Horses can be affected by peer pressure, a study has found.
Stabling a horse next to one showing behaviours such as cribbing or wind-sucking can increase the likelihood of similar behaviour in the other.
The conclusion was reached after a study involving 287 horses at nine Hungarian riding schools.
Behaviours in horses are considered abnormal when they occur without any primary function and can be damaging to the health or performance of the animal.
Such behaviours are called abnormal stereotypic behaviour (ASB), or stereotypies.
The most common are crib-biting, wind-sucking, weaving, and box walking. They are known to cause or contribute to gastric ulcers, colic, tooth wear, weight loss, and fatigue.
Such behaviour is seen are in up to 5% of horses, studies have shown.
The authors of the study, headed by Krisztina Nagy and entitled "Possible influence of neighbours on stereotypic behaviour in horses", noted that other studies had already shown social isolation, housing, management conditions, and feeding regime had a strong effect on developing such behaviours.
There was, they said, a common belief that exposure to horses behaving in this way increased the risk of it occurring in horses that observe, although it has never been substantiated.
The team's research showed that having a neighbouring horse behaving in this way represented a "significant risk factor" in determining whether a horse will adopt such behaviour.
Also, aggressive behaviour towards other horses increased the odds of stereotypy in the aggressor.
Such behaviours are difficult to change once established, so preventing them from developing is important, the authors said.
"This process is reversible by environment enrichment, but only at early life," they said.
The effect of neighbours on horses with such established behaviours is controversial, they said.
"The aim ... was to investigate risk factors of abnormal stereotypic behaviour (ASB) in horses. We show that exposure to a stereotypic neighbour may have significant effect on the odds of horses performing abnormal stereotypies.
The researchers observed horses known to show such behaviours at all the riding schools in the study, focusing on housing, management conditions, food regime, stereotypies, and problematic behaviour performed by the individual horse or by a horse in its sight.
In the 287 horses sampled, the prevalence of crib-biting/wind-sucking was 4.53% (13 horses), wood-chewing 10.10% (29 horses), weaving 2.79% (8 animals), and box-walking 3.83% (11 animals).
The occurrence of any of the four listed abnormal behaviours was 16.70%, or 48 horses in total. Aggression towards horses was noted in 17.07% (49 horses) and aggression towards people in 9.41% (27 animals).
Behaviour problems during riding were noted in 9.06% of the horses.
Stereotypies were more common in non-competition horses compared to competition horses (33 out of 126 non-competition horses and 12 out of 161 competition horses).
Researchers revealed several risk factors for developing such behaviours, including the presence of a horse that was already weaving, crib-biting or box-walking, and the presence of an aggressive horse in the neighbouring stall. Other risk factors included receiving oats more than twice a day and more than two riders using the one horse.
The authors said the presence of the behaviours in the Hungarian riding schools used in the study was no different from that reported in other countries.
"However, factors associated with [these behaviours] revealed by our analysis are in apparent contrast with previous epidemiological surveys."
Management methods restricting natural behaviour have been suggested as major risk factors for developing these behaviours.
They pointed to earlier studies that indicated mouth-related behaviours - crib-biting, wind-sucking and wood-chewing - are mostly associated with diet and the restriction of normal grazing behaviour, whereas locomotor stereotypies such as box-walking and weaving might be activated by social isolation or inadequate physical exercise.
"It is not surprising that many risk factors reported by others were not identified in our study, because management practices in the studied nine riding schools were very similar and lacked most of the risk factors identified by previous studies; and other factors, like behaviour of the neighbouring horses, have not been included in previous surveys.
"Only two management conditions emerged as risk factors for ASB. First, receiving concentrates more than twice a day increased the odds of wood-chewing and ASB in general, in accordance with previous studies.
"Second, having more than two riders increased the odds of weaving, but diminished the odds of box-walking.
"The number of riders as a possible risk factor has not been studied previously ... We can only speculate that horses ridden by many riders are more stressed (increasing the odds of weaving) but also receive more exercise, thus reducing the odds of box-walking.
"Disentangling the possible effects of number of riders, amount of exercise and riding by strangers needs further investigations."
All other risk factors, they said, such as aggression toward horses or presence of a neighbour performing aggression or some kind of stereotypy, are behavioural variables.
The nature of the relation between aggression and stereotypies is poorly understood. Dominant horses are usually more aggressive, and another study showed foals of dominant mares are more likely to develop abnormal behaviour.
Others believe that stereotypies and aggression are a common consequence of frustration.
"The precise nature of the possible effect of neighbours on ASB should be studied experimentally before concrete suggestions can be given for practical management. "However, as the majority of managers believe that stereotypies are not just inherited but also learned from other horses, we cannot avoid the responsibility to make some general remarks."
The nature of the study meant the authors could not say horses displaying abnormal behaviours caused such behaviour in neighbours, only that it increased the risk of it occurring.
"Our aim in presenting these findings was to draw attention to the possibility of such effects so that other researchers would include this variable in their surveys.
"Even if such effects are substantiated by other studies, isolating stereotypic horses is bad management because social deprivation enhances stress and attenuates ASB.
"Careful monitoring of the horses for early signs of [this] behaviour is important to prevent the establishment of stereotypies.
"Horses susceptible to developing stereotypies might be moved away from stress agents, including stereotypic neighbours."