Whether you are a dressage enthusiast or just an occasional rider reading equestrian magazines and message boards, chances are you have heard the recent debate about "rollkur".

Rollkür, now officially known as "hyperflexion of the neck," is a training technique used by many dressage riders today. Hyperflexion has also been used in other equestrian disciplines, most notably show jumping.

There has long been controversy over the welfare implication of using this technique, and recently the issue has come to a head. An incident occurred at Odense on the 18th October 2009, where Patrik Kittel warmed up his horse Scandic for approximately two hours, during which time the horse underwent various degrees of hyperflexion, with very little respite. Most notably, the horse's tongue hung outside the its mouth and was clearly blue, showing signs of hypoxia (a lack of oxygen to the tissue). This footage can be seen here.

In accordance with the FEI; prolonged or excessive use of the technique is not allowed. But how long is too long? Anky van Grunsven (dressage world champion 2008) was asked in an interview by Epona.tv how she uses her controversial trademark training tool. Do bouts of hyperflexion of the neck last for 10 seconds, a couple of minutes, ¼ of an hour or longer? "I don't have a clue," she admits. "Time flies when you are having fun".

Despite its use by highly successful international dressage riders such as Anky, the message from the riding public is clearly ‘ban rollkür', as shown by over 6000 signatories to a petition the FEI to ban hyperflexion at competition venues.

A further question should also be considered - is hyperflexion the sole problem, or should we be focusing on the training methods used to achieve hyperflexion? An interesting question is postured by Epona.tv in their article ‘Necks, Lies and Videotapes': "What if I taught my horse - using treats or carefully timed pressure and release - to bend its neck extremely for a brief moment?  That would be hyperflexion. But would it be evil? Should it be banned?".

It is proposed that hyperflexion itself is not the biggest problem, but rather the methods employed in order to achieve and sustain a hyperflexed frame. Hyperflexion can be achieved and perpetuated over long periods of time by the application of unyielding pressure from the bit and spurs. The reins are used to pull - and sometimes ‘saw' - on the horse's mouth until it assumes the hyper flexed position. The pressure is usually maintained to some extent for the duration of the hyperflexion. If the horse then tries to stretch its neck, it is corrected by applying an aversive stimulus, such as the curb bit and chain, or the spurs.

Epona.tv have documented this on film, and report that often the horses are made to go in the hyper flexed position - not for a few moments as sanctioned as acceptable by the FEI - but for several minutes at a time.

Over the years there has been very little study into the effects of rollkür. However there are a small handful of studies. The limited results available have tended to raise further questions, and the time is certainly right for further research to be conducted in this area.

Oldruitenborgh-Oosterbaan et al's (2006) study used eight riding school horses in two identical tests, one test ridden in rollkür and one in a natural frame. Heart-rate monitoring and blood sampling were used to collect quantitative data on workload and stress. There was a significant increase in heart-rate for those horses trained using rollkür at trot compared to those trained in natural frame at trot. Blood lactate and pH also rose significantly during rollkür while the remainder of the blood test showed no real change. The researchers interpreted their results as showing that while the workload was higher during rollkür, stress levels did not increase. However, the hyperflexion used in this test was neither pronounced nor as prolonged as would be used in Dressage horses. In a separate study, Ödberg & Bouissou (1999) noted that the use of draw reins can result in a loss of impulsion and false collection if used continuously.

One of the more recent pieces of research is a study conducted by von Borstel et al (2009) which takes a behavioural approach to quantify the effect of rollkür on the horse. Horses were ridden through a Y maze where one route resulted in the horse being ‘on the bit' and the other being in rollkür. During rollkür it was noted that the horses were moving notably slower and showed more behavioural signs of discomfort, such as tail swishing, head tossing or attempted bucks. The horses were ridden through the maze randomly alternating the sides about 30 times. After this initial period, the horses were left to choose which side to follow. Out of the 15 horses involved in this trial, 14 (93%) chose significantly more often the maze-arm associated with the non- rollkür frame.

It has been noted by researchers that further studies are needed to assess the reaction of horses to gradual training of rollkür, as opposed to coercively obtained hyperflexion, in order to decide whether the practice should continue to be allowed.

Given the limitations of the above studies, it can be identified that more definitive studies are required on stress, kinematics, and also the psychological effects on the horse. Restrictions on vision and respiratory function imposed on the horse during rollkür, and the possibility of the horse entering a state of learned helplessness should perhaps be studied.

As noted in the FEI report into the use of rollkür in international competition (FEI, 2006), further research should be conducted to confirm whether there are welfare issues involved in training techniques using hyperflexion. In ensuring this is conducted, the FEI could go some way towards re-establishing its commitment to prioritise the health and wellbeing of the competition horse.