Does social motivation overshadow fear caused by a sudden sound in horses?

doi:10.21203/rs.3.rs-2575920/v1

Living in a herd has multiple advantages for social species and is a primary survival strategy for prey. The presence of conspecifics, identified as a social buffer, may mitigate the individual stress response. Social isolation is, therefore, particularly stressful for horses, which are gregarious animals. However, they are not equally vulnerable to separation from the group. We tested whether more and less socially dependent horses and independent individuals would differ in their responses to novel and sudden sounds occurring in two contexts: non-social and social motivation. Twenty warmblood horses were first exposed to two social tests: to evaluate the level of social dependence (rate of restless behaviour; social isolation) and the quantity and the quality of interactions in which they were involved (stay on a paddock). Two fear audio tests were then performed to compare the responses to sudden sounds while feeding (non-social motivation; control trial) and while moving towards the herd (social motivation; experimental trial). Socially dependent horses showed more pronounced avoidance behaviour and needed much more time to resume feeding during the control trial. Hence, dependent individuals appeared to be more fearful. However, during an experimental trial, horses of both groups tended to ignore the sound or paid only limited attention to the stimulus, continuing to move forward towards their conspecifics. Thus, social motivation may overshadow fear caused by a frightening stimulus and make fearful and dependent horses more prone to face a potentially stressful event. This finding should be taken into account in horse training and management.

horse

gregariousness

social dependence

fearfulness

social buffer

Horses’ reactivity to different stressors and environmental stimuli is modified by the animal’s motivation, fearfulness or curiosity (Kozak et al. 2018). The variety of these responses may stem from differences in coping styles or temperamental traits (Visser et al. 2002) and has major implications in animal husbandry (Rothmann et al. 2014). Responding to suddenness and novelty may compromise both the safety of the handler and the welfare of a horse (Acton et al. 2020; Vidament et al. 2021). It is, therefore, crucial to be aware of the possible reactions of the horses (Corgan et al. 2021). Temperament is defined as any characteristic of an individual that appears to be stable over both time and situation (Roberts et al. 2016). Fearfulness and gregariousness, i.e. reactivity to social isolation from conspecifics (Lansade et al. 2008b), are among the most well-known and particularly interesting dimensions of equine temperament (Lansade et al. 2008a; Lansade et al. 2017; Górecka-Bruzda et al. 2022). Indeed, these traits relate directly to the evolution of the horse, which is a highly social prey animal that lives in a herd and remains in constant readiness to confront the threat (Marliani et al. 2021).

A stable and organised social structure, manifested by dominance and submissive interactions, enables horses to avoid the deleterious energy and health costs that would be incurred in aggressive encounters (Narciso et al. 2021). The group improves foraging strategies, provides learning opportunities (Mendonça et al. 2021) and facilitates localising and avoiding a potential threat (Hothersall and Casey 2012). For social species, living in a herd is the primary antipredator defence mechanism (Apfelbach et al. 2005). Although predation pressure is relaxed in domestic horses (Janczarek et al. 2021), they experience multiple anthropogenic stressors (Squibb et al. 2018). According to the risk-disturbance hypothesis, various sounds, objects and events may trigger a response analogous to the presence of a predator (Frid and Dill 2002). In this context, social support from other members of the herd consists of social buffering that mitigates individual stress levels (Marliani et al. 2021, Hartmann et al. 2012).

Horses are highly motivated to any kind of physical contact with other individuals (Søndergaard et al. 2011). They strengthen relationships through affiliative approaches, mutual grooming or keeping proximity (Wolter et al. 2018). However, if direct contact is not possible, they at least strive for indirect interactions (Yarnell et al. 2015). For horses, social isolation is a particularly strong stressor that negatively affects their behavioural and physiological reactivity (Kay and Hall 2009; Żelazna and Jezierski 2018). When returning to the group after a previous restriction of social opportunities, horses may manifest a rebound effect, which refers to increased intensity of interactions with conspecifics (Christensen et al. 2002a). Stress elicited by social isolation may even overshadow mild somatic pain (Reid et al. 2017). However, horses do not respond to the separation from the group with the same strength (Wolff et al. 1997). Socially dependent individuals have difficulty functioning when out of the herd. In contrast, independent animals are not excessively bonded to other horses or humans and are self-reliant when left away from the herd (Burattini et al. 2020). In the previous study (Janicka et al. 2022), we showed that the motivation for social interaction rapidly decreased the potential of a self-applied acoustic stimulus in creating virtual barriers.

Here, we assessed the differences in reaction to the sudden sounds occurring in two different contexts (social and non-social motivation) between the horses differing in the level of social dependence. Given the significance of the social motivation, we hypothesised that dependent and independent individuals would differ in their fear responses in the non-social sound test but not the social motivation sound test. We predicted that social motivation, which is the willingness to remain in contact with a group (Lansade et al. 2008b), would overshadow the fear caused by a sudden sound.

Ethical statement

All experimental procedures were carried out in accordance with the European directive 2010/63/EU and Polish laws - the Act of January 23, 2021 - Law on the organization of breeding and reproduction of farm animals (Journal of Laws 2021, Item 36) and were approved by the Local Ethics Review Committee for Animal Experiments (no 58/2022), Lublin, Poland.

Animals and living conditions

The study involved 20 adult warmblood horses (aged 6 – 16 years), comprising ten mares and ten geldings. The horses knew each other since they had been housed in the same equestrian centre for at least two years. They were released to an extensive pasture or occasionally to several paddocks for about six hours a day, where they could engage in social interaction. All of the subjects were familiar with the paddocks and pastures used in the study. Thus, habituation to the environment could have been omitted.

The subjects were kept in individual box stalls bedded with wheat straw and equipped with a manger, hay feeder, salt lick and an automatic waterer. They were fed three times a day with a mixture of oats and bran and had unlimited access to hay. The horses worked under the saddle for a maximum of two hours a day, six days a week. The subjects were under permanent veterinary control and were regularly observed by an experienced caretaker during daily handling. No physical or behavioural disorders were found before or after the study.

The tests were carried out on calm, windless days (< 0.3 m/s) during the spring period of 2022.

Social tests

Prior to the main part of the experiment (Sound tests), the horses were subjected to two different social tests to assess the level of social dependence and engagement in social interactions. The tests were conducted on separate, consecutive days.

Social dependence test – arena test

Dependence on the other members of the social group was assessed with a modified arena test (Seaman et al. 2002). The test consisted of separation from a herd that stayed on the pasture about 300 m from a tested subject. Horses were individually released to a small paddock (400m²) and were observed for a subsequent five minutes using the focal animal sample method (Altmann 1974). This time appeared to be sufficient because individual differences in reactions to an unusual situation were shown to be the strongest within the first few minutes (Wolff et al. 1997). The most frequent behavioural variables observed during isolation stress are vocalisation, defecation, locomotion and vigilance (Lansade et al. 2008b) and these were included in the ethogram (Table 1).

Table 1 Behaviours recorded during herd dependence test based on modified ethogram worked out by Seaman et al. (2002)

Behaviour	Description
Sustained walk (s)	Walking energetically, looking in front or around
Trot (s)	A two-beat gait
Canter (s)	* A three-beat gait
Vigilance (s)	Standing still with an elevated neck, intently orientated head and ears
Whinny (freq.)	* A high amplitude call of long duration that fluctuates in frequency and is given on expiration, produced during social separation, among others (Laurijs et al. 2021)
Snoring (freq.)	* A non-vocal inhalation sound, a stress-releasing behaviour (Laurijs et al. 2021)
Defecation (freq.)	The elimination of faeces

* added behaviours/modified descriptions

Since behaviours assessed as frequency (whinny, snoring and/or defecation) were displayed only by a few horses, the level of social dependence was established based on the percentage of time of displaying other restless behaviours. According to that, horses were divided into three groups:

independent (I): < 15% (n = 9) of the observation time was accounted for by restless behaviour – horses able to function without the social support of conspecifics (Stachurska et al. 2021),
moderately dependent (MD): 15 – 50% (n = 5),
dependent (D): > 50% (n = 6) – horses not self-reliant, reactive to social isolation from conspecifics (Burattini et al. 2020).

Social interaction test – introduction to a paddock

To determine the engagement in social interactions, the horses were observed just after being released into a large paddock (5000 m²) in the morning (8.00 a. m.). The procedure was repeated on three consecutive days, with a single observation session lasting ten minutes. Affiliative and agonistic interactions (McDonnell 2003) (Table 2) in which each focal animal was involved were recorded every 30 s using the one-zero sampling method (Altmann 1974). The total number of affiliative, aggressive/threatening and submissive behaviours exhibited by each horse was then calculated. The measurement of spatial proximity (horses standing with body contact or within two horse-lengths), commonly used to assess social bonds (Wolter et al. 2018), was not taken into account because of the short-term character of the social test, whereas horses show a mean latency of changing the spatial distribution of group members every 8 min (Christensen et al. 2002b).

Table 2 Behaviours recorded during social interaction test (McDonnell 2003)

Affiliative behaviour	Aggressive and threatening behaviour
mutual grooming (nipping/nuzzling/rubbing each other’s neck, mane, chest, back, rump or tail) laying the head on the neck, back or rump approach (forward movement toward another horse and staying close) following (moving along behind another horse, no attempt to attack) playing with conspecifics (any reaction or sequence of behaviour during play in motion, occurring socially)	head, bite or kick threat bite, kick head bump strike (one or both forelegs rapidly extended forward to contact another horse, often accompanied by a squeal or snort) rear (raising of the forelegs into the air, supporting the body on the hind legs only) herding (combination of head threat and ears laid back with forward locomotion) chase (pursuit of another horse, usually at a gallop)
* Submissive behaviour – avoidance of aggressive/threatening interaction

Sound tests

Sound tests were the startling fearfulness tests, assessing horses’ reactions to a sudden auditory stimulus (Seaman et al. 2002). However, these tests were conducted in two different contexts: without (a control trial) and with (an experimental trial) a social motivation.

Non-social motivation sound test (NMST) – a control trial

The horses were individually subjected to a controlled assessment of their fear response to a new and unexpected sound while eating (non-social motivation sound test; NMST). The testing arena (140 x 30 m) was designated with tape for an electric fence on one of the known paddocks. It was located next to another paddock, providing proximity (approximately 50 m) to conspecifics during the test to avoid the impact of social isolation stress and lack of concentration on the food provided. The horses in the neighbouring paddock were habituated to the test sound a few days before the study.

A plastic container for the later provision of attractive food was placed at the front of the testing arena. A loudspeaker (JBL Charge 4, rated power of 30W) was positioned 1 m in front of the container, allowing the horse a long escape distance after the sound playback, if necessary. Prior to the test, horses were taught to feed from the container.

Each horse was released into a testing arena 15 minutes before the start of the test to habituate to the experimental setting. The experimenter then brought the horse to the start point (3 m from the container). At this time, the second experimenter put chopped carrots inside the container. The horse was unleashed and could start feeding, and the experimenters left the arena. After 30 s, a sudden and unknown sound (sound A; squeal of a pig, 20s, 40 dB) was played via Bluetooth by a Samsung Galaxy A02s device (Samsung Electronics Co., Ltd., South Korea).

Latency to respond to the stimulus was proved to be the objective and easy measure of horses’ responsiveness in fear tests (Górecka-Bruzda et al. 2011). Therefore, latency to resume feeding (s) was measured in the current study. If a horse did not approach the food within 300 s, it was given 301 s (Górecka-Bruzda et al. 2011). Additionally, a 5 – point behavioural scale describing horses’ reactions after sound playback was developed. Receiving 5 points indicated the highest level of stress (Table 3).

Table 3 Behavioural scale used to assess horses’ responsiveness during the sudden sound test

Points	Behavioural response
1	The horse pays no attention to the sound, continues to take food in an uninterrupted manner and is calm.
2	The horse raises its head, puts up its ears and resumes feeding. It is not spooked, and it does not walk away from the food.
3	The horse lifts its head, ears up, walks away a few steps, looks at it, approaches cautiously and starts to eat, may snoring.
4	The horse abruptly removes its head and jumps back. Spins, impatient, may toss head, hesitantly approaches food, still spooked.
5	The horse abruptly jumps back, walks away/runs a considerable distance, snorts, is afraid to approach food for a very long time (hesitant, tense) or does not approach at all.

Social motivation sound test (SMST) – an experimental test

The social motivation sound test (SMST) was a modified version of the sound tunnel test developed in the previous study (Janicka et al. 2022) to assess the potential of the sounds in creating electric-free virtual fences. During the test, the horses moved through the same corridor (55 m length, 5 m width, designated with tape for an electric fence) leading towards a known pasture, where the other members of the herd (except for individuals included in the study) were staying. The start-end of the corridor was closed after the introduction of a horse to a start zone (first 5 m), whereas the finish-end was permanently open, allowing the animal to leave it after each run. The experimental paddock, at which the corridor was designated, was separated from the pasture by metal railings. There was a loudspeaker placed outside the corridor at the finish line. The day before the SMST, horses were brought into the experimental paddock for two hours to familiarise themselves with the corridor.

SMST was conducted in the morning hours on three consecutive days (three trials). During each trial, horses went through the corridor at least three times: 1) being led by an experimenter, 2) freely, without an experimenter, and 3) freely, but with playing a sudden, unknown sound (sound B; futuristic characteristics, 20s, 100 dB) when the individual reached a distance of 10 m from the finish line. The initial two trials evaluated the horse's willingness to walk towards the pasture and other horses located approximately 150 meters away. If the horse exhibited this behaviour, a third trial was conducted to observe its reactions.

We showed that the motivation for social interaction rapidly decreased the audio virtual fencing effect (Janicka et al. 2022). Thus, we focused on subtle differences in reaction to the sound between horses with different levels of social characteristics. The sound barrier effect, completion of the SMST and behaviour after completion of the test were recorded (Table 4).

Table 4 Behavioural variables assessed during social motivation sound test

Behaviour	Description
Barrier effect (scale 1 – 5)	Reaction to the sound; 1 - continuing forward movement, 2 – slowing down, 3 – stopping, 4 – going away, 5 – running away
Completion of the test (scale 1 – 2)	Leaving a corridor before the end of the 20 s sound; 1 – no, 2 – yes
Finish behaviour (scale 1 – 6)	Behaviour after completion of the test; 1 – free walking/resting near the finish line; 2 – free walking/resting away from the finish line, 3 – looking around nervously + no locomotion, 4 – walking + looking around nervously, 5 – trot/canter + looking around nervously, 6 – not reaching a finish line within 20 s of the sound

Statistical analyses

The statistical analysis was performed with the SAS 9.4 software (version 9.4, SAS Institute Inc., Cary, NC, USA). Since the data did not conform to a normal distribution, as determined by the Kolmogorov-Smirnov and Shapiro-Wilk tests (p < 0.05), non-parametric tests were applied. Correlations between the results of behavioural tests were checked with Spearman's rank correlation. Kruskal-Wallis test was used to evaluate if social dependence affected the horses’ responses to the sounds and engagement in affiliative and dominant-submissive interactions. Dwass-Steel-Critchlow-Fligner (DSCF) multiple comparisons were then performed to check for the significance of differences between the groups. Additionally, the differences between mares and geldings during the tests were checked (Mann-Whitney U test). Even though the significance of results was calculated via non-parametric tests, for better clarity and readability for readers, the results were presented in tables as mean value ± standard error of the mean (SE). The minimum level of significance for all tests used was accepted at p < 0.05.

Sex differences

Mares were more restless when separated from the group during the herd dependence test (Mann-Whitney U test; U = 256.50, p = 0.0037), and needed more time to resume feeding after sound playback in NMST (U = 297.00, p = 0.0234) and responded to this sound more violently (they had a higher score on the behavioural scale) (U = 243.00, p = 0.0019) than geldings (Table 1). Showing affiliative interactions was common and comparable (U = 427.50, p = 0.7412). Although mares showed submissive behaviour more often (U = 297.00, p = 0.0234), agonistic interactions were rarely presented. No differences between mares and geldings were found in response to the sound during SMST (p > 0.05). The sound barrier effect (stopping/limiting further movement) was weak, and both mares and geldings finished the test with a similar frequency. The only difference concerned behaviour after completion of the SMST. Mares were more agitated while waiting to be reunited with the herd (U = 260.00, p = 0.0045).

Table 5 Sex differences during social and sound tests

Variable	Mare	Gelding
Restless behaviour (%)	45.37 ± 6.78 b	24.07 ± 5.71 a
Affiliative interactions (freq.)	21.50 ± 1.72 a	22.30 ± 2.19 a
Dominance behaviour (freq.)	3.50 ± 0.67 a	3.80 ± 0.60 a
Submissive behaviour (freq.)	3.00 ± 0.29 b	2.00 ± 0.22 a
Latency to resume feeding in NMST (s)	156.40 ± 26.87 b	52.50 ± 16.23 a
Behavioural scale in NMST (1-5)	3.60 ± 0.25 b	2.50 ± 0.19 a
Barrier effect in SMST (1-5)	2.27 ± 0.26 a	1.90 ± 0.21 a
Completion of the SMST (1-2)	1.80 ± 0.07 a	1.83 ± 0.07 a
Finish behaviour in SMST (1-6)	4.00 ± 0.33 a	2.00 ± 0.36 b

Means (± SE) marked with different letters differ significantly between mares and geldings at p < 0.05 (Mann-Whitney U test); NMST – non-social motivation sound test, SMST – social motivation sound test

Dependence on the social group and need for social interactions

Most of the horses (n = 11) showed more or lesser signs of stress when they were physically isolated from the group (Table 6). Of all the other horses, those categorised as dependent spent the most time demonstrating a sustained walk (DSCF = 3.8730, p = 0.0170; DSCF = 5.0767, p = 0.0010; in comparison to moderately dependent and independent, respectively). The time of vigilance was similar in the dependent and moderately dependent groups (DSCF = 0.2582, p = 0.9818) and definitely longer than in independent horses (DSCF = 4.1779, p = 0.0088; DSCF = 4.2567, p = 0.0074, respectively). Canter, whinnies and defecation rarely occurred and did not differ significantly between the groups (p > 0.05). Dependent horses trotted (DSCF = 3.9485, p = 0.0145) and snorted (DSCF = 3.7255, p = 0.0229) more than independent horses but not moderately dependent individuals (DSCF = 2.7749, p = 0.1217; DSCF = 2.9863, p = 0.0875, respectively). However, snorting was also rarely noted.

Although horses responded differently to social isolation, the quantity and the quality of interactions during their stay on a paddock (social interaction test) were similar between the groups (p > 0.05). The horses usually displayed affiliative interactions, whereas dominance and submissive behaviours were rarely observed. For this reason, agonistic behaviours were not included in further analysis.

Table 6 Comparison of the time budget of horses with different levels of social dependence based on the social dependence test

Behaviour	Dependent (N = 6)	Moderately dependent (N = 5)	Independent (N = 9)
Time budget
Sustained walk (s)	188.67 ± 22.04 c	34.00 ± 7.31 b	0.00 ± 0.00 a
Trot (s)	23.33 ± 8.44 b	3.60 ± 2.91 ab	1.22 ± 1.22 a
Canter (s)	2.67 ± 1.26 a	2.40 ± 1.47 a	0.56 ± 0.55 a
Vigilance (s)	45.83 ± 10.36 b	46.00 ± 13.85 b	9.44 ± 3.02 a
Whinny (freq.)	1.00 ± 0.52 a	1.60 ± 1.60 a	0.00 ± 0.00 a
Snoring (freq.)	4.67 ± 1.61 b	0.60 ± 0.60 ab	0.22 ± 0.15 a
Defecation (freq.)	0.33 ± 0.21 a	0.00 ± 0.00 a	0.11 ± 0.11 a
Frequency of social interactions
Affiliative interactions	23.67 ± 3.29 a	24.80 ± 4.69 a	19.11 ± 4.34 a
Dominance behaviour	4.17 ± 1.92 a	2.00 ± 0.71 a	4.22 ± 1.14 a
Submissive behaviour	3.83 ± 0.54 a	2.00 ± 0.71 a	1.88 ± 0.35 a

Means (± SE) marked with different letters differ significantly in rows at p < 0.05 (DSCF)

Reactions to sounds

Horses’ responses to a sudden sound in NMST varied, but they generally interrupted feeding for some time and showed movement in the opposite direction. On the contrary, they rarely stopped forward movement and turned back towards the start zone in SMST. Their reactions were usually limited to slowing down after hearing a sound while moving towards the other herd members.

Dependent horses were more fearful during NMST than independent horses (Table 7). They needed much more time to resume feeding after the sound was played (DSCF = 4.3270, p = 0.0063) and had higher scores on the behavioural scale, indicating the strength of the fear response (DSCF = 3.9483, p = 0.0145). In spite of this, no differences were found between the groups during the SMST (p > 0.05). The sound barrier effect was poor and usually limited to slowing down or stopping for a while, and eventually, most of the horses finished the test.

Table 7 Impact of the social dependence of the horses on reactions to sounds

Behaviour	Social characteristics of the horses
	Dependent (N = 6)	Moderately dependent (N = 5)	Independent (N = 9)
Sound barier (SMST)	2.44 ± 0.35 a	1.80 ± 0.31 a	2.00 ± 0.23 a
Completion of SMST	1.83 ± 0.09 a	1.87 ± 0.09 a	1.78 ± 0.08 a
Finish behaviour (SMST)	3.00 ± 0.46 a	3.40 ± 0.49 a	2.89 ± 0.36 a
Refeeding latency (SST)	171.00 ± 32.18 b	121.60 ± 39.15 ab	50.55 ± 17.63 a
Behavioural scale (SST)	3.67 ± 0.27 b	3.20 ± 0.43 ab	2.55 ± 0.21 a

Means (± SE) marked with different letters differ significantly in rows at p < 0.05 (DSCF)

There were no correlations between the horses’ reactions in two different sound tests, except for the first reaction after the sound playback (Table 8). In the case of the horses that gained more points on the behavioural scale during NMST (stronger stress response), the sound barrier effect in SMST was greater (Spearman's rank correlation; r_s = 0.280, p = 0.0303). However, this correlation was weak.

Both refeeding latency (r_s = 0.405, p = 0.0013) and behavioural scale scoring (r_s = 0.431, p = 0.0006) were moderately correlated with the time spent on restless behaviour in the herd dependence test. Simultaneously, restlessness (indicating the level of herd dependence) and the responses during SMST were not correlated (p > 0.05).

There was no correlation between the frequency of affiliative interactions and social dependence (restlessness) (r_s = 0.1716, p = 0.1898). However, in the case of the horses that interacted more, the sound barrier effect was stronger (more points in the behavioural scale) (r_s = 0.46105, p < 0.0002) and the horses finished the test with a lower probability (r_s = -0.38931, p < 0.0021). They also received more points on the behavioural scale during NMST (r_s = 0.38527, p < 0.0024).

Table 8 Correlations between the results of behavioural tests

	Sudden sound test, herd dependence test and social motivation sound test
Sudden sound test and social tests	Latency (NMST)	Behav. (NMST)	Restles.	Barrier (SMST)	Compl. (SMST)	Finish (SMST)
Latency (NMST)	-	-	0.405 *	0.236	-0.095	0.171
Behav. (NMST)	-	-	0.431 *	0.280 *	-0.115	0.186
Restles.	-	-	-	0.048	0.086	0.129
Affiliative behaviour	0.217	0.385 *	0.171	0.461 *	-0.389 *	0.068

* statistically significant correlations; Abbreviations: latency (NMST) – refeeding latency in NMST, behav. (NMST) – behavioural scale in NMST, restless. – restless behaviour (%) in herd dependence test, barrier (SMST) – effect of sound barrier in SMST, compl. (SMST) – completion of SMST, finish (SMST) – finish behaviour in SMST; NMST – non-social motivation sound test, SMST – social motivation sound test

Social characteristics of the horses

Horses evolved as prey species for which living in a herd is a survival strategy (Marliani et al. 2021). Thus, isolation from conspecific is a particularly strong stressor for horses (Kay and Hall 2009), which was confirmed in the current study. The arena test revealed that most of the horses feel uncomfortable when left alone, but their responses vary from low to high levels of stress. Similar observations were made by Wolff et al. (1997), Górecka-Bruzda et al. (2022) and Lansade et al. (2008b). The isolation stress was mainly displayed by sustained walking, vigilance and trotting, whereas neighing was infrequent. However, snorting was presented almost exclusively by individuals identified as dependent. In turn, Lansade et al. (2008b) point to the frequency of neighing as the most interesting behavioural parameter, which was well correlated with other parameters measured in the same isolation tests, such as defecation, locomotion and vigilance. The differences in these observations may be due to the individual characteristics of the horses as well as to the various ages of tested horses (young vs. adult horses in the current study). The current study further confirmed that the arena test is an appropriate test to measure the level of social dependence (Wolff et al. 1997; Seaman et al. 2002), which is manifested by reactivity to the separation of the group (Le Scolan et al. 1997; Lansade et al. 2008b).

Horses are highly social animals and strengthen mutual bonds through affiliative behaviour (Wolter et al. 2018). After the introduction to a large paddock during the social interaction test, the horses presented a wide range of affiliative interactions, including mutual grooming, following, approach and motor playing, whereas agonistic interactions were rare and limited exclusively to threatening. Similarly, Kimura (1998) noticed that in some months of the year, dominance–submission interactions were not frequent. In the case of our study, it could stem from a short period of observation, and this should be taken into account in future studies. However, we aimed to conduct a simple and repeatable social test. The frequency of affiliative relationships did not vary between mares and geldings and horses of different levels of gregariousness. Thus, the need for interaction with conspecifics was similar in the group of tested subjects. As shown by Mendonça et al. (2021), social status and sex generally affect the kind of presented affiliative interactions, but this factor was not included in our study. Generally, the similar engagement in interactions of horses with different social characteristics stresses the importance of social contact with conspecifics. Stachurska et al. (2021) also noted that separating some mares from the group, regardless of their social rank, resulted in increased agitation among the other group members. In the current study, we noticed that, compared to geldings, mares were more restless when separated from the herd. This observation is in agreement with the study of Górecka-Bruzda et al. (2022), who showed that mares were more socially dependent than geldings.

Reactions to the sounds

As we predicted, horses of different levels of social dependence varied in their responses to the sudden sound during the test in a non-social (but not a social) motivation context when they moved towards a herd on a pasture. Horses classified as dependent and moderately dependent were more fearful compared to independent individuals in a non-social motivation context. They responded to the sound more violently (jumped back and walked/run away more often) and needed significantly more time to resume feeding than self-reliant animals. Briard et al. (2015) also noted that fearless horses were simultaneously less gregarious. In turn, Villalba et al. (Villalba et al. 2009) showed that sheep cautious in accepting new food were more reactive during social isolation. Indeed, in many species, these traits have been used to characterise ‘bold’ and ‘shy’ behavioural types (Briard et al. 2015). We found no correlations between the rate of restlessness during separation from the herd and the results of the social motivation sound test. Horses’ reactions were usually limited to slowing down after hearing the sound or eventually to stopping for a while. Even when hesitating, they decided to move forward to stay closer to the herd. Thus, the motivation to join the group was too strong for an unexpected, novel sound to stop both independent and dependent horses from further movement. Jouven et al. (2012) showed that some of the sheep, previously trained not to exceed the virtual fence, followed their conspecifics regardless of the electric shocks they received. Since, in our study, dependent and moderately dependent individuals showed a greater level of stress during a control sound test and then reacted similarly to independent horses in the social motivation sound test, we may conclude that the strong need to stay in proximity to members of the group overshadows fear caused by a frightening stimulus. As reported by Burattini et al. (2020), boldness and independence are two important behavioural traits that influence horses' fearfulness, assertiveness and sociability. Socially dependent horses are not at ease when left alone, away from the herd. Instead, they rely more on their conspecifics (Burattini et al. 2020). Thus, by comparing the results of a sound test in social and non-social motivation contexts, we confirmed the role of conspecifics as a social buffer and the importance of social motivation to confront a stressor. Ricci‑Bonot et al. (2021) and Christensen et al. (2008) also demonstrated the calming effect of a companion. We conclude that this factor should be considered in horse management when handling or schooling animals. For example, in the study of Dai et al. (2019), horses were taught loading into a trailer in groups, which allowed their gregarious nature to be used in the learning process.

Similar observations were made for the factor ‘sex’. Mares were definitely more anxious during a control sound test but reacted the same as the geldings during the sound test in the social motivation context. Thus, willingness to be reunited with conspecifics overshadowed the effect of a frightening stimulus. The only difference concerned behaviour after completion of a test, when the horses were waiting to join the herd. Again, the mares were more agitated, similar to the social isolation test (social dependence test). Lansade et al. (2022) noted that weaned horses still preferred their dam as well as other mares from their natal group after five months of separation, but this tendency was stronger in fillies than in colts. A stronger need to stay in a group in the case of mares may be due to the social structure of the species. In the wild, horses lived in groups comprising mainly females, and this is probably why social support is particularly important for them (Górecka-Bruzda et al. 2022).

Another important observation we made is the correlation between the quantity of affiliative behaviours and responses during two sound tests. Horses more engaged in affiliative interactions were more agitated after sound playback in a non-social motivation context, were also more sensitive to the appearance of a sound in social motivation context and finished that test with a lower probability. Taking into account the fact that horses with different levels of gregariousness made a similar number of social interactions, this tendency was due to greater fearfulness rather than social dependence. We further confirmed that social dependence does not result from the need for direct interactions but rather from the very need to stay close to the herd and is best observed during social isolation (Le Scolan et al. 1997; Wolff et al. 1997; Seaman et al. 2002; Lansade et al. 2008b; Lansade and Simon 2010).

A weak correlation was observed between the first reaction to sound playback in the two sound tests. For the horses which showed stronger stress responses in a control sound test, the sound barrier effect in the social motivation sound test was greater. However, no other correlations were found between the results of the two different sound tests. Thus, the response to a specific stimulus in one test allowed the horse's response in the second test to be predicted only partially. Lansade et al. (2008c) showed that within each sense, the greater the horses’ response to one stimulus, the greater their response to the other. However, in the current study, the same sensory stimulus was used in two different situations – social and non-social motivation. It is, therefore, important to regard the context of presenting a testing stimulus during behavioural observations.

Horses are highly motivated to stay close to their conspecifics, but they differ in their responses to social isolation. Socially dependent horses are not at ease being left alone outside the herd, in contrast to those more independent. However, while on the paddock, they engage in social interactions in a similar way. Dependent individuals are generally more fearful than independent animals. However, when facing sudden, novel audio stimulus in a social motivation context (motivation to join a herd), they react similarly to more self-reliant horses. Thus, social motivation may overshadow fear caused by a frightening event in horses. This finding should be taken into account in horse training and management.

Authors’ contribution statements: The conception for the paper was conceived by Wiktoria Janicka and Izabela Wilk. The experiments were designed by Wiktoria Janicka and Izabela Wilk and performed by Wiktoria Janicka. The data were analyzed by Wiktoria Janicka and Tomasz Próchniak. The paper was written by Wiktoria Janicka and was revised critically by Izabela Wilk and Tomasz Próchniak. All authors read and agreed to the published version of the manuscript.

Funding: This research was supported by project no. SD/61/ZiR/2022 provided by University of Life Sciences in Lublin, Poland.

Competing interests: The authors have no competing interests to declare that are relevant to the content of this article.

Ethics approval: All experimental procedures were carried out in accordance with the European directive 2010/63/EU and Polish laws - the Act of January 23, 2021 - Law on the organization of breeding and reproduction of farm animals (Journal of Laws 2021, Item 36) and were approved by the Local Ethics Review Committee for Animal Experiments (no 58/2022), Lublin, Poland.

Acton AS, Gaw CE, Chounthirath T, Smith GA (2020) Nonfatal horse-related injuries treated in emergency departments in the United States, 1990–2017. Am J Emerg Med 38:1062-1068. https://doi.org/10.1016/j.ajem.2019.158366
Altmann J (1974) Observational study of behavior–Sampling methods. Behaviour 49:227-267.
Apfelbach R, Blanchard CD, Blanchard RJ, Hayes RA, McGregor IS (2005) The effects of predator odors in mammalian prey species: A review of field and laboratory studies. Neurosci Biobehav Rev 29:1123-1144. https://doi.org/10.1016/j.neubiorev.2005.05.005
Briard L, Dorn C, Petit O (2015) Personality and Affinities Play a Key Role in the Organisation of Collective Movements in a Group of Domestic Horses. Ethology 121:888-902. https://doi.org/10.1111/eth.12402
Burattini B, Fenner K, Anzulewicz A, Romness N, McKenzie J, Wilson B, McGreevy P (2020) Age-Related Changes in the Behaviour of Domestic Horses as Reported by Owners. Animals 10:2321. https://doi.org/10.3390/ani10122321
Christensen JW, Ladewig J, Søndergaard E, Malmkvist J (2002a) Effects of individual versus group stabling on social behaviour in domestic stallions. Appl Anim Behav Sci 75:233-248. https://doi.org/10.1016/S0168-1591(01)00196-4
Christensen JW, Malmkvist J, Nielsen BL, Keeling LJ (2008) Effects of calm companion on fear reactions in naïve test horses. Equine Vet J 40:46-50. https://doi.org/10.2746/042516408X245171
Christensen JW, Zharkikh T, Ladewig J, Yasinetskaya N (2002b) Social behaviour in stallion groups (Equus przewalskii and Equus caballus) kept under natural and domestic conditions. Appl Anim Behav Sci 76:11-20. https://doi.org/10.1016/S0168-1591(01)00208-8
Corgan ME, Grandin T, Matlock S (2021) Evaluating the Reaction to a Complex Rotated Object in the American Quarter Horse (Equus caballus). Animals 11:1383. https://doi.org/10.3390/ani11051383
Dai F, Costa AD, Bonfanti L, Caucci C, Di Martino G, Lucarelli R, Padalino B, Minero M (2019) Positive Reinforcement-Based Training for Self-Loading of Meat Horses Reduces Loading Time and Stress-Related Behavior. Front Vet Sci 6:350. https://doi.org/10.3389/fvets.2019.00350
Frid A, Dill L (2002) Human–caused Disturbance Stimuli as a Form of Predation Risk Conserv Ecol 6:11. https://doi.org/10.5751/ES-00404-060111
Górecka-Bruzda A, Jastrzębska E, Drewka M, Nadolna Z, Becker K, Lansade L (2022) Female Horses are more socially dependent than geldings kept in riding clubs. Appl Anim Behav Sci 254:105714. https://doi.org/10.1016/j.applanim.2022.105714
Górecka-Bruzda A, Jastrzębska E, Sosnowska Z, Jaworski Z, Jezierski T, Chruszczewski MH (2011) Reactivity to humans and fearfulness tests: Field validation in Polish. Appl Anim Behav Sci 133:207-215. https://doi.org/10.1016/j.applanim.2011.05.011
Hartmann E, Søndergaard E, Keelinga LJ (2012) Keeping horses in groups: A review. Appl Anim Behav Sci 136:77-87. https://doi.org/10.1016/j.applanim.2011.10.004
Hothersall B Casey R (2012) Undesired behaviour in horses: A review of their development, prevention, management and association with welfare. Equine Vet Educ 24:479-485. https://doi.org/10.1111/j.2042-3292.2011.00296.x
Janczarek I, Stachurska A, Kędzierski W, Wnuk-Pawlak E, Wilk I, Zyglewska K, Paszkowska A, Ryżak M, Wiśniewska A (2021) Heart rate variability in Konik and purebred Arabian horses in response to different predator vocalisations. Animal 15:100045. https://doi.org/10.1016/j.animal.2020.100045
Janicka W, Wilk I, Próchniak T, Janczarek I (2022) Can Sound Alone Act as a Virtual Barrier for Horses? A Preliminary Study. Animals 12:3151. https://doi.org/10.3390/ani12223151
Jouven M, Leroy H, Ickowicz A, Lapeyronie P (2012) Can virtual fences be used to control grazing sheep? Rangel J 34:111-123. https://doi.org/10.1071/RJ11044
Kay R, Hall C (2009) The use of a mirror reduces isolation stress in horses being transported by trailer. Appl Anim Behav Sci 116:237243. https://doi.org/10.1016/j.applanim.2008.08.013
Kimura R (1998) Mutual grooming and preferred associate relationships in a band of free-ranging horses. Appl Anim Behav Sci 59:265-276. https://doi.org/10.1016/S0168-1591(97)00129-9
Kozak A, Zięba G, Tietze M, Rozempolska-Rucińska I (2018) Consistency of emotional reactivity assessment results obtained in different behavioural tests. Appl Anim Behav Sci 205:54-60. https://doi.org/10.1016/j.applanim.2018.05.013
Lansade L, Bouissou M-F, Erhard HW (2008a) Fearfulness in horses: A temperament trait stable across time and situations. Appl Anim Behav Sci 115:182-200. https://doi.org/10.1016/j.applanim.2008.06.011
Lansade L, Bouissou M-F, Erhard HW (2008b) Reactivity to isolation and association with conspecifics: A temperament trait stable across time and situations. Appl Anim Behav Sci 109:355-373. https://doi.org/10.1016/j.applanim.2007.03.003
Lansade L, Lévy F, Parias C, Reigner F, Górecka-Bruzda A (2022) Weaned horses, especially females, still prefer their dam after five months of separation. Animal 16:100636. https://doi.org/10.1016/j.animal.2022.100636
Lansade L, Marchand AR, Coutureau E, Ballé C, Polli F, Calandreau L (2017) Personality and predisposition to form habit behaviours during instrumental conditioning in horses (Equus caballus). PLoS ONE 12:e0171010 https://doi.org/10.1371/journal.pone.0171010
Lansade L, Pichard G, Leconte M (2008c) Sensory sensitivities: Components of a horse’s temperament dimension. Appl Anim Behav Sci 114:534-553. https://doi.org/10.1016/j.applanim.2008.02.012
Lansade L, Simon F (2010) Horses’ learning performances are under the influence of several temperamental dimensions. Appl Anim Behav Sci 125:30-37. https://doi.org/10.1016/j.applanim.2010.02.010
Laurijs KA, Briefer EF, Reimert I, Webb LE (2021) Vocalisations in farm animals: A step towards positive welfare assessment. Appl Anim Behav Sci 236:105264. https://doi.org/10.1016/j.applanim.2021.105264
Le Scolan N, Hausberger M, Wolff A (1997) Stability over situations in temperamental traits of horses as revealed by experimental and scoring approaches. Behav Processes 41:257-266. https://doi.org/10.1016/S0376-6357(97)00052-1
Marliani G, Sprocatti I, Schiavoni G, Bellodi A, Accorsi PA (2021) Evaluation of Horses’ Daytime Activity Budget in a Model of Ethological Stable: A Case Study in Italy. J Appl Anim Welf Sci 24:200-213. https://doi.org/10.1080/10888705.2020.1857252
McDonnell SM (2003) The equid ethogram: A practical field guide to horse behavior. Eclipse Press, Lexington
Mendonça RS, Pinto P, Inoue S, Ringhofer M, Godinho R, Hirata S (2021) Social determinants of affiliation and cohesion in a population of feral horses. Appl Anim Behav Sci 245:105496. https://doi.org/10.1016/j.applanim.2021
Narciso MHPM, da Luz MPF, Maia CM, Nicolau J, Filho PP (2021) Dominance and Leadership in the Equine Social Structure: A Preliminary Study About Mules and Sex Influence. J Equine Vet Sci 99:103392. https://doi.org/10.1016/j.jevs.2021.103392
Reid K, Rogers CW, Gronqvist G, Gee EK, Bolwell CF (2017) Anxiety and pain in horses measured by heart rate variability and behavior. J Vet Behav 22:1-6. https://doi.org/10.1016/j.jveb.2017.09.002
Ricci‑Bonot C, Romero T, Nicol C, Mills D (2021) Social buffering in horses is influenced by context but not by the familiarity and habituation of a companion. Sci Rep 11:886. https://doi.org/10.1038/s41598-021-88319-z
Roberts K, Hemmings AJ, Moore-Colyer M, Parker MO, McBride SD (2016) Neural modulators of temperament: A multivariate approach to personality trait identification in the horse. Physiol Behav 167:125-131. https://doi.org/10.1016/j.physbeh.2016.08.029
Rothmann J, Christensen OF, Søndergaard E, Ladewig J (2014) Behavior Observation During Conformation Evaluation at a Field Test for Danish Warmblood Horses and Associations with Rideability and Performance Traits. J Equine Vet Sci 34:288-293. https://doi.org/10.1016/j.jevs.2013.06.007
Seaman SC, Davidson HPB, Waran NK (2002) How reliable is temperament assessment in the domestic horse (Equus caballus)? Appl Anim Behav Sci 78:175-191. https://doi.org/10.1016/S0168-1591(02)00095-3
Søndergaard E, Jensen MB, Nicol HJ (2011) Motivation for social contact in horses measured by operant conditioning. Appl Anim Behav Sci 132:131-137. https://doi.org/10.1016/j.applanim.2011.04.007
Squibb K, Griffin K, Favier R, Ijichi C (2018) Poker Face: Discrepancies in behaviour and affective states in horses during stressful handling procedures. Appl Anim Behav Sci 202:34-38. https://doi.org/10.1016/j.applanim.2018.02.003
Stachurska A, Wiśniewska A, Kędzierski W, Różańska–Boczula M, Janczarek I (2021) Behavioural and Physiological Changes in a Herd of Arabian Mares after the Separation of Individuals Differently Ranked within the Dominance Hierarchy. Animals 11:2694. https://doi.org/10.3390/ani11092694
Vidament M, Lansade L, Danvy S, Priest BDS, Sabbagh M, Ricard A (2021) Personality in young horses and ponies evaluated during breeding shows: Phenotypic link with jumping competition results. J Vet Behav 44:1-11. https://doi.org/10.1016/j.jveb.2020.09.003
Vilanova XM, Beaver B, Uldahl M, Turner PV (2021) Recommendations for Ensuring Good Welfare of Horses Used for Industrial Blood, Serum, or Urine Production. Animals 11:1466. https://doi.org/10.3390/ani11051466
Villalba JJ, Manteca X, Provenza FD (2009) Relationship between reluctance to eat novel foods and open-field behavior in sheep. Physiol Behav 96:276-281. https://doi.org/10.1016/j.physbeh.2008.10.010
Visser EK, van Reenena CG, van der Werf JTN, Schilderc MBH., Knaapd JH, Barneveld A, Blokhuis HJ (2002) Heart rate and heart rate variability during a novel object test and a handling test in young horses. Physiol Behav 76:289-296. https://doi.org/10.1016/s0031-9384(02)00698-4
Wolff A, Hausberger M, Le Scolan N (1997) Experimental tests to assess emotionality in horses. Behav Processes 40:209-221. https://doi.org/10.1016/s0376-6357(97)00784-5
Wolter R, Stefanski V, Krueger K (2018) Parameters for the Analysis of Social Bonds in Horses. Animals 8:191. https://doi.org/10.3390/ani8110191
Yarnell K, Hall C, Royle C, Walker SL (2015) Domesticated horses differ in their behavioural and physiological responses to isolated and group housing. Physiol Behav 143:51-57. https://doi.org/10.1016/j.physbeh.2015.02.040
Żelazna R, Jezierski T (2018) Behavioural reactions of horses (Equus caballus) to separation stress in conspecifics. A pilot study on emotional contagion in the horse. Anim Sci Pap Rep 36:333-338. https://doi.org/10.3390/fi13100250

No competing interests reported.

Does social motivation overshadow fear caused by a sudden sound in horses?

Status:

Journal Publication

Version 1

Abstract

Introduction

Materials And Methods

Ethical statement

Social tests

Sound tests

Statistical analyses

Results

Sex differences

Dependence on the social group and need for social interactions

Reactions to sounds

Discussion

Social characteristics of the horses

Reactions to the sounds

Conclusions

Declarations

References

Additional Declarations

Status:

Journal Publication

Version 1