Skip to main content
NCBI home page
Current Zoology logoLink to Current Zoology
. 2025 Aug 18;72(3):419–427. doi: 10.1093/cz/zoaf056

Trade-offs between proximity and physical contact during group integration in pigs (Sus scrofa domesticus)

Sunil Khatiwada 1,✉, Victoria E Lee 2, Simon P Turner 3, Irene Camerlink 4,✉,b
Editor: Zu-Shi Huang
PMCID: PMC13290397  PMID: 42358757

Abstract

During behavioral trade-offs, individuals have to decide whether to express a behavior which may lead to a reward or potential costs when engaging in a risky situation. Social integration forces animals to make such trade-offs. We hypothesized that animals predominantly demonstrate nontactile behavior and hence less tactile behavior in a high-risk context such as during social integration, while using tactile behavior more than nontactile in less risky situations such as under social stability. Pigs (Sus scrofa domesticus) typically are in close physical contact to each other, but physical contact also relates to increased aggression. We investigated 18 groups (142 pigs) across different phases of social stability, thereby observing snout proximity and snout contact. Additionally, aggression (reflecting costs) and growth performance (reflecting benefits) were measured. Data were analyzed using mixed models while accounting for group stability. Snout proximity was indeed most frequent during social instability and reduced as stability increased, while snout contact remained more constant. The high occurrence of snout proximity during social instability suggests conflict avoidance and thus risk aversion. Animals that showed more frequent snout proximity grew slower, while initiators and recipients of frequent snout contact had a better growth performance. The causality of these effects cannot be ascertained, but it is possible that slower growing, and thus weaker individuals may have made a behavioral trade-off by choosing proximity rather than contact during social instability. The results further emphasize the importance of distinguishing between nuances in behavior.

Keywords: aggression, behavioral trade-offs, group dynamics, social behavior, social nosing, Sus scrofa

Understanding the implications of variation in behavioral expression, particularly those important for building social connections, advances our knowledge of mammalian societies. The ability to trade between behaviors based on perceived risks and benefits suggests that animals can prioritize one behavior over another.

Graphical Abstract

Graphical Abstract.

For image description, please refer to the figure legend and surrounding text.

Open in a new tab

Behavioral expression, a signaling process that transmits information from initiators to recipients (Demartsev et al. 2023), can assist social members in various scenarios, such as in recognizing group members and maintaining social relationships. Animals are expected to balance the costs and benefits from decisions in any given context (Vázquez et al. 2020; Carsten et al. 2023) and attempt to initiate interactions via a signaling process that can reduce costs (Tibbetts and Dale 2007). To transmit information, group-living animals have developed different forms of communication, such as nontactile (e.g., olfactory and vocal) and tactile contact (e.g., physical aggression), of which some may pose more risk than others.

Tactile behavior is common in many species’ behavioral repertoire, such as muzzle contact in vervet monkeys (Chlorocebus pygerythrus) (Nord et al. 2021), post-conflict affiliative muzzle contact in domestic goats (Capra hircus) (Schino 1998), reconciliatory contact in mandrills (Mandrillus sphinx) (Schino and Marini 2011), and mouth–mouth interaction in beluga calves (Delphinapterus leucas) (Hill et al. 2019). Individuals derive benefits from tactile behavior throughout life as it contributes to the maintenance and formation of social bonds (Uvnäs-Moberg 1998). Gentle tactile behavior is argued to serve a function in regulatory, affiliative and communicative processes (Gliga et al. 2019) and increases endogenous oxytocin secretion contributing to health and development (Uvnäs-Moberg 1998) including growth (Uvnäs-Moberg and Petersson 2005; Champagne et al. 2008). However, tactile behavior can also lead to aggression which may have serious consequences, and therefore nontactile contact may in some contexts be preferred to minimize risks of injury. It is therefore beneficial for social species to engage in their relevant species-specific gentle tactile behavior (e.g., allogrooming, snout contact, licking) or nontactile behavior depending on the situation.

Given the potential costs and benefits of tactile behavior, it is of great importance for the individuals of social species to be able to switch between tactile and nontactile behavior. Switching between tactile and nontactile behavior, and adopting the strategy that bears the least risk can provide benefits such as increased information gathering (Partan and Marler 2005). This may be particularly important during risky situations, such as during a time of social instability where both the potential costs (Prevolnik et al. 2021) and benefits of tactile behavior may be increased. The establishment of new social groups, resulting in social instability, is in domestic pigs a common occurrence due to the way they are managed (Coutellier et al. 2007). Pigs form dominance relationships (McGlone 1985) and use, amongst others, olfactory signals for social recognition (Kristensen et al. 2001). For this they require to be in close proximity because of the functioning of the vomeronasal organ (VNO) (Schild and Rørvang 2023). It has been proposed that nontactile snout proximity relates to social recognition while tactile snout contact relates at least in part to affiliative behavior and may benefit the recipient's health in the form of an increased growth performance (Camerlink et al. 2012; Camerlink and Turner 2013). However, it may also risk the costs of receiving aggression.

The objective of this study was to investigate the occurrence of snout proximity and snout contact during social encounters in pigs, while exploring the costs and benefits individuals may accrue from these behaviors. We hypothesized that pigs demonstrate more nontactile snout proximity in a situation where the associated costs of direct contact are high, and express tactile snout contact when the situation is less risky. We predict more nontactile snout proximity behavior when the risks are high, which is in the early stage of group integration when there is frequent aggressive behavior, and that it reduces with increasing group stability (prediction 1). At the same time, we predict that tactile snout contacts increase with increasing social stability (prediction 2), as evidenced through low aggression. Further, we hypothesize that these behaviors will relate to individual costs and benefits in terms of growth performance. If true, then subtle differences in social behavior (proximity versus contact) can have opposite outcomes for growth performance and will be important to consider in the study of animal behavior.

Materials and methods

Animals and housing

Male (n = 72) and female pigs (n = 70) (Large White × Landrace sows × Danish Duroc boars) (n = 18 groups) were studied on the Easter Howgate Pig Research Unit across three batches (i.e., cohorts), starting when animals were 4 weeks of age. Approximately half of the piglets (10–14 days old) were socialized preweaning by allowing two litter groups to mingle (the piglets could interact, but the sows could not) through an opening between two farrowing pens before the start of the experiment. The experiment started when pigs (4 weeks of age) were weaned from the sow and transferred to another research building. Here, different sibling groups were combined to form new social groups. Groups were composed of eight pigs (males and females) from two different litters unfamiliar to each other (four siblings per litter), thereby resulting in each pig being housed with three familiar and four unfamiliar pigs. The sex ratio and number of socialized pigs per group were balanced where possible but due to variation in birthdate and number of males and females this deviated from a 1:1 ratio in approximately half of the groups. The groups were either housed in enriched conditions (n = 9 groups; pen of 6.9 m × 2.1 m, with deep straw bedding, Porcichew pig toys and hessian ropes) or in less-enriched conditions (n = 9 groups; pen of 4.6 m × 2.1 m, with rubber mats and wood shavings but no straw or toys). All pigs within a group were exposed to the same housing condition. Sibling groups were balanced across housing conditions. Pigs were fed ad libitum commercial pig feed and had water available via nipple drinkers. All pigs were weighed regularly, and average daily weight gain (ADWG) in kg per day was calculated for week 4–5 of age, week 6–8 and week 10–11, where ADWG = (weight last week − weight first week)/days between weeks. To enable individual identification, pigs were ear-tagged and received a spray mark on their back with animal marker spray. In the course of the observation period, three individuals were taken off trial and removed from the observed groups due to health issues.

Behavioral observations

The behaviors snout proximity, snout contact and aggressive behavior were recorded following the ethogram in Table 1, while noting both the initiator and recipient. When an initiator approached and made contact to a recipient within <1 s, the behavior was scored as snout contact without scoring snout proximity separately. Behavioral observations were conducted on all pigs from week 4 until week 11 (without observation in week 9) using continuous behavioral sampling live and from video (Bateson and Martin 2021). Each group was observed continuously for 15 min per session, with 3–4 observation sessions per day. This resulted in 45 min to 1 h of continuous observation per group per day. Each study week included two observation days (except week 4 in which 3 days were observed), resulting in 2 h of observations per group per week. Observations were randomized and done across 7 weeks (resulting in ca. 14 h of observations per group). The behaviors were recorded by a single observer, using the “Animal Behavior pro” application installed on an ipad mini 2.

Table 1.

Ethogram of observed behaviors in 4–11 weeks old pigs (n = 142).

Behavior Description
Snout proximity Pig proactively brings its snout within 30 cm distance of the snout of another pig, by turning its head towards the other, without making contact (irrespective of the duration).
Snout contact Pig proactively touches the snout (area below the eyes, including nose disc) of another pig with its nose disc (irrespective of the duration).
Aggression Pig delivers a head knock, threat, push, and/or bite to another pig, or engages in a mutual fight, including a rapid sequence of biting. Can include moments in which the opponents stand still (pause) to regain breath. Continues until one retreats or until other behavior (excl. standing still) is performed for at least 3 s.
Open in a new tab

Dominance relationships

From the overall frequency of aggressive interactions, the Modified David Score (MDS) was calculated as an indicator of dominance rank. The Modified David Score is an adaptation of the original David's Score, which is a method used to rank individuals in a social group based on their dominance interactions. The MDS was calculated using the “steepness” package (version 0.3-0) (Leiva and de Vries 2022) in R (version 4.3.2) (R Core Team 2023). Due to the relatively infrequent occurrence of aggression in some weeks, dominance relationships could not be reliably calculated by week (to show the change over time) and were instead calculated across the whole observation period.

Social stability

From literature it is well-known that pigs at regrouping show a peak of aggression, after which there is a gradual reduction in agonistic interactions across the 3 weeks following (Meese and Ewbank 1973; Prevolnik et al. 2021). Therefore, observation weeks were chosen at different stages of group stability. Weeks 4–5 of age (weeks 1–2 after regrouping) represents social instability as they will need to form new dominance relationships (Meese and Ewbank 1973; McGlone 1985). Weeks 6–8 of age (weeks 3–5 after regrouping) represents medium social stability, as most dominance relationships will have formed but some changes in hierarchy may still occur (Meese and Ewbank 1973). Weeks 10–11 of age (weeks 7–8 after regrouping) represents high social stability.

As aggression was too low in some weeks to calculate the stability of dominance positions, we verified the stability of the phases by the frequency of aggression. Analysis of differences in aggression between the social stability phases was conducted using a mixed model with aggression as response variable, phase as predictor variable and batch as random variable. This showed that the low and medium stability phases had significantly more aggression than the high stability phase (low vs. high: P < 0.001, medium vs. high: P < 0.001) (Figure 1). The low and medium stability phase did not significantly differ (P = 0.23; Figure 1) but were kept separate as the early phase contains higher outliers and keeping them separate aligns with the time required for dominance establishment as reported in the literature (Meese and Ewbank 1973). Data for behavior and average daily weight gain (ADWG) were therefore divided into these three phases for social stability: low (weeks 4–5 of age), medium (weeks 6–8 of age) and high (weeks 10–11 of age).

Figure 1.

Boxplot showing frequency of aggression in pigs across low, medium, and high social stability.

Open in a new tab

Frequency of aggressive behavior, based on behavioral observations of agonistic behavior as described in Table 1, across phases of social stability: low stability (1–2 weeks after regrouping, 4–5 weeks of age), medium stability (3–5 weeks after regrouping, 6–8 weeks of age), and high stability (7–8 weeks after regrouping, 10–11 weeks of age).

Data analyses

There was a moderate correlation between initiating and receiving behaviors, both for snout proximity (r = 0.31, P < 0.001), snout contact (r = 0.38, P < 0.001), and aggressive behavior (r = 0.32, P < 0.001) (Cor.test function). This suggests that individuals who initiate these behaviors are also the recipients of the behavior, as the behavior may be reciprocated. Based on the correlation, we included only information from the initiator in analyses of social behavior while taking into account both initiators and recipients for the association with ADWG. Additionally, correlation analysis was performed between MDS and ADWG, and between aggressive behavior and snout proximity, and snout contact. The correlation was performed using the “cor.test” function in R. Not all animals had the same amount of observations, and therefore the behaviors were divided by the total number of 15-min sessions per animal to come to an average frequency per 15 min.

We performed three linear mixed models in R using the package lmerTest (version 3.1-3) (Kuznetsova et al. 2017). In model 1, the response variable was the frequency of snout proximity or frequency of snout contact per session (two separate analyses). The predictor variables were social stability phase, housing condition, socialization, sex, and MDS. The random variables were batch and pig ID nested within group. In model 2, ADWG was the response variable (for the initiator and recipient separately). The predictor variables were snout proximity, snout contact, aggression, housing condition, socialization, and sex, with as random variables batch and pig ID nested within group. In model 3, the frequency of snout proximity and snout contact towards related (littermates) and unrelated conspecifics (nonlittermates) was analyzed in a mixed model. The response variable was the frequency of snout proximity or snout contact (two separate analyses). The predictor variables were housing condition, sex, socialization, and social stability, with as random variables pig ID and group nested within batch.

We selected the best fit model based on lowest AIC value after backward selection on the full model using the function step(). Due to the backward selection step the final model had only the variables that best fitted the model. Hence in model 1, the phase of social stability, housing condition and sex best predicted snout proximity behavior whereas phase of social stability, housing condition and MDS best predicted the snout contact behavior. In model 2, snout proximity and snout contact best predicted the ADWG for initiator whereas, snout proximity, snout contact, and aggressive behavior was in the final model for recipient. In model 3, the phase of social stability, housing condition, and individual sex best fitted the model for snout proximity towards unrelated individuals. In contrast, only the phase of stability and housing condition was in the final best fit model for the snout proximity towards related individual. Similarly, the phase of social stability and housing only best fitted the model for snout contact towards unrelated as well as related conspecifics. Model diagnostics were performed on the best fit model to check for normality and homogeneity of residuals and collinearity between predictor variables. Residuals were normally distributed, as assessed from Shapiro–Wilk statistics. Homogeneity was validated by visual inspection of residual plots against fitted values from sjPlot package (version 2.8.15) (Lüdecke 2024). Cook distances were calculated and plotted to identify influential data points (the Cook distance for any data point was smaller than 1 in all models). The variance inflation factor was performed to detect multicollinearity (i.e., correlation between predictor variables) from the “car” package (version 3.1-2) (Fox and Weisberg 2018). The variance inflation factor did not indicate strong multicollinearity (<2). Post hoc analyses were performed using the “emmeans” (version 1.8.9) package (Russell 2020) to report least square means (LSmeans) with standard errors (SE). Reported values are LSmeans with SE unless stated otherwise. P values were obtained for the model and considered significant at ≤0.05. Figures were made using the “ggplot2” package (version 3.5.1) (Wickham 2016).

Results

Hypothesis 1: Individuals choose nontactile over tactile behavior when risks are high

We predicted that pigs would show snout proximity when costs (in terms of aggression) are high and show more snout contact when risks are low. Pigs showed, on average, 0.46 ± 0.33 (means ± SD) times snout proximity per pig per 15 min, and 0.82 ± 0.48 times snout contact per pig per 15 min. The frequency of snout proximity differed depending on social stability (low–mid: 030 ± 0.026; low–high: 030 ± 0.026; P < 0.001, model 1), with the highest frequency during low social stability, while there was no significant difference between the medium and high stability phase (Figure 2). The frequency of snout contact was lower during the medium stability phase (low–mid: 0.10 ± 0.039, P = 0.02; MID–HIGH: −0.13 ± 0.040, P = 0.002, model 1) while being numerically but not significantly higher in the phase of high social stability (Figure 2). There was a positive correlation between snout proximity and snout contact in all phases of social stability (low: r = 0.75, P < 0.001; medium: r = 0.52, P < 0.001; high: r = 0.61, P < 0.001).

Figure 2.

Comparative bar graph showing difference between snout proximity and snout contact.

Open in a new tab

Average frequency (per pig per 15 min of observation) of snout proximity (left) and snout contact (right) by phase of social stability. The asterisk indicates significant differences (P < 0.05) between the phases.

Aggressive behavior was positively correlated with snout proximity (rrange: 0.34–0.65; P < 0.001) and snout contact (rrange: 0.4–0.66; P < 0.001) across the phases of social stability. The Modified David score (MDS) did not relate to the expression of snout proximity (P = 0.42; model 1) but did relate to the expression of snout contact (b = 0.006 ± 0.002; P = 0.03; model 1) indicating that pigs with higher dominance value initiated more snout contact. There was no correlation between dominance (MDS) and weight gain (ADWG; r = 0.01; P = 0.89). Males expressed more snout proximity than females (males: 0.54 ± 0.126; females: 0.46 ± 0.126 times/15 min; P = 0.002) but did not differ from females in the frequency of snout contact (P = 0.48). Pigs in the more enriched housing conditions showed less frequent snout proximity (enriched: 0.46 ± 0.114 times/15 min vs. less-enriched: 0.52 ± 0.114 times/15 min; P = 0.02, model 1) and snout contact (enriched: 0.81 ± 0.184 times/15 min vs. less-enriched: 0.94 ± 0.184 times/15 min; P = 0.01, model 1) than the less-enriched pigs. Preweaning socialization did not influence the snout proximity (P = 0.80) or contact (P = 0.65).

Hypothesis 2: Individuals derive benefits from tactile contact

Pigs that initiated or received a greater frequency of snout proximity had a lower average daily weight gain (ADWG) than those who did so less (initiator: b = −0.65 ± 0.056 kg/day; P ≤ 0.001; recipient: b = −0.62 ± 0.060 kg/day; P ≤ 0.001; model 2) (Figure 3). In line with our hypothesis, frequent initiators and recipients of snout contact had a greater ADWG than those who initiated or received less snout contact (initiator: b = 0.20 ± 0.041 kg/day; P ≤ 0.001; recipient: b = 0.18 ± 0.046 kg/day; P ≤ 0.001; model 2) (Figure 3). To confirm the costs associated with aggression, pigs that received more aggressive behavior had a lower ADWG (b = −0.11 ± 0.030 kg/day; P ≤ 0.001; model 2) whereas for the initiator aggression was not significantly associated with ADWG (P = 0.43).

Figure 3.

Scatterplot with the relationship between average daily weight gain and social behaviors.

Open in a new tab

Relationship between the average daily weight gain (ADWG, in kg/day) and behavioral expression of snout proximity, snout contact, and aggressive behavior for initiators (teal) and recipients (orange) of the behavior. The x axis shows the average frequency of behavior per 15 min. Predicted values refer to the fitted response value from the model.

Factors affecting snout proximity and contact

Unrelated conspecifics showed more snout proximity towards each other compared to related conspecifics (nonlittermates: 0.31 ± 0.013, vs. littermates: 0.24 ± 0.077, P < 0.0001) and more snout contact (nonlittermates: 0.31 ± 0.077 vs. littermates: 0.24 ± 0.077, P < 0.0001). In the phase of low social stability, related pigs (littermates) showed more frequent snout proximity towards each other than in the later phases (P < 0.001, model 3; Figure 4).

Figure 4.

Boxplot with snout proximity towards related and unrelated pigs across social stability phases.

Open in a new tab

Frequency of snout proximity (per 15 min) directed towards unrelated pigs (nonlittermates; blue) and related pigs (littermates; orange) during the phases of low, medium, and high social stability.

Discussion

As hypothesized, nontactile snout proximity was highest during low social stability and reduced with increasing social stability. However, unexpectedly, tactile behavior in the form of snout contact did not show a linear increase over time as we predicted. Instead, snout contact remained reasonably constant, with only a slight decrease in the intermediate phase. This suggests that snout contact is an important behavior regardless of social stability. This is supported by the positive relationship between body weight gain (ADWG) and the initiation and receipt of snout contact, which aligns with our second hypothesis that snout contact translates into individual costs and benefits in growth.

Snout proximity reduced after the low stability phase, with no significant difference between the phases of medium and high social stability, while aggression was significantly lower only in the high stability phase. This suggests that snout proximity mostly serves a function in social recognition (Penn and Frommen 2010). This is further confirmed by the more frequent nosing behaviors between the unrelated pigs. However, even between littermates nosing behavior was higher in the low social stability phase, although an increase in interactions is common during a change in the physical or social environment (e.g., increased aggression is seen in stable groups when they are relocated; D’Eath et al. 2010).

We further hypothesized that snout contact would relate to benefits. We indeed found a positive correlation between snout contact and body weight gain, although here as well, the directionality of the effect is uncertain. Individuals with better growth may have more energy to invest in social relationships (Barber 1991), as is reported for play behavior (Brown et al. 2015; Franchi et al. 2023), and they may have physiological advantages when taking risks (Mathot et al. 2015). However, the relationship between growth performance and the receipt of snout contact suggests that close interactions are associated with faster growth. Previous studies have shown that recipients of nosing behavior (contact with any body part, without distinction between snout proximity and contact) had a better growth performance (Camerlink et al. 2012), which is suggested to occur through the release of oxytocin (Uvnäs-Moberg and Petersson 2005; Camerlink et al. 2012). Animals can express their affiliative behaviors using body contact, allogrooming (Mooring et al. 1998; Tønnesen et al. 2008; Schneider and Krueger 2012), and licking (Johnsen et al. 2016). Snout-to-body contact, which may include nudging and snout-to-snout contact, is also used as a play marker (Horback 2023). Snout contact specifically can consolidate the social bond between the sow and her piglets (Newberry et al. 1988; Blackshaw and Hagelsø 1990; Petersen et al. 1990) and is therefore a biologically relevant behavior for pigs. The results show that the differential use of nosing behavior such as snout proximity, snout-to-snout contact and snout-to-body contact can have different implications for individuals. Similarly, Yasui and Idani (2017) have argued that for elephants the shape of the trunk when touching other conspecifics implies different functions. Based on the current results, we suggest that snout proximity is used more for social recognition whereas snout contact may relate more to social affiliation, and that these two behaviors may have an inverse relationship with body weight gain.

In social species which maintain a dominance hierarchy, dyadic encounters determine their rank order, which can be through aggressive behavior and may escalate into a fight (e.g., Clark and D’Eath 2013; Camerlink et al. 2022). Individuals may therefore avoid direct contact upon encounter with unfamiliar conspecifics. While animals are primed to make optimal decisions (Bogacz 2007), the positive correlation between snout proximity and contact could suggest that the expression of nosing behavior does not relate to optimization of social information but rather conflict avoidance. Pigs that showed more snout proximity may have avoided conflict, and thus costs, but did have a lower growth performance. Rather than snout proximity resulting in costs, such as energy expenditure and receipt of aggression (Vøllestad and Quinn 2003), it is more plausible that less well growing individuals, and thus weaker ones, will have had more reason to be risk aversive by showing more snout proximity. The benefits of nontactile over tactile behavior may depend on individuals’ assessment of costs and benefits (Georgiev et al. 2013; Maestripieri 2021), and our results suggest that pigs adjust their behavior based on social context, which may especially be the case for the individuals who have a lower resource holding potential to engage in fights (Camerlink et al. 2022). Those individuals may trade-off the opportunity of tactile contact by choosing a lower risk of aggression by engaging in snout proximity instead.

The relationships between nosing behavior and growth could be due to a third confounding factor, such as dominance. Dominant pigs are typically (but not always) heavier and grow faster (e.g., Scheel et al. 1977) due to their preferential access to feed resources, which was not found in the current study. On the other hand, dominance is not the sole reason associated with weight gain (Meese and Ewbank 1973), and other factors such as high cortisol responses to stress (Hewagalamulage et al. 2016), sex differences, fearfulness (Page et al. 2019), or social stress (Tamashiro et al. 2007) can have an important effect. Dominant pigs showed more snout contact, but not more snout proximity. This may reflect dominant pigs’ social ease in their higher ranked position, as they have less to fear from an aggressive response because the social hierarchy is maintained through active avoidance of dominant individuals by subordinates (Gonyou 2001). While the Modified David's Score (which was here based on overall aggressive encounters) is one of many standard approaches for observing dominance hierarchies, calculating the rank based on winners and losers of aggressive interactions might provide a more precise picture regarding the relationship between dominance, snout contact, and weight gain.

Behaviors expressed in dyadic interactions are affected by conditions such as familiarity (Patison et al. 2010; Kutsukake et al. 2019) and the physical and social environment (Rodenburg and Koene 2007), whereby dyadic interactions are formed based on individuals’ ability to utilize personal and social information (Dall et al. 2005). Pigs use olfactory cues to distinguish familiar and unfamiliar conspecifics (Kristensen et al. 2001) and need to come in close proximity for olfactory recognition (Schild and Rørvang 2023). It is therefore reasonable that pigs that were unfamiliar to each other had more frequent snout proximity. However, the finding of more snout contact between unrelated pigs contradicts the theory that snout contact would be more associated with affiliative behavior (Camerlink and Turner 2013).

Males showed more snout proximity, but not contact, than females. This relates to previous work that shows that males show more ritualized agonistic behavior and take longer before engaging in contact, supposedly because the costs of escalated aggression between males are greater than between females (Camerlink et al. 2022). Pigs housed in enriched conditions exhibited less frequent snout proximity and contact than pigs in less-enriched conditions even when the space was adequately available. This is in line with the studies on behavioral differences between barren and enriched housing conditions, in which animals in enriched conditions perform fewer social interactions due to more frequent engagement with objects in the environment (e.g., Luo et al. 2020). Further, early life socialization did not influence snout contact or snout proximity. Early life socialization has been suggested to reduce pigs’ aggressive behavior later in life (D’Eath 2005; Verdon et al. 2016) and to influence sociocognitive development (Weller et al. 2020), although it did not influence social play behavior (Weller et al. 2019).

Time, and thus the animals’ age, presents a natural confound with social stability. However, social behavior remains largely unaffected by age in juvenile (O’Malley et al. 2022) and sub-adult pigs (Seddaiu et al. 2024) but it can fluctuate depending on the social context (O’Malley et al. 2022). For example, regrouping has a profound effect on the level of aggression (e.g., D’Eath et al. 2010; Turner et al. 2017). Indeed, in the current study, littermates showed a stable frequency of snout proximity and contact, aside from the low stability phase when they were regrouped with unfamiliar pigs. We can therefore be reasonably confident that the differences in behavior are related to the level of social stability rather than the change in age.

In conclusion, the positive association between snout contact and body weight gain, and the negative association between snout proximity and body weight gain suggests some behavioral trade-offs. Although the direction of the relationships would benefit from further study, the data suggest that in a situation of low social stability, slower growing individuals, who likely would suffer greater costs from escalated contact, may adapt by showing nontactile behavior to avoid direct physical contact, while snout contact has a relationship with enhanced growth performance. Hence, during group formation, snout proximity seems more related to conflict avoidance and conspecific recognition, whereas snout contact may have a more affiliative role in social interactions. These results add to the body of literature that shows that regrouping pigs into new social groups is detrimental for their welfare. It further emphasizes the relevance of observing nuances in animals’ social behavior to gain a better understanding of how they adapt to adverse circumstances.

Acknowledgments

We would like to thank the animal technicians at SRUC for their help with the experiment.

Contributor Information

Sunil Khatiwada, Animal Behaviour and Welfare Department, Institute of Genetics and Animal Biotechnology of the Polish Academy of Sciences, Postępu 36a Jastrzębiec 05-552, Poland.

Victoria E Lee, Animal Behaviour & Welfare, Animal and Veterinary Sciences Department, Scotland's Rural College (SRUC), West Mains Rd., Edinburgh EH9 3JG, UK.

Simon P Turner, Animal Behaviour & Welfare, Animal and Veterinary Sciences Department, Scotland's Rural College (SRUC), West Mains Rd., Edinburgh EH9 3JG, UK.

Irene Camerlink, Animal Behaviour and Welfare Department, Institute of Genetics and Animal Biotechnology of the Polish Academy of Sciences, Postępu 36a Jastrzębiec 05-552, Poland.

Funding

The animal experiment was funded by the BBSRC (Biotechnology and Biological Sciences Research Council), grant number BB/V001515/1, and was conducted in collaboration with Queen's University Belfast and the Pig Improvement Company (PIC). SRUC receives support from the Scottish Government. The behavioral data collection and analysis for this study is part of a project funded by Narodowym Centrum Nauki Poland (NCN), grant number 2020/39/B/NZ8/02508.

Ethics statement

This noninvasive behavioral study adhered to the guidelines approved by SRUC's Animal Experiments Committee. The experiment was authorized by the UK Home Office (Project Licence: PP1403242). The study was part of a larger overarching research project which imposed different housing conditions as treatments. These are accounted for in the statistical analyses but are not the focus of this study.

References

  1. Barber  N, 1991. Play and energy regulation in mammals. Q Rev Biol  66(2):129–147. [DOI] [PubMed] [Google Scholar]
  2. Bateson  M, Martin  P, 2021. Measuring Behaviour: An Introductory Guide. Cambridge, UK: Cambridge University Press. [Google Scholar]
  3. Blackshaw  JK, Hagelsø  AM, 1990. Getting-up and lying-down behaviours of loose-housed sows and social contacts between sows and piglets during day 1 and day 8 after parturition. Appl Anim Behav Sci  25(1–2):61–70. [Google Scholar]
  4. Bogacz  R, 2007. Optimal decision-making theories: linking neurobiology with behaviour. Trends Cognit Sci  11(3):118–125. [DOI] [PubMed] [Google Scholar]
  5. Brown  SM, Klaffenböck  M, Nevison  IM, Lawrence  AB, 2015. Evidence for litter differences in play behaviour in pre-weaned pigs. Appl Anim Behav Sci  172:17–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Camerlink  I, Bijma  P, Kemp  B, Bolhuis  JE, 2012. Relationship between growth rate and oral manipulation, social nosing, and aggression in finishing pigs. Appl Anim Behav Sci  142(1–2):11–17. [Google Scholar]
  7. Camerlink  I, Farish  M, Arnott  G, Turner  SP, 2022. Sexual dimorphism in ritualized agonistic behaviour, fighting ability and contest costs of Sus scrofa. Front Zool  19(1):13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Camerlink  I, Turner  SP, 2013. The pig's nose and its role in dominance relationships and harmful behaviour. Appl Anim Behav Sci  145(3–4):84–91. [Google Scholar]
  9. Carsten  T, Fievez  F, Duque  J, 2023. Movement characteristics impact decision-making and vice versa. Sci Rep  13(1):3281. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Champagne  DL, Bagot  RC, van Hasselt  F, Ramakers  G, Meaney  MJ  et al. , 2008. Maternal care and hippocampal plasticity: evidence for experience-dependent structural plasticity, altered synaptic functioning, and differential responsiveness to glucocorticoids and stress. J Neurosci  28(23):6037–6045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Clark  CC, D’Eath  RB, 2013. Age over experience: Consistency of aggression and mounting behaviour in male and female pigs. Appl Anim Behav Sci  147(1–2):81–93. [Google Scholar]
  12. Coutellier  L, Arnould  C, Boissy  A, Orgeur  P, Prunier  A  et al. , 2007. Pig’s responses to repeated social regrouping and relocation during the growing-finishing period. Appl Anim Behav Sci  105(1–3):102–114. [Google Scholar]
  13. Dall  SRX, Giraldeau  LA, Olsson  O, McNamara  JM, Stephens  DW, 2005. Information and its use by animals in evolutionary ecology. Trends Ecol Evol  20(4):187–193. [DOI] [PubMed] [Google Scholar]
  14. D’Eath  RB, 2005. Socialising piglets before weaning improves social hierarchy formation when pigs are mixed post-weaning. Appl Anim Behav Sci  93(3–4):199–211. [Google Scholar]
  15. D’Eath  RB, Turner  SP, Kurt  E, Evans  G, Thölking  L  et al. , 2010. Pigs’ aggressive temperament affects pre-slaughter mixing aggression, stress and meat quality. Animal  4(4):604–616. [DOI] [PubMed] [Google Scholar]
  16. Demartsev  V, Gersick  AS, Jensen  FH, Thomas  M, Roch  MA  et al. , 2023. Signalling in groups: New tools for the integration of animal communication and collective movement. Methods Ecol Evol  14(8):1852–1863. [Google Scholar]
  17. Fox  J, Weisberg  S, 2018. An R Companion to Applied Regression. 3rd edn. Thousand Oaks (CA): Sage Publications. [Google Scholar]
  18. Franchi  GA, Larsen  ML, Kristoffersen  IH, Winters  JF, Pedersen  LJ  et al. , 2023. Play behaviour positively relates to weight gain, feeding behaviour and drinking behaviour in weaner pigs (Sus scrofa). Appl Anim Behav Sci  259:105836. [Google Scholar]
  19. Georgiev  AV, Klimczuk  AC, Traficonte  DM, Maestripieri  D, 2013. When violence pays: a cost-benefit analysis of aggressive behavior in animals and humans. Evol Psychol  11(3):678–699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gliga  T, Farroni  T, Cascio  CJ, 2019. Social touch: A new vista for developmental cognitive neuroscience?  Dev Cogn Neurosci  35:1–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Gonyou  HW, 2001. The social behaviour of pigs. In: Social Behaviour in Farm Animals. Wallingford, UK: CABI Publishing, 147–176. [Google Scholar]
  22. Hewagalamulage  SD, Lee  TK, Clarke  IJ, Henry  BA, 2016. Stress, cortisol, and obesity: A role for cortisol responsiveness in identifying individuals prone to obesity. Domest Anim Endocrinol  56:S112–S120. [DOI] [PubMed] [Google Scholar]
  23. Hill  HM, Dietrich  S, Guarino  S, Banda  M, Lacy  K, 2019. Preliminary observations of an unusual mouth interaction between beluga calves (Delphinapterus leucas). Zoo Biol  38(2):149–156. [DOI] [PubMed] [Google Scholar]
  24. Horback  K, 2023. Nosing around: Play in pigs. Anim Behav Cogn  2(2):186. [Google Scholar]
  25. Johnsen  JF, Zipp  KA, Kälber  T, de Passillé  AM, Knierim  U  et al. , 2016. Is rearing calves with the dam a feasible option for dairy farms? Current and future research. Appl Anim Behav Sci  181:1–11. [Google Scholar]
  26. Kristensen  HH, Jones  RB, Schofield  CP, White  RP, Wathes  CM, 2001. The use of olfactory and other cues for social recognition by juvenile pigs. Appl Anim Behav Sci  72(4):321–333. [DOI] [PubMed] [Google Scholar]
  27. Kutsukake  N, Teramoto  M, Honma  S, Mori  Y, Ikeda  K  et al. , 2019. Behavioural and hormonal changes during group formation by malechimpanzees. Behaviour  156(2):109–129. doi: 10.1163/1568539X-00003532. [DOI] [Google Scholar]
  28. Kuznetsova  A, Brockhoff  PB, Christensen  RH, 2017. lmerTest package: Tests in linear mixed effects models. J Stat Softw  82(13):1–26. [Google Scholar]
  29. Leiva  D, de Vries  H, 2022. Steepness: testing steepness of dominance hierarchies. R package v0.3–0. [accessed 2024 July]. Available from: https://CRAN.R-project.org/package=steepness.
  30. Lüdecke  D, 2024. sjPlot: Data visualization for statistics in social science. R package version 2.8.16. [accessed 2024 July]. Available from: https://CRAN.R-project.org/package=sjPlot.
  31. Luo  L, Reimert  I, Middelkoop  A, Kemp  B, Bolhuis  JE, 2020. Effects of early and current environmental enrichment on behavior and growth in pigs. Front Vet Sci  7:268. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Maestripieri  D, 2021. Costs of aggression. In: Shackelford  TK, Weekes-Shackelford  VA, editors. Encyclopedia of Evolutionary Psychological Science. Cham: Springer International Publishing, 1540–1544. [Google Scholar]
  33. Mathot  KJ, Nicolaus  M, Araya-Ajoy  YG, Dingemanse  NJ, Kempenaers  B, 2015. Does metabolic rate predict risk-taking behaviour? A field experiment in a wild passerine bird. Funct Ecol  29(2):239–249. [Google Scholar]
  34. McGlone  JJ, 1985. A quantitative ethogram of aggressive and submissive behaviours in recently regrouped pigs. J Anim Sci  61(3):559–565. [PubMed] [Google Scholar]
  35. Meese  GB, Ewbank  R, 1973. The establishment and nature of the dominance hierarchy in the domesticated pig. Anim Behav  21(2):326–334. [Google Scholar]
  36. Mooring  MS, Gavazzi  AJ, Hart  BL, 1998. Effects of castration on grooming in goats. Physiol Behav  64(5):707–713. [DOI] [PubMed] [Google Scholar]
  37. Newberry  RC, Wood-Gush  DGM, Hall  JW, 1988. Playful behaviour of piglets. Behav Processes  17(3):205–216. [DOI] [PubMed] [Google Scholar]
  38. Nord  CM, Bonnell  TR, Dostie  MJ, Henzi  SP, Barrett  L, 2021. Tolerance of muzzle contact underpins the acquisition of foraging information in vervet monkeys (Chlorocebus pygerythrus). J Comp Psychol  135(3):349–359. [DOI] [PubMed] [Google Scholar]
  39. O’Malley  CI, Steibel  JP, Bates  RO, Ernst  CW, Siegford  JM, 2022. The social life of pigs: Changes in affiliative and agonistic behaviors following mixing. Animals (Basel)  12(2):206. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Page  KD, Ruykys  L, Miller  DW, Adams  PJ, Bateman  PW  et al. , 2019. Influences of behaviour and physiology on body mass gain in the woylie (Bettongia penicillata ogilbyi) post-translocation. Wildl Res  46(5):429–443. [Google Scholar]
  41. Partan  SR, Marler  P, 2005. Issues in the classification of multimodal communication signals. Am Nat  166(2):231–245. [DOI] [PubMed] [Google Scholar]
  42. Patison  KP, Swain  DL, Bishop-Hurley  GJ, Robins  G, Pattison  P  et al. , 2010. Changes in temporal and spatial associations between pairs of cattle during the process of familiarisation. Appl Anim Behav Sci  128(1–4):10–17. [Google Scholar]
  43. Penn  DJ, Frommen  JG, 2010. Kin recognition: an overview of conceptual issues, mechanisms and evolutionary theory. In: Kappeler  P, editor. Animal Behaviour: Evolution and Mechanisms. Berlin, Heidelberg: Springer, 55–85. [Google Scholar]
  44. Petersen  V, Recén  B, Vestergaard  K, 1990. Behaviour of sows and pig during farrowing under free-range conditions. Appl Anim Behav Sci  26(1–2):169–179. [Google Scholar]
  45. Prevolnik Povše  M, Mesarec  N, Skok  J, Škorjanc  D, 2021. Agonistic interactions between littermates reappear after mixing multiple litters at weaning in pigs. Agriculture  11(9):844. [Google Scholar]
  46. R Core Team , 2023. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. [Google Scholar]
  47. Rodenburg  TB, Koene  P, 2007. The impact of group size on damaging behaviours, aggression, fear and stress in farm animals. Appl Anim Behav Sci  103(3–4):205–214. [Google Scholar]
  48. Russell  L, 2020. emmeans: Estimated marginal means, aka least-squares means. R package version 1.4.7. [accessed 2024 July]. Available from: https://cran.r-project.org/package=emmeans.
  49. Scheel  DE, Graves  HB, Sherritt  GW, 1977. Nursing order, social dominance and growth in swine. J Anim Sci  45(2):219–229. [Google Scholar]
  50. Schild  SLA, Rørvang  MV, 2023. Pig olfaction: The potential impact and use of odors in commercial pig husbandry. Front Anim Sci  4:1215206. [Google Scholar]
  51. Schino  G, 1998. Reconciliation in domestic goats. Behaviour  135(3):343–356. [Google Scholar]
  52. Schino  G, Marini  C, 2011. Know your enemy: Accessibility and danger modulate the use of conciliatory patterns in mandrills. Anim Behav  81(5):1009–1014. [Google Scholar]
  53. Schneider  G, Krueger  K, 2012. Third-party interventions keep social partners from exchanging affiliative interactions with others. Anim Behav  83(2):377–387. [Google Scholar]
  54. Seddaiu  P, Turner  SP, Camerlink  I, 2024. Cross-seasonal and diurnal variation in physical contact between sub-adult pigs. Appl Anim Behav Sci  278:106379. [Google Scholar]
  55. Tamashiro  KL, Hegeman  MA, Nguyen  MM, Melhorn  SJ, Ma  LY  et al. , 2007. Dynamic body weight and body composition changes in response to subordination stress. Physiol Behav  91(4):440–448. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Tibbetts  EA, Dale  J, 2007. Individual recognition: It is good to be different. Trends Ecol Evol  22(10):529–537. [DOI] [PubMed] [Google Scholar]
  57. Tønnesen  HG, Bøe  KE, Andersen  IL, 2008. Group size and social interactions in goats. In: Proceedings of the 18th Nordic Symposium of the ISAE  Oscarsborg, Norway: Finnish Society for Applied Ethology, 16–18. [Google Scholar]
  58. Turner  SP, Nevison  IM, Desire  S, Camerlink  I, Roehe  R  et al. , 2017. Aggressive behaviour at regrouping is a poor predictor of chronic aggression in stable social groups. Appl Anim Behav Sci  191:98–106. [Google Scholar]
  59. Uvnäs-Moberg  K, 1998. Oxytocin may mediate the benefits of positive social interaction and emotions. Psychoneuroendocrinology  23(8):819–835. [DOI] [PubMed] [Google Scholar]
  60. Uvnäs-Moberg  K, Petersson  M, 2005. Oxytocin, a mediator of anti-stress, well-being, social interaction, growth and healing. Z Psychosom Med Psychother  51(1):57–80. [DOI] [PubMed] [Google Scholar]
  61. Vázquez  J, Fargallo  JA, Jiménez  N, Aguilar-Montiel  F, Rodríguez-Martínez  L, 2020. Dear enemy effect in the Mexican volcano mouse Neotomodon alstoni: Implications of sex in the agonistic behaviour among neighbours. Behav Processes  181:104251. [DOI] [PubMed] [Google Scholar]
  62. Verdon  M, Morrison  RS, Hemsworth  PH, 2016. Rearing piglets in multi-litter group lactation systems: Effects on piglet aggression and injuries post-weaning. Appl Anim Behav Sci  183:35–41. [Google Scholar]
  63. Vøllestad  LA, Quinn  TP, 2003. Trade-off between growth rate and aggression in juvenile coho Salmon, Oncorhynchus kisutch. Anim Behav  66(3):561–568. [Google Scholar]
  64. Weller  JE, Camerlink  I, Turner  SP, Farish  M, Arnott  G, 2019. Socialisation and its effect on play behaviour and aggression in the domestic pig (Sus scrofa). Sci Rep  9(1):4180. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Weller  JE, Turner  SP, Futro  A, Donbavand  J, Brims  M  et al. , 2020. The influence of early life socialisation on cognition in the domestic pig (Sus scrofa domestica). Sci Rep  10(1):19077. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Wickham  H, 2016. ggplot2: Elegant Graphics for Data Analysis. New York: Springer. [Google Scholar]
  67. Yasui  S, Idani  G, 2017. Social significance of trunk use in captive Asian elephants. Ethol Ecol Evol  29(4):330–350. [Google Scholar]

Articles from Current Zoology are provided here courtesy of Oxford University Press

RESOURCES