Heart Rate Reduction by Grooming in Horses (Equus caballus)
Normando S, Haverbeke A, Meers L, Ödberg FO, Ibañez Talegón M
Department of Animal Nutrition, Genetics, Production and Ethology
Ghent University, Heidestraat 19, B-9820 Merelbeke, Belgium
Voice 32-9-2647804    Fax 32-9-2647849 frank.odberg@rug.ac.be

Introduction

Horses usually perform social grooming at a site dorsal of the neck and at the withers (Waring, 1983; Feh and de Mazières, 1993). Allo-grooming has been associated with physiological effects on heart rate (HR) (Lynch and Gantt, 1968; Lynch et al, 1974; Boccia et al, 1989; Aureli et al, 1999), immunity (Gust et al, 1996) and beta-endorphins (Keverne et al, 1989).

            Feh and de Mazières (1993), using eight adult Camargue horses and eight foals (gender not mentioned), found that manual imitation of grooming at the withers decreased the HR of the groomed individual significantly. Grooming at a non preferred site (shoulder) did not decrease HR as compared to rest frequency. They used a stethoscope and dictated on tape the number of heartbeats every 30 seconds. 

We proposed to repeat the Feh and de Mazières’ study using telemetric recording of HR while grooming four areas, using a more homogenous population of horses and by using a Latin Square design.

Methods

Subjects. Sixteen Spanish crossbred geldings (mostly Spanish x Selle Français), aged 3 to 12 years, were housed in individual loose boxes. They belonged to the Spanish Police in Madrid (Cuerpo Nacional de Policía). All were previously accustomed to wearing the girth with the electrodes.

Material. The Polar Accurex Plus Heart Rate Monitor (Polar®, Kempele, Finland) was used to telemetrically record HR, expressed as the mean rate per minute, measured at five seconds intervals. The electrodes (with gel) were fixed on a girth belt, which was fitted just behind the scapula: one on the cardiac area and another on the upper left thorax.

Protocol.  The following areas on the near side were groomed:

- At the base of the neck, 10 cm around the withers and 10 cm starting from the root of the "mane" (quotation marks indicate henceforth used term)

- The "withers"

- The "shoulder" (an area centred around the scapulo-omeral joint)

- The "hip" (an area just below the spina iliaca ventralis).

The shoulder is reported to be a site at which aggressive actions, such as bites and bite threats are often directed (Waring, 1983). The hip was chosen as a presumed neutral zone.

Recordings. The horses were not exercised nor fed one hour before recordings, which took place between 0900 and 1300 hours. Each horse was moved to a separate room in order to minimise disturbances and attached with a rope to the wall. The same person groomed all horses through a firm action of the fingertips.

Data summary and statistics. A Latin Square design with an extra period in four x four horses determined the sequences of the areas. Every three minutes of manual imitation of allo-grooming were followed by one minute of rest in order to minimise possible carry over effects (Morris, 1999). A three-minute control recording without grooming always preceded the Latin Square. Normality and equality of variance were checked before a two-way ANOVA assessed whether there were significant (p<0.05) differences between the horses’ HR when groomed on the different body locations. The ANOVA were followed by Multiple Range Tests (Newman-Keuls test) for the different areas with and without inclusion of the control period.

Results

Grooming all four areas significantly decreased (F₄=41.43, p<0.0001) HR compared to HR during the control period (mane: 10.8%, shoulder: 14.3%, hip: 15.2%, average: 15.6%), but the effect was significantly stronger when the withers were groomed (22.1%). The post-hoc Multiple Range Test (Newman-Keuls test) showed a difference between all the areas and the control period and between some of the areas (Table 1). Repetition of the test without including the control period yielded similar results. HR significantly differed also between individuals (F₁₄=147.38, p<0.0001) and periods (F₅=6.63, P<0.0001). HR continued to decrease during grooming, reaching a 22.45% reduction at period 6 (Table 2). 

Table 1: Effect on HR of grooming different areas

*Different letters denote significant difference (P<0.05) in HR (Multiple Range Test: Newman-Keuls test). Means are given ± SD.

            Table 2: Mean HR levels during the different successive periods of grooming 

Means are given ± SD

Discussion

These results confirm Feh and de Mazières' (1993) main conclusion that manual imitation of grooming the withers decreases HR. A decrease was also detected at areas where a negative or neutral reaction was expected, i.e. the shoulder and the hip. One should however consider the possibility of a conditioning effect. Horses petted regularly could associate an originally neutral or negative stimulus with some agreeable feelings until it acquires a positive value. This is in agreement with the reasoning that, as horses spontaneously rasp, lick or rub but do not slap, the possible rewarding effect of the latter (which is one of the most spontaneous ways humans reward a horse) develops through a conditioning procedure. 

The surprising weaker decrease in HR during grooming at the mane compared to the shoulder and the hip was due to one abnormally high HR (110 beats/min) recorded in horse 12 when tested in period 2. This horse was fed just after the experiment and may have perceived some anticipatory stimuli. Deleting this record decreases the average HR during grooming the mane to 40 beats/min, which is close to that found when grooming the withers. The average HR during period 2 drops to 44 beats/min, increasing the difference with the control period without affecting the habituation sequence.

The inhibiting effects found were much greater at the withers (22.1%) than at other areas (10.8%, 14.3% and 15.2%) but were generally stronger than those found by Feh and de Mazières (1993) (13.5% for foals and 11.4% for adult horses). The horses had not been exercised or fed one hour before each test but their average HR was rather high (52 beats/min) during the control period. A breed or climatic effect is possible, but the following factors could also be responsible: they were led away from their own stable, which often signals imminent work, to an area where they are sometimes treated or vaccinated; the presence of and manipulation by unfamiliar people; the effects of isolation (Jezierski, 1983); the electrode girth which could be associated with saddling in anticipation of work; a beep sounding every five seconds. 

HR decreased gradually from control period one down to the last grooming period (period 6, repetition of 5): 8.14%, 16.45%, 15.36%, 22.20% and 22.45%, suggesting habituation occurred. Absolute HR values have been affected, but the Latin Square design demonstrates that, relatively speaking, grooming site does have an effect. 

The main conclusion is that in horses grooming the withers has a calming effect.

Acknowledgements

Thanks are due to J Creten, RUCA, for statistical advice; S Martín Emmler, UCM, for her practical assistance; and the Cuerpo Nacional de Policía, Madrid for permission to use their horses.

References

Aureli F, Preston SD, de Waal FBM (1999) Heart rate responses to social interactions in free-moving rhesus macaques (Macaca mulatta).  A pilot study. J of Comp Psych 113(1): 59-65.

Boccia M, Reite M, Laudenslager M (1989) On the physiology of grooming in a pigtail macaque.  Physiol Behav 45(3): 667-670.

Feh C, de Mazières J (1993) Grooming at a preferred site reduces heart rate in horses. Anim Behav 46(6): 1191-1194.

Gust DA, Gordon TP, Wilson, ME, Brodie AR, Ahmed Ansari A, McClure HM (1996) Group formation of female pigtail macaques (Macaca nemestrina). Am J Primatol 39(4): 263-273.

Jeziersky T (1993) Effect of social isolation on heart rate in horses: Proceeding of the International Congress on Applied Ethology, Berlin 1993, pp 387-388.

Keverne E, Martenz N, Tuite B (1989) Beta-endorphin concentrations in cerebrospinal fluid of monkeys are influenced by grooming relationships. Psychoneuroendocrinology 14(1-2): 155-161.

Lynch JJ, Gantt WH (1968) The heart rate component of the social reflex in dogs. The conditional effects of petting and person. Cond Reflex 3(2): 69-80.

Lynch JJ, Fregin GF, Mackie JB, Monroe RR (1974) Heart rate changes in the horse to human contact. Psychopharmacology 11(4): 472-478.

Morris TR (1999) Experimental Design and Analysis in Animal Sciences. CABI Publishing, Wallingford.

Waring GH (1983) Horse Behavior. Noyes Publishing, Park Ridge, NJ.