). Doubleday, Bantam Doubleday Dell, New York, NY. Van Weeren, R., Schrijer, S., 2003. Auf dem falschen rucken ausgetragen. Reiterrevue. 7, 13-15. Warren-Smith, A.K., McGreevy, P.D., 2008. Equestrian coaches’ understanding and application of learning theory in horse training. Anthroz}oos. 21, 153-162.