Lying behavior in horses in relation to box size

doi:10.1016/j.jevs.2005.11.015

Journal of Equine Veterinary Science

Volume 26, Issue 1, January 2006, Pages 11-17

https://doi.org/10.1016/j.jevs.2005.11.015 Get rights and content

Abstract

To analyze if horses' lying behavior is influenced by the size of their lying area, 8 horses were kept, first, in a large box—(2.5 × height of the horse)²m²—and then in a small box—(1.5 × height of the horse)²m²—or vice versa. After a 5-day adaptation period, the lying behavior (frequency and duration of sternal and lateral recumbency, rotating behavior just before lying down, and rolling behavior just before getting up) was observed from video recordings on 3 consecutive nights. The results showed that the duration of sternal recumbency was significantly longer in the large boxes than in the small boxes (P = .002). Furthermore, box size exerted an influence on the frequency of the rolling behavior shown before getting up, but box size was not the only factor affecting this behavior. We conclude that observations of the resting behavior of horses are necessary to make recommendations for the minimal size of stalls for horses.

Introduction

In contrast with other farm animals, horses are often housed in old buildings. In addition, the size of modern riding horses is increasing. As a result, the size of the boxes in which most horses are kept is relatively small. According to Danish recommendations, the box area measured in m² should be at least twice the height at the withers squared, and the shortest side of the box should be at least 1.5 times the height at the withers. Recommendations in other European countries are similar. In the UK, the British Horse Society recommends a box size of 3.6 × 3.6 m. In Sweden, the Swedish Board of Agriculture recommends that the minimum area for a small horse should be 8.0 m², the shortest side being 2.35 m, and for a large horse 9.0 m², the shortest side being 2.5 m. For horses higher than 1.7 m at the withers, the area in m² should be at least 1.8 times the height at the withers squared. However, all of these recommendations are based on practical experience rather than systematic observation of the behavior of horses, particularly their lying behavior.

Pedersen et al¹ conducted an experiment on the influence of bedding material on the time horses spend recumbent. The experiment revealed that most of the horses made a rolling behavior before getting up. It was suggested that this behavior is due to difficulties in getting up, possibly because of too small a box size. Except for a brief note in Waring,² this rolling behavior before getting up has, to our knowledge, not been previously described. Therefore, the question is raised whether the behavior is normal or whether it is a result of too small a box size. If so, the behavior could possibly be used as an indicator of how big a box should be to allow normal resting behavior in horses. In addition, the rolling behavior increases the risk that the horse gets stuck against the box wall, a condition that is rare but can be fatal.

The number of studies on equine sleep is limited. The main focus has been on the entity of sleep and its effect on the behavior of the horse or on the effect of the behavior of the horse on its sleep.³ Although many studies have examined sleep in animals, its purpose is still unclear. Some studies suggest that sleep might serve as a thermoregulatory function,4, 5 that it plays a role in memory consolidation,6, 7, 8 permits vigilance recovery,⁹ conserves energy,4, 10 and is a restorative process.¹⁰ These studies also show that short-term deprivation of sleep will lead to an increase in total sleep time during recovery.

These and other studies make it obvious that systematic observations are necessary to ensure that horses are able to perform normal sleeping behavior, and to obtain an optimal amount of sleep, when reconsidering the recommendations for the minimal box size for horses.

The objective of the study was to analyze the effect of the box size on the lying behavior of horses and to observe whether the frequency of the rolling behavior before getting up is influenced by the size of the box. Based on video recordings, the duration and frequency of sternal and lateral recumbency of 8 horses kept both in a small and in a large box were calculated and the behavior before lying down and getting up was observed. Since determination of sleep necessitates electroencephalographic recording, particularly determination of slow wave sleep and rapid eye movement sleep (REM), observations in the current study were limited to body position rather than the occurrence of sleep.

Section snippets

Materials and methods

The study was done at Søborggaard Horse Centre in the northern part of Sealand (Denmark) from October 2003 to February 2004.

Statistics

For the statistical analysis, SAS was used. Because the data set was not normally distributed, it was transformed to normality with a Box-Cox transformation, with γ = 0.3 for the statistical analysis. The statistical program Proc Mixed was used for analysis with this model. To test for correlations between the occurrence of the rolling behavior and the horses' height at the withers, girth, and abdomen, a Spearman rank correlation test was done.

Lying behavior

The mean total duration of recumbency of each horse per night was 140 minutes in the large box and 135 minutes in the small box (Fig 1). When the estimated average duration of lying episodes was analyzed (ie, after transformation of the data set), there was a significant difference between the large box and the small box (P = .002; Table 2), in that the horses spent more time recumbent in a large box. When lying was divided into sternal and lateral recumbency, it was found that the horses spent

Discussion

The results of our study show that horses are recumbent for a longer time in the large boxes than in the small ones. Although the difference is small, it was found to be statistically significant (P = .002; Table 2). Furthermore, box size exerts an influence on the rolling behavior shown before getting up. When examining the treatment (ie, from small to large box or vice versa), it was revealed that the horses that are moved from a large to a small box perform the rolling behavior almost 3

Conclusion

We conclude that box size affected the lying behavior of horses in that they spent more time recumbent in a large box than in a small one. Although the difference was statistically significant, it was found to be so small that it is questionable whether it has any implications for the horse. We further conclude that box size was a factor that affected the occurrence of the rolling behavior before getting up and that the change was more pronounced in horses that went from a large box to a small

References (28)

PedersenGR et al.
The influence of bedding on the time horses spend recumbent
J Equine Vet Sci
(2004)
BergerRJ et al.
Energy conservation and sleep
Behav Brain Res
(1995)
RechtschaffenA et al.
Sleep deprivation in the rat by disk-over-water method
Behav Brain Res
(1995)
GiudittaA et al.
The sequential hypothesis of the function of sleep
Behav Brain Res
(1995)
SmithC
Sleep states and memory processes
Behav Brain Res
(1995)
DukasR et al.
Sustained vigilance and animal performance
Anim Behav
(1995)
DallaireA
Rest behavior
Equine Pract
(1986)
HouptKA et al.
Night-time behavior of stabled and pastured peri-parturient ponies
Appl Anim Behav Sci
(1986)
WebbWB et al.
Attempts to modify the sleep patterns in the rat
Physiol Behav
(1971)
MetzJHM
The reaction of cows to a short-term deprivation of lying
Appl Anim Behav Sci
(1985)

CarsonK et al.

Equine behaviour, II: a review of the literature on feeding, elinimative and resting behaviour

Appl Anim Ethol

(1983)

HaleLA et al.

The electroencephalogram of the normal “grade” pony in sleep and wakefulness

Comp Biochem Physiol

(1980)

DewasmesG et al.

Pattern of rapid-eye movement sleep episode occurrence after an immobilization stress in the rat

Neurosci Lett

(2004)

SchiffelholzT et al.

Novel object presentation affects sleep-wake behavior in rats

Neurosci Lett

(2002)

Cited by (43)

Assessing the cost efficiency and benefits of sawdust, pinewood, and rubber wood shavings as bedding materials
2024, Journal of Equine Veterinary Science
This study aimed to evaluate the benefits of using sawdust, pinewood shavings, and rubber wood shavings as horse bedding materials. Nine horses were used in a repeated measure design, where each horse was provided with each of the bedding materials for two consecutive weeks. The weight, cost, mucking out time, ammonia emissions, and urine holding capacity were measured and calculated for each bedding material. Results showed that pinewood shavings had the highest urine-holding capacity, while rubber wood shavings had the lowest absorbency for urine. Pinewood shavings also had lower ammonia emissions compared to sawdust and rubber wood shavings but were significantly more expensive. Sawdust was the cheapest of the three bedding materials, but it was used in greater quantities, which required more time for mucking out. The findings from this study could be used to make informed decisions on bedding materials for stables, taking into consideration the trade-off between cost, maintenance requirements, and labor hours.
The Behavior of Horses Stabled in a Large Group at Essential Resources (Watering Point and Lying Halls)
2022, Journal of Equine Veterinary Science
Citation Excerpt :
Stabled horses also spend more time lying in the sternal than in the lateral position. Furthermore, the sternal position could be extended by increasing the size of single boxes [50]. Likewise, space allowance of the littered area affects lying behavior in group-housed horses [13].
Since there is a practical need, this study focused on the behavior of 51 horses kept in one large group in order to make management suggestions regarding the animal:watering point ratio and the design of lying halls. Drinking events and interactions between pairs of horses were analyzed for 18 days as well as the occupancy time of the watering point. The time period had significant effects on drinking events with visible interactions (P < .05). The highest least square means (LSM) and standard errors (SE) of drinking events with and without visible interaction were observed between 12 noon-2 p.m. (13.6 ± 2.2 and 8.6 ± 1.6 drinking events, respectively) which was also the time period of main occupancy of the watering point (41.9 ± 2.7 min/2 h). About 57% of drinking events were classed as with visible interaction, of which over 90% were mild (i.e. laying back of ears, tightening of face). Horses’ behavior (resting while standing/others) and lying positions (sternal/lateral) in different lying halls were analyzed over 36 days by generalized mixed models. The sternal position was preferred (12.4% ± 1.4%–39.2% ± 3.0%) compared to the lateral one (1.4% ± 0.4%–8.3% ± 1.6%). Older horses spent significantly less time in the sternal position (13.8% ± 1.6%) and rested more standing (45.4% ± 3.3%) compared to younger horses (26.4% ± 4.5% and 29.1% ± 4.9%, respectively) (P < .05). Overall, this study provides first insights into drinking and lying behavior of horses kept in a large group.
Horses’ resting behaviour in shelters of varying size compared with single boxes
2022, Applied Animal Behaviour Science
Citation Excerpt :
Full rolling prior to rising has been observed in horses on pasture (Hansen et al., 2007), indicating that a full roll may be a comfort behaviour when there is enough space. In this study, frequency of rolling prior to rising behaviour varied from 14% to 55%, compared with around 30% in other studies (Pedersen et al., 2004; Raabymagle and Ladewig, 2006; Hansen et al., 2007). How rising behaviour varies between housing systems and whether it is a potential indicator of welfare status need to be studied in more detail.
Lying behaviour in horses is affected by factors such as bedding, housing and available lying area. For group-housed horses, social factors may influence access to available lying area and affect their ability to meet their need for sleep, including essential REM sleep. REM sleep can only be achieved when the whole body, including the head, is supported by the ground, so lateral recumbency is important to meet horses’ sleep requirements. This study investigated the effects of available lying area in shelters on horses’ lying and rising behaviour, on disturbance behaviour by horses, and on lying bouts by individual horses. Lying and rising behaviour was video-recorded for eight horses in single boxes (control treatment) and in an open-barn with three available lying area of 8, 18 and 28 m²/horse, respectively in the shelters. The results revealed significantly less lateral recumbency in the shelter with 8 m² lying area/horse (22 min, p = 0.04) compared with the single boxes (52 min), and a tendency for more lateral recumbency with 18 m² lying area/horse (48 min, p = 0.07) compared with 8 m² lying area/horse. Rising without prior rolling was the most common rising behaviour in the single boxes. Frequency of rolling prior to rising varied from 14% to 55% for all housing systems, compared with previous observations of ~30% irrespective of available lying area. This may be due to inter-individual differences, indicating a need for detailed studies of rising behaviour. Lying behaviour was affected by the behaviour of other horses and also significantly affected by available lying area. With more available area in the shelter, horses lay down for almost twice as many bouts (p = 0.01) and for almost twice as long as compared to a smaller area (p = 0.001). Number of lying bouts (p = 0.001) and behaviour during rising from the lying position were also affected by available lying area. It is therefore likely that the space requirement to meet horses’ need for rest will be larger in group-housed horses than for horses in individual boxes.
A Preliminary Study Investigating the Influence of Auditory Stimulation on the Occurrence of Nocturnal Equine Sleep-Related Behavior in Stabled Horses
2019, Journal of Equine Veterinary Science
The physical environment is known to influence nocturnal behavioral time budgets of the stabled horse, but less evidence exists to suggest how this might be affected by including additional sensory stimuli. This study aimed to establish the impact of novel auditory stimuli on the frequency of equine sleep-related behavior. Seven horses stabled for 24 hours per day on the same yard receiving the same daily management routine were observed from 2030 to 0630 over nine nights. Frequency of nocturnal behavior was recorded using focal intermittent sampling against a predetermined ethogram and an infrared CCTV camera system. Data were recorded under the following conditions: without music for two nights (phase A1), exposure to music for five nights (Beethoven's ninth Symphony) played at an average of 62.3 decibels (phases B1 [nights 3-4] and B2 [nights 6-7]), and two further nonconsecutive nights (phase A2) when music was no longer played. A general linear model was used to determine differences in the frequency of parametric behavioral data with a significantly higher occurrence of “ingestion” (F [3,18] = 7.910, P = .001) during phases in B compared with A, and a significant decrease in the occurrence of “other” behavior (F [3,18] = 10.25, P = .000) comparing phase A1 with all other phases. The Wilcoxon signed rank test highlighted significant differences in the frequency of “lateral recumbency” between specific phases (P < .05). The addition of music appears to have a significant effect on the equine nocturnal time budget that might be beneficial from an equine sleep perspective.
Stall architecture influences horses’ behaviour and the prevalence and type of stereotypies
2019, Applied Animal Behaviour Science
Citation Excerpt :
Here we demonstrate that even when there is a small possibility for close social contact, this outdoor view is a source of more agitation and chronic stress-related behaviours (study 1) and that the effect is immediate (study 2). This is especially remarkable in the broodmare population where this was the only difference between the two types of stalls: mares that had grids on the half door performed less stereotypic behaviours and spent more time foraging and resting, the latter behaviours known to occur more in calm and positive situations (Benhajali et al., 2009; Greening et al., 2013; Kwiatkowska-Stenzel et al., 2016; Greening et al., 2013; Pessoa et al., 2016; Raabygmale and Ladewig, 2005) than those who could have the head outside and watch out-of-reach conspecifics and humans walking around. Another remarkable result is that the type of stereotypic behaviours performed varied according to stall architecture.
Despite the spatial and social restrictions it causes, single stall housing still prevails in sport and riding school horses, leading to the emergence of abnormal behaviours such as stereotypic or abnormal repetitive behaviours (SB/ARB). In the present study, we investigated the impact of the type (visual/tactile) and amount of social information that could be exchanged (i.e. distance between the individuals) on the expression of welfare indicators, including, but not limited to, STB. Additional observations were made on the production of snorts, recently described as a potential indicator of positive emotions, according to the type of stall horses were housed in. Two complementary studies were performed. One observational study on 32 sport horses, all living in the same place, being of the same breed and sex, whose aim was to compare the behaviours of horses maintained for a long time in two types of stalls differing mostly in the possibilities of contact with close neighbours versus looking outdoors. The second, experimental study, consisted in moving purebred Arab broodmares from one condition to another randomly every day for 66 days, the two types of stalls differing only by the possibility or not to put the head outside above the open top half door. The results show clear statistical relations between stall architecture and horses’ behaviour, especially STB, their prevalence and type differing according to the type of stall in both studies. Overall, the access to outdoor visibility and its degree (possibility to put the head out or not) had a major effect on the horses’ behaviours, which was the same in both studies, despite the differences between populations in terms of breed, sex and type of work. The experimental study also reveals that changes in behaviours can be rapid after a change of housing.
Assessing equine emotional state
2018, Applied Animal Behaviour Science
Citation Excerpt :
This phenomenon of Post Inhibitory Rebound behaviour has been reported in horses post-confinement by both Freire et al. (2009) and Chaplin and Gretgrix (2010). The occurrence of Rebound behaviour has been shown to be subject to moderation through management activities such as the provision of social contact and stable size (Raabmagle and Ladewig 2006), but worryingly has also been shown to have serious impacts on horse welfare and human safety (e.g. Pessoa et al., 2016). The emotional mediation in rebound behaviour was highlighted by McGreevy and Nicol (1998) in their investigation of the motivational consequences of short-term prevention of crib-biting.
The scientific study of animal emotion has recently become an important focus for animal behaviour and welfare researchers. For horses used by humans for work, recreation or sport, the question of the significance of their life experiences in terms of their emotional response, is an important one if we are to provide for their welfare needs. Horses have received relatively less scientific attention than many livestock species when it comes to investigating emotional state or affective experience, although their behavioural responses during sporting or recreational performance are often described anecdotally using terminology indicating an underlying presumption of equine emotions. Indeed, the international governing body for equestrian sport, the Fédération Equestre Internationale (FEI), include the concept of ‘the Happy Equine Athlete’ into their rules, as a key objective during training and competition.
This review presents available evidence to date of the physiological, behavioural and cognitive components of equine emotion and evaluates the extent to which the question concerning ‘how horses feel’ can be answered. The characterization of equine emotion in terms of level of arousal and valence, based on physiological, behavioural and cognitive indicators, offers a way forward to determine the impact of different situations and experiences on horses during their working lives. There is a need to develop robust validated methods for accessing equine emotions, to underpin a universally agreed method for/approach to providing an accurate assessment of equine welfare that can be utilized in a variety of contexts. This will provide a means of monitoring and improving the horse’s experience, ensuring that the horse enjoys a good life, rather than one that is just worth living.

View all citing articles on Scopus

: Refereed

View full text

Scientific PaperLying behavior in horses in relation to box size

Abstract

Introduction

Section snippets

Materials and methods

Statistics

Lying behavior

Discussion

Conclusion

J Equine Vet Sci

Behav Brain Res

Behav Brain Res

Behav Brain Res

Behav Brain Res

Anim Behav

Equine Pract

Appl Anim Behav Sci

Physiol Behav

Appl Anim Behav Sci

Appl Anim Ethol

Comp Biochem Physiol

Neurosci Lett

Neurosci Lett

Scientific Paper
Lying behavior in horses in relation to box size