TYPE Original Research PUBLISHED 28 July 2022 DOI 10.3389/fanim.2022.909038 Investigating the effects of two OPEN ACCESS weaning methods and two EDITED BY Ruth Catriona Newberry, Norwegian University of Life genetic hybrids on play behavior Sciences, Norway REVIEWED BY in weaner pigs (Sus scrofa) Marek Spinka, Czech University of Life Sciences Prague, Czechia Guilherme A. Franchi 1*, Mona L. V. Larsen 1,2, Jennifer Weller, Queen’s University Belfast, Jeanet F. M. Winters 1, Margit Bak Jensen 1 United Kingdom and Lene Juul Pedersen 1 *CORRESPONDENCE 1 Department of Animal Science, Aarhus University, Tjele, Denmark, 2 Katholieke Universiteit (KU) Guilherme A. Franchi Leuven, Department of Biosystems, Division Animal and Human Health engineering, M3-BIORES: amorimfranchi@anis.au.dk Measure, Model and Manage Bioresponses, Heverlee, Belgium SPECIALTY SECTION This article was submitted to Animal Welfare and Policy, a section of the journal Frontiers in Animal Science In conventional pig production, suckling pigs are typically weaned between 3 and RECEIVED 31 March 2022 5 weeks of age. This involves separation from the sow, dietary change, and ACCEPTED 04 July 2022 regrouping in a novel environment, which challenge the welfare of pigs. We PUBLISHED 28 July 2022 investigated the effects of two weaning methods [conventional weaning: two CITATION litters mixed in a weaner pen of different size and design (MOVE) vs. litter staying Franchi GA, Larsen MLV, Winters JFM, in the farrowing pen after removing the sow (STAY)] and two genetic hybrids Jensen MB and Pedersen LJ (2022) Investigating the effects of two [DanBred Yorkshire × Landrace (approximately 21 total pigs born/litter; DB) vs. weaning methods and two genetic Topigs Norsvin TN70 Yorkshire × Landrace (approximately 16 total pigs born/ hybrids on play behavior in weaner litter with higher individual birth weight and weaning weight than DB; TN)] on pigs (Sus scrofa). Front. Anim. Sci. 3:909038. play behavior across weaning. Both genetic hybrids were inseminated with doi: 10.3389/fanim.2022.909038 semen of DanBred Duroc boars. Litters were reduced to the number of COPYRIGHT functional teats at birth. The durations of locomotor-rotational play (LOC) and © 2022 Franchi, Larsen, Winters, Jensen social play (SOC) of 24 indoor-housed litters [pigs/litter: (average ± SD) 13 ± 2; and Pedersen. This is an open-access article distributed under the terms of age at the weaning day: 26 ± 2 days] were video-recorded continuously between the Creative Commons Attribution 14:00 h and 22:00 h on days −1, 1, and 2 relative to weaning and statistically License (CC BY). The use, distribution or reproduction in other forums is analyzed with mixed-effects modeling at the individual level. Before weaning, TN permitted, provided the original author pigs performed LOC longer than DB pigs. On day 2 post-weaning, STAY pigs (s) and the copyright owner(s) are engaged in more SOC than MOVE pigs. Moreover, TN pigs and STAY pigs credited and that the original publication in this journal is cited, in displayed a steeper increase in LOC from days 1 to 2 than DB pigs and MOVE accordance with accepted academic pigs, respectively. We demonstrated that pigs belonging to the genetic hybrid practice. No use, distribution or reproduction is permitted which does with higher weight at birth and weaning spent more time playing on the day not comply with these terms. before weaning. Additionally, weaning pigs in the farrowing pen and, hence, avoiding social mixing and relocation to an unfamiliar environment had a positive effect on social play after weaning. Our study illustrates that weaning stress in pigs may be reduced by using a genetic hybrid featuring higher birth and weaning weight and by keeping litters intact in a familiar environment after weaning. This study also supports the use of play behavior as an animal welfare indicator. KEYWORDS locomotor-rotational play, social play, farrowing pen, genetic hybrid, video analysis, positive animal welfare Frontiers in Animal Science 01 frontiersin.org Franchi et al. 10.3389/fanim.2022.909038 Introduction effects of early weaning in pigs, for example, the use of housing systems that allow the dam and her offspring to voluntarily In a natural setting, a sow seeks a secluded site to farrow and spend time apart, from day 11 pp at the earliest, to habituate the remains close to her pigs, who are nursed approximately 30 young to periods of separation and to encourage early times daily for the first week after farrowing (Weary et al., 2008). consumption of solid food aiming to improve post-weaning Approximately 10 days post-partum (pp), nursing declines and performance (e.g., Damm et al., 2003; Hötzel et al., 2004). pigs gradually become more nutritionally and socially Alternatively, offering a diet with increased energy density independent, and by 4 to 5 months of age, pigs are fully prior to weaning reportedly led to increased food intake habituated to solid food (Jensen and Stangel, 1992). lasting for days after weaning (Pajor et al., 2002) and reduced Conversely, in modern pig production systems, pigs are post-weaning frequency of vocalizations (Weary et al., 1999). abruptly weaned from their dam at approximately 3 to 5 Furthermore, weaning pigs in the farrowing pen may avoid weeks of age (Š pinka, 2017). In addition to the abrupt stressful handling and transportation as well as keep litters separation from the sow, pigs typically face changes in the intact, thus limiting disruption of social organization and physical and social environments, as well as a shift from fighting following weaning (Puppe et al., 1997; Colson highly digestible sow milk to a less digestible dry solid feed et al., 2012). (Dybkjær, 1992; Weary et al., 2008; Campbell et al., 2013). Selecting a genetic hybrid with an expected increased Hence, weaning is acknowledged to challenge the welfare of viability may contribute to increase the resilience of pigs to pigs as they are too physiologically and behaviorally immature to weaning. In Denmark, DanBred Landrace × Yorkshire hybrid cope with these synchronous stressors. For instance, the abrupt sows are commonly used in commercial farms (Früh et al., 2014) dietary change can result in undernourishment, reduced growth, and they are characterized by their high prolificity [in 2020, the and increased susceptibility to illness such as diarrhea (Le average size litter 5 days pp was 19.6 pigs (Hansen, 2021)]. The Dividich and Sève, 2000; Leeb et al., 2014). Additionally, the high prolificity leads to economic and environmental benefits early separation from the mother and subsequently the social due to an increased number of weaned pigs per sow and mixing in a new environment can increase the frequency of consequent reduced carbon emissions per kilogram of aggression and occurrence of injuries (Pitts et al., 2000) and lead slaughtered pig, respectively (Rutherford et al., 2013). to abnormal behaviors such as belly nosing and re-directed However, the increased litter size is obtained at the expense of suckling of ears or tails of other pigs (Fraser, 1975; Puppe more live-born pigs with low birth weight (Pedersen et al., 2011) et al., 1997). These practices also reduce play behavior. For and low average rectal temperature within 24 h pp (Schild et al., instance, Donaldson et al. (2002) and Hötzel et al. (2010) found a 2020), leading to increased pig mortality and reduced viability. reduced occurrence of play behavior in 3- to 4-week-old pigs on Another disadvantage is that sows’ functional teats are being the first days following weaning. Additionally, Colson et al. outnumbered by litter size (Moustsen and Nielsen, 2017), which (2012) reported reduced time spent playing in 26-day-old pigs leads to increased teat competition (Milligan et al., 2001). As socially mixed in a novel environment at weaning. total colostrum yield is reportedly independent of litter size Indeed, one way of investigating the welfare consequences of (Foisnet et al., 2010), the consequences of this high prolificity are weaning in pigs can be through play behavior. Pigs perform play low pig survival chances and viability (Ocepek et al., 2017; Schild behavior from the first day of life (Š pinka, 2017), and this et al., 2020). An alternative breed, the Topigs Norsvin TN70 behavior is considered a relevant welfare indicator because it is hybrid [Norsvin Landrace × Z-line (Yorkshire)], is characterized typically performed when the animal’s primary needs, such as by a slightly reduced litter size (approximately 16 pigs 5 days pp; food and safety, have been met (Newberry et al., 1988; Š pinka Schild et al., 2020), but live-born pigs with higher birth weight, et al., 2001; Held and Š pinka, 2011), while suppression of play which is associated with higher body weight on the expected day behavior may indicate the presence of physical, physiological, or of weaning, higher average rectal temperature within 24 h pp psychological challenges. In addition to indicating the fulfillment (Quiniou et al., 2002; Kammersgaard et al., 2011), and higher of primary needs, and the fact that negative affective states are growth from birth to weaning in outdoor organic systems absent, the performance of play behavior by animals is compared with the DanBred genetic hybrid (Schild et al., associated with the presence of positive-valence affective states 2020). Additionally, Schild et al. (2019) reported that Topigs due to its correlation with increased activity of brain areas Norsvin TN70 sows were more protective towards their pigs mediating hedonic properties of rewards (Burgdorf and during pig handling (i.e., earmarking and castration) than Panksepp, 2007; Berridge and Kringelbach, 2008; Horback, DanBred sows. Increased sow attention to the whole litter, 2014). Hence, play behavior is suggested to be a positive acquisition of passive immunity, and thermoregulation are animal welfare indicator (Š pinka et al., 2001; Boissy et al., 2007). crucial aspects for pigs’ viability and performance during the As reviewed by Weary et al. (2008), different management suckling and post-weaning period (Baxter et al., 2008; Declerck practices have been found to limit, at least partly, the negative et al., 2016). Frontiers in Animal Science 02 frontiersin.org Franchi et al. 10.3389/fanim.2022.909038 Accordingly, the aims of the current study were to batches were included in this study. The farrowing unit had two investigate the single and interactive effects of two weaning sections, each containing eight 3.0 × 2.2 m farrowing pens for loose- methods (conventional weaning: pigs moved to and mixed in housed sows (FT30, Jyden A/S, Denmark). In each pen, the floor a weaner pen vs. weaning in a farrowing pen for loose-housed was 41% slatted and 59% solid concrete, the sides of the pen were sows) and two genetic hybrids (DanBred vs. Topigs Norsvin made of metal bar and equipped with sloped walls to reduce the risk TN70) on play behavior in domestic pigs (Sus scrofa) during the of pig crushing by the sow. A hayrack was placed on the entrance days around weaning at 26 days of age. We hypothesized that, next to the covered pig creep area (0.9 m2). The solid floor surface before weaning, Topigs Norsvin TN70 pigs would play more was heated, and the creep area was equipped with a 150 W infrared than DanBred pigs due to a higher liveweight and consequently heat lamp (InterHeat CO., South Korea) providing 35°C in the improved thermoregulation and nutritional status. Additionally, creep area at farrowing. The temperature was gradually decreased we hypothesized that Topigs Norsvin TN70 pigs would be more until the heater was turned off in week 3 pp. One week before the resilient to weaning and consequently play more than DanBred expected date of farrowing, three DB and three TN gilts/sows were pigs after weaning. Moreover, we hypothesized that the practice allocated to each of the two rooms and, within room, sows were of keeping weaned pigs in the farrowing pen would represent a randomly allocated to farrowing pen. The farrowing pen was less stressful management practice and thus result in a higher equipped with a low 80 × 28 cm polyconcrete feed trough (Jyden performance of play compared with mixing litters in a weaner A/S, Denmark) allowing pigs to eat together with the sow. A water pen after separation from the dam. drinking nipple was placed 5 cm above the deepest point of the feed trough for the sow, permitting mixing of the lactation diet with water. This water drinking nipple was closed at weaning to omit Materials and methods water in the feed trough where the weaner diet was delivered post- weaning. A 31 × 17 × 11 cm pig water trough (Aqua-Level system Experimental design with hinged trough, Jyden A/S, Denmark) positioned next to the feed trough on the slatted floor provided ad libitum access to water The present study was conducted from February to throughout the experimental period. November 2020 at the Danish Pig Research Centre, Aarhus Sows received analgesic (Metacam®) and all pigs were University, Foulum, Denmark. The animal experimental weighted, ear tagged, and had their umbilical cord cut procedures and care of animal under study were carried out in following farrowing, if it ended before 15:00 h, or on the next accordance with the Ministry of Food, Agriculture and Fisheries, morning, if farrowing ended after 15:00 h. The number of own The Danish Veterinary and Food Administration under act 474 pigs nursed by the sows was adjusted to be equal to the number of 15 May 2014 and executive order 2028 of 14 December 2020, of functional teats. Surplus pigs within sow hybrid were cross and under consideration of the ARRIVE Guidelines (Du Sert fostered within the first 3 days pp to sows that had fewer pigs et al., 2020). The present study was part of a longer study than functional teats. If no sows were available for cross- examining the clinical, physiological, and behavioral effects of fostering, surplus pigs were euthanized by blunt force trauma. weaning intact litters from sows of different hybrids in farrowing Non-viable pigs and pigs below 700 g were euthanized first pens for loose-housed sows (Winters et al., submitted). followed by randomly selected pigs. On the day before weaning, The experimental design followed a 2 × 2 factorial arrangement the numbers of pigs per pen in the DB and TN genetic hybrids with the following factors: weaning method [two litters grouped in a were 12 ± 2 and 13 ± 2, respectively. Pigs received 1 ml of conventional two-climate weaner pen (MOVE) vs. intact litter that subcutaneous iron injection (Viloferron®, 20%) on day 4 pp and stayed in the farrowing pen at weaning after the sow was removed male pigs were castrated after receiving a local anaesthetic (0.5 (STAY)] and sow genetic hybrid [DanBred Landrace × Yorkshire ml of Procamidor®Vet 20 mg/ml per testis) and analgesic (0.2 (approximately 21 total pigs born/litter; DB) vs. Topigs Norsvin ml of Melovem® 20 mg/ml i.m.). Pigs were vaccinated with TN-70 Yorkshire × Landrace (approximately 16 total pigs born/ Ingelvac MycoFLEX® on day 7 pp and again at weaning. No pigs litter; TN)]. Both genetic hybrids were inseminated with semen of were tail-docked or tooth-clipped. DanBred Duroc boars. The artificial light was off from 22:00 h to 07:00 h. Every morning, feed troughs were emptied before first feeding, and manure was removed. Straw racks were filled with wheat straw Animals, housing, and management daily and approximately 130 g of fine chopped wheat straw were provided on the solid floor daily. Approximately 400 g of sawdust Lactation period was provided in the pig creep area daily. At farrowing, the room Twelve gilts (6 DB and 6 TN gilts) were included in one batch. temperature was set at 21°C and gradually decreased to 19°C After farrowing and lactation, these sows returned in a second batch at weaning. as second-parity sows. In total, 306 pigs (50% castrated males, 50% Sows were fed a pelleted standardized lactation diet (Die females) from 24 litters [pigs/litter: (average ± SD) 13 ± 2] over two Profil Lac, DLG, Frederica, Denmark, metabolizable energy: 12.9 Frontiers in Animal Science 03 frontiersin.org Franchi et al. 10.3389/fanim.2022.909038 MJ/kg, 15.6% crude protein) following the Danish were ad libitum fed a pelleted standardized weaner diet (Prime recommended feeding curve (Tybirk et al., 2018) delivered 4 Midi Piller, DLG, Denmark, metabolizable energy: 14.8 MJ/kg, times daily at 08:30 h, 11:00 h, 15:30 h, and 22:00 h. If the feed 19.3% crude protein), with no added zinc oxide, delivered once trough was empty in the morning, from day 14 pp until weaning, daily in the morning. No pigs were treated with antibiotics the sows’ daily feed allowance was increased by 2% to ensure that during the experimental period. there was feed in the trough for the main part of the 24 h to encourage pigs to start eating solid feed. Pigs did not receive weaner diet before weaning, but they had free access to the sow Play behavior recording diet during the entire lactation period. Play behavior was observed from video recordings on days −1, Post-weaning period 1, and 2 relative to the weaning day (day 0). A 2D video camera In each batch, half of the DB and TN litters were randomly (model DS-2CD2145FWD-I, Hikvision, China) was placed above allocated to one of the two weaning methods, STAY or MOVE. This each pen with a top-view angle, overlooking the entire pen area, resulted in the following total number of pens and pigs per pen in except under the creep area in the farrowing pen and under the each weaning method × genetic hybrid combination: STAY × DB: cover in the weaner pen, from a height of approximately 2.3 m in four pens (12 ± 2 pigs/pen), MOVE × DB: four pens (25 ± 3 pigs/ the farrowing pen and 2.8 m in the weaner pen. Video data were pen), STAY × TN: four pens (14 ± 2 pigs/pen), and MOVE × TN: processed with the software Blue Iris v.5 (Blue Iris Security, USA). four pens (26 ± 2 pigs/pen). Litters assigned to the STAY treatment Each pig was identified on the video recordings by use of markings stayed in the farrowing pens for loose-housed sows after the sow on the back made with a livestock marker on days −2, 1, and 2. was removed out of sensory range of detection by the pigs. Litters Play behavior was observed using behavioral sampling and assigned to the MOVE treatment were moved to one of four continuous recording (Bateson and Martin, 2021) of the identical 5.40 × 2.45 m two-climate weaner pens (approximately individual pig from 14:00 h to 22:00 h on each experimental 0.51 m2/pig) with approximately 33% slatted, 33% drained, and day (−1, 1, and 2) using the Noldus Observer XT 15 (Noldus 33% solid floor in the same weaner unit. The solid floor was roofed Information Technology, The Netherlands) software package. by a manually adjustable fiber panel positioned approximately 85 Pigs were expected to play with the highest frequency during cm above the floor level (Jyden A/S, Denmark) providing two zones light hours (07:00 to 22:00 h). However, the hours from 07:00 to in the pen (a cooler slatted and drained floor area and a warmer and 14:00 h included experimental and management procedures darker area under the roof). The water trough was of the same type disturbing the pig such as general management, marking, and as used in the farrowing pens. Two weaner feeders (TR4, Rotecna®, weighing of the pigs, and data sampling. Thus, it was chosen to Spain) were placed next to each other in the partially slatted floor only observe the pigs during the undisturbed hours from 14:00 h area providing eight 14 × 18 cm feeding places per weaner pen. On to 22:00 h on each day. The durations of two types of play the day of weaning, experimental pigs were (average ± SD) 26 ± 1 behavior were recorded: locomotor-rotational play (LOC; days old and weighed 7.4 ± 1.9 kg (DB: 7.0 ± 2.0 kg; TN: 7.8 ± 1.8 Table 1) and social play (SOC; Table 2). When a behavior kg). To accommodate all-in all-out farrowing for hygiene purposes, duration was registered, the observer returned the video to the weaning took place on 1 day per week, as close as possible to the day start time of this behavior and continued the observations from when litters were 28 days old. that time point to ensure that no play behavior was missed. If Pigs were weaned at 14:00 h. Litters in treatment STAY were necessary, the identities of the pigs participating in the play were locked inside the pig creep area while the sow was moved out of noted to avoid registering the same events twice. The durations the farrowing pen. Lactation diet leftovers were removed from all of LOC and SOC of the individual pig comprised the sum of feed troughs, the water nipple was turned off and a weaner diet durations of all behavioral elements constituting each type of was offered. At the time of weaning and every morning post- play, respectively, as displayed by the individual pig during the weaning, all STAY pens were provided approximately 130 g of 8-h observation period. Whenever pigs were disturbed (i.e., a chopped wheat straw on the solid floor and approximately 400 g human was within the framework of the pen, the pigs were not in of sawdust in the creep area. Litters allocated to treatment the pen, or the pigs were being moved in or out of the pen), this MOVE were grouped with another entire litter of the same was noted as disturbance, and no behavior registrations were breed and walked to a conventional weaner pen in another made. The duration of disturbance across pens ranged from 0 to section of the building located approximately 80 m from the 150 s on day −1, 0 to 47 s on day 1, and 0 to 51 s on day 2. farrowing unit. The MOVE pens received double the amount of Three experimenters (A, B, and C) were assigned to observe straw and sawdust (to maintain same provision per animal) play behavior from video recordings, balanced for genetic hybrid placed on the drained floor and on the solid floor, respectively. and weaning method. Two experimenters (A and B) observed The room temperature was increased to 24°C and gradually batch 1, with experimenter A also observing batch 2 together decreased to 21°C 56 days post-weaning in both treatments. Pigs with a third experimenter (C) (Table 3). Each experimenter was Frontiers in Animal Science 04 frontiersin.org Franchi et al. 10.3389/fanim.2022.909038 TABLE 1 Ethogram of behavioral elements constituting locomotor-rotational play in domesticated pigs (Sus scrofa). Behavior Description Modified from Locomotor- A pig performs one of the locomotor play behavior elements: gambolling or scamper; or one of the rotational play behavior elements: rotational flop, head toss, hop/pivot, turn, or shake straw. Behavioral elements are mutually exclusive and recorded with exact start and stop play times. If a pig stops playing for a full second, it is noted as a new event. Elements of locomotor-rotational play Gambolling Gallop-like energetic running (the two forelimbs move approximately in phase, followed by the two hind limbs) and hopping in Brown et al. (2015) forward motions within the pen environment. Often associated with vigorous ear flapping, moving across a large area of the pen, and occasionally bouncing into other pigs. As part of a bout of gambolling, the pig may turn and change direction. The behavior ends when the pigs move at a slower pace or go out of sight. Scamper Performing at least two hops forward in rapid succession usually accompanied by ear flapping and sometimes accompanied by head Donaldson et al. tossing. The behavior ends when the pigs move at a slower pace or go out of sight. (2002); Martin et al. (2015); Zupan et al. (2016) Flop A rapid drop from an upright position to sternal or lateral recumbency in which the pig appears to fall by itself and not because of Donaldson et al. contact with another pig. (2002) Head toss Exaggerated lateral movement of the head and neck in the horizontal plane, involving at least one full movement to each side. Donaldson et al. (2002) Hop/Pivot Making a short, bouncing leap on the spot during which all feet are simultaneously lifted off the ground, body oriented in the same Newberry et al. direction throughout the hop or rotated rapidly in the horizontal plane. (1988); Donaldson et al. (2002); Zupan et al. (2016) Turn Rapid turning around the body axis on the spot during which at least one foot is touching the ground. Zupan et al. (2016) Shake straw Perform vigorous lateral movements of head and neck while holding straw, which protrudes from mouth. Newberry et al. (1988) Throw Holding straw in the mouth and throwing it through the air by a rapid movement of the head and neck. Zupan et al. (2016) straw trained in observing play until a concordance correlation (0.5 to 0.7), high (0.7 to 0.9), or very high (0.9 to 1.00). The CCC coefficient (CCC) ≥ 0.80 was achieved between experimenters calculations accounted for the total duration of play (i.e., LOC within each batch. The CCC is an integrated measure of and SOC durations combined) of the individual pig. During the correlation as well as accuracy and precision (Lawrence and training, each experimenter within a given batch was given the Lin, 1989) and was herein interpreted according to the criteria same 1-h video to observe for each of the three housing proposed by Hinkle et al. (2003) indicating the level of situations: farrowing pen with sow, farrowing pen without agreement: negligible (0.0 to 0.3), low (0.3 to 0.5), moderate sow, and two-climate weaner pen. After observing these 3 h of TABLE 2 Ethogram of behaviors constituting social play in domesticated pigs (Sus scrofa). Behavior Description Modified from Social play Energetic mutual interaction between two or more upright pigs initiated by play fighting and may include the following elements: climb, follow/chase, lever, nudge, and push (see below). Pigs display neither biting nor avoidance. The social play ends when the pigs are no longer in contact, or when one pig no longer shows interest, or when they stop follow/chase, or if at least one of the pigs moves out of sight. Elements of social play Climb Placing both front hoofs on the back of another pig. Brown et al. (2015) Follow/ Walking/running (often with physical contact) in the same direction as another pig. Zupan et al. chase (2016) Lever With the snout lifting another pig so that at least two feet are lifted off the floor. Donaldson et al. (2002) Nudge Use of snout to gently touch another pig’s body, not including naso–naso contact. Brown et al. (2018) Push Using head, neck, or shoulders with minimal or moderate force to drive into another pig’s body. Brown et al. (2018) Play fight Two pigs mutually push and head-knock each other. A general mild intensity of the performed fighting behaviors and a lack of biting or Brown et al. avoidance distinguish play fight from aggressive fighting. (2015) Frontiers in Animal Science 05 frontiersin.org Franchi et al. 10.3389/fanim.2022.909038 TABLE 3 Breakdown of the number of pens and pigs per pen following a 2 × 2 factorial arrangement of the factors weaning method (litters mixed in a conventional weaning pen; MOVE vs. litter stayed in the farrowing pen after removing the sow; STAY) and genetic hybrid (DanBred; DB vs. Topigs Norsvin TN70; TN) observed by each experimenter on each day (−1, 1, and 2) relative to weaning. Day Treatment combination Experimenter A B C -1 MOVE × DB 4 pens, 11 ± 2 2 pens, 13 ± 1 2 pens, 15 ± 1 MOVE × TN 4 pens, 13 ± 2 2 pens, 13 ± 1 2 pens, 13 ± 0 STAY × DB 2 pens, 13 ± 1 1 pen, 14 1 pen, 10 STAY × TN 2 pens, 12 ± 3 1 pen, 15 1 pen, 15 1 MOVE × DB 1 pen, 28 2 pens, 24 ± 3 1 pen, 24 MOVE × TN 3 pens, 27 ± 1 — 1 pen, 23 STAY × DB 2 pens, 13 ± 1 1 pen, 14 1 pen, 10 STAY × TN 2 pens, 12 ± 3 1 pen, 15 1 pen, 15 2 MOVE × DB 1 pen, 28 2 pens, 24 ± 3 1 pen, 24 MOVE × TN 3 pens, 27 ± 1 — 1 pen, 23 STAY × DB 2 pens, 13 ± 1 1 pen, 14 1 pen, 10 STAY × TN 2 pens, 12 ± 3 1 pen, 15 1 pen, 15 Number of pigs per pen is presented as average ± standard deviation, when more than one pen was observed. video, CCC was calculated by the Noldus Observer XT software effects. Additionally, a continuous-time autoregressive and any disagreement between experimenters was discussed. covariance structure was included to account for repeated This process was repeated on 1-h videos until the acceptable measures of each pig over days. Model assumptions of CCC was achieved (batch 1: 0.86, batch 2: 0.87, all combinations normality and homoscedasticity were confirmed through of experimenters across batches: 0.82). Intra-experimenter graphical inspection of the residuals. To achieve a satisfactory agreement was also tested for the experimenter observing both fit of the model, the initial SOC model was reduced by removing batches, achieving a CCC of 0.84. the three-way interaction. Tukey-adjusted post-hoc analyses were performed for each model (library emmeans v.1.6.3; Lenth, 2021). Statistical analyses Statistical analyses were performed in R v.4.1.1 (R Core Results Team, 2021). p-values < 0.05 were considered significant, and p- values of 0.1 > p ≥ 0.05 were considered tendencies. The A significant interaction between weaning method and day durations (s/pig/8-h) of LOC and SOC were analyzed over on LOC was detected (Table 4; Figure 1). On each day, no experimental days at the individual pig level, accounting for statistical difference in time spent in LOC between MOVE and the nested structure in data with pigs nested within pens. STAY was observed. However, performance of LOC in both Because of the percentage of zero observations (day −1: 11%; weaning method groups decreased from days −1 to 1 and STAY day 1: 49%; day 2: 26%), LOC was square-root-transformed to pigs increased time spent in LOC from days 1 and 2. A fulfill the assumptions of normality and homoscedasticity and significant interaction between genetic hybrid and day on LOC analyzed with a linear mixed-effects model with a Gaussian was also detected (Table 4; Figure 2). On day −1, TN pigs distribution (library glmmTMB v.1.1.2; Brooks et al., 2017). Due performed LOC longer than DB pigs. From days −1 to 1, to the high percentage of zero observations (day −1: 67%; day 1: performance of LOC decreased in both genetic hybrid groups. 95%; day 2: 92%), SOC was converted into a binary variable Conversely, TN pigs performed more LOC on day 2 than on day (played; did not play) and analyzed with a mixed-effects logistic 1. A tendency for interaction among weaning method, genetic regression model (library glmmTMB v.1.1.2). Both LOC and hybrid, and day on LOC was detected (Table 4; Figure 3). SOC models initially included sex (castrated male; female), Irrespective of weaning method or genetic hybrid, the odds of weaning method (MOVE; STAY), genetic hybrid (DB; TN), displaying SOC were higher on day −1 than days 1 and 2 [day −1/ day (−1; 1; 2), all possible two-way interactions between weaning day 1: (odds ratio ± standard error) 9.4 ± 3.05; day −1/day 2: 6.6 ± method, genetic hybrid, and day, and the three-way interaction 1.98], while there was no statistical difference between days 1 and 2 among weaning method, genetic hybrid, and day. Batch, pig, pen (0.7 ± 0.27) (Table 4). A significant interaction between weaning before weaning, and pen after weaning were included as random method and day on SOC was detected (Table 4; Figure 4). On day 2, Frontiers in Animal Science 06 frontiersin.org Franchi et al. 10.3389/fanim.2022.909038 TABLE 4 Test statistics (c2 test) and p-values for the duration (s/pig/8-h) of locomotor-rotational play (LOC) and social play (SOC) in weaner pigs used in a 2 × 2 factorial arrangement of treatment factors: weaning method (litters mixed in a conventional weaning pen; MOVE vs. litter stayed in the farrowing pen after removing the sow; STAY) and genetic hybrid (DanBred; DB vs. Topigs Norsvin TN70; TN), on days −1, 1, and 2 relative to the weaning day. Variable Statistics Sex1 Weaning Genetic Day2 Weaning method × Weaning Genetic Weaning method × method1 hybrid1 Genetic hybrid1 method × hybrid × Genetic hybrid × Day2 Day2 Day2 LOC c2 test 0.04 0.18 0.06 403.67 3.15 6.80 6.62 5.17 p-value 0.828 0.666 0.804 < 0.075 0.033 0.036 0.075 0.001 SOC c2 test 3.77 0.24 0.21 72.75 0.69 20.56 5.76 — p-value 0.051 0.623 0.651 < 0.404 < 0.001 0.056 — 0.001 1 Degrees of freedom = 1; 2Degress of freedom = 2. STAY pigs had a higher probability of displaying SOC than MOVE within 24 h after farrowing. Conversely, TN sows typically farrow pigs. A tendency for interaction between genetic hybrid and day on slightly smaller litters composed of heavier live-born pigs with SOC was detected (Table 4; Figure 5). Castrated male pigs tended to higher rectal temperature 24 h post-farrowing than the DB genetic show higher odds of displaying SOC than female pigs (1.5 ± hybrid (Schild et al., 2020). Reduced birth weight and reduced rectal 0.33; Table 4). temperature are accompanied by lower body energy reserves and higher sensitivity to cold (Pedersen et al., 2011), jeopardizing the ability of pigs to reach the teats and delaying the first suckling event Discussion (Le Dividich and Sève, 2000; Quiniou et al., 2002). The consequence of this cascade is increased risk of mortality and reduced weight on Pre-weaning period the expected day of weaning (Pedersen et al., 2015), which may decrease pigs’ resilience to abrupt, earlier than natural, weaning Our finding that TN pigs engaged more in LOC compared with (Wolter and Ellis, 2001; Collins et al., 2017). In the current study, DB pigs before weaning may be linked to the respective breeding TN pigs were heavier than the DB pigs on the day of weaning. goals, and consequent pig characteristics at birth, of each genetic Additionally, Winters et al. (submitted), accounting for a larger hybrid. Litters originating from DB sows feature a high number of population of pigs (including the current study’s population), live-born pigs with low birth weight and low rectal temperature reported higher average daily weight gain during the week prior FIGURE 1 Time spent performing locomotor-rotational play (LOC), in s/pig/8-h, in pigs moved to a weaner pen (MOVE; red bars) or kept in the farrowing pen (STAY; green bars) after weaning, between 14:00 h and 22:00 h on days −1, 1, and 2 relative to weaning. The bars represent least squares means and the error bars indicate standard error. Estimates and standard error bars were square root transformed prior to analysis, and the back-transformed least squares means are shown within a box on the top of each bar. Different superscript letters indicate statistical difference at p < 0.05 both between and within days. Frontiers in Animal Science 07 frontiersin.org Franchi et al. 10.3389/fanim.2022.909038 FIGURE 2 Time spent performing locomotor-rotational play (LOC), in s/pig/8-h, in DanBred pigs (DB; gray bars) or Topigs Norsvin TN70 pigs (TN; yellow bars) between 14:00 h and 22:00 h on days −1, 1, and 2 relative to weaning. The bars represent least squares means and the error bars indicate standard error. Estimates and standard error bars were square root transformed prior to analysis, and the back-transformed least squares means are shown within a box on the top of each bar. Different superscript letters indicate statistical difference at p < 0.05 both between and within days. to the weaning day and higher weight at weaning in TN pigs than in Consequently, the difference in time spent playing between the DB pigs. As play behavior has been demonstrated to be sensitive to two genetic hybrids could have been greater if litter size had not negative physical and physiological conditions (Siviy and Panksepp, been harmonized to a similar number of suckling piglets in the two 1985; Newberry et al., 1988), our findings support our hypothesis genetic hybrids. Commercial farms using DB sows increasingly that TN pigs would be more playful than DB pigs because of their allow a higher number of piglets suckle the sow than the number of higher weights prior to weaning. Had the litter size in DB and TN her functional teats to manage the large litter size, a strategy that groups not been equalized to the number of functional teats, the results in increased within-litter competition and reduced weight weight gain of the DB pigs would likely have been less uniform, and gain of suckling piglets (Kobek-Kjeldager et al., 2020a; Kobek- lower, due to increased within-litter competition for teats. Kjeldager et al., 2020b). FIGURE 3 Time spent performing locomotor-rotational play (LOC), in s/pig/8-h, in pigs moved to a weaner pen (MOVE; red bars) or kept in the farrowing pen (STAY; green bars) and DanBred pigs (DB; solid bars) or Topigs Norsvin TN70 pigs (TN; striped bars) between 14:00 h and 22:00 h on days −1, 1, and 2 relative to weaning. The bars represent least squares means and the error bars indicate standard error. Estimates and standard error bars were square root transformed prior to analysis, and the back-transformed least squares means are shown within a box on the top of each bar. Frontiers in Animal Science 08 frontiersin.org Franchi et al. 10.3389/fanim.2022.909038 FIGURE 4 Probabilities of displaying social play (SOC) in pigs moved to a weaner pen (MOVE; red bars) or kept in the farrowing pen (STAY; green bars) after weaning, between 14:00 h and 22:00 h on days −1, 1, and 2 relative to weaning. The bars represent probability, and the error bars indicate standard error. Estimates and standard error bars were back-transformed from the model using the inverse link function. Different superscript letters indicate statistical difference at p < 0.05 both between and within days. Post-weaning period and an abrupt separation from the sow and change to solid feed, likely jeopardizes their welfare, as reported in other animal species Performance of play behavior was lower on the first days after (e.g., dairy calves, Krachun et al., 2010; deer fawns, Muller-Schwarze weaning compared to the day before weaning, irrespective of et al., 1982). Our findings corroborate the understanding that early genetic hybrid or weaning method. This pattern in performance weaning challenges the welfare of pigs, who are too young and of play is in accordance with previous studies on play behavior in immature to cope easily with simultaneous stressors associated with pigs around weaning (e.g., Donaldson et al., 2002; Hötzel et al., this management practice involving separation from the sow, 2010; Colson et al., 2012). It has been suggested that pigs of the age abrupt transition from milk to solid feed, and social mixing in a used in this study are still nutritionally dependent of the sow’s milk, novel environment. FIGURE 5 Probabilities of displaying social play (SOC) in DanBred pigs (DB; gray bars) or Topigs Norsvin TN70 pigs (TN; yellow bars) between 14:00 h and 22:00 h on days −1, 1, and 2 relative to weaning. The bars represent probability, and the error bars indicate standard error. Estimates and standard error bars were back-transformed from the model using the inverse link function. Frontiers in Animal Science 09 frontiersin.org Franchi et al. 10.3389/fanim.2022.909038 Our results did not support the hypothesis that TN pigs would Our results show that LOC durations rose from days 1 to 2, be more playful than DB pigs after weaning. The TN pigs showed a especially in STAY pigs and TN pigs. If pigs had been observed for a steeper decrease in time spent performing LOC from days −1 to 1 longer period following weaning, we may have seen a level of play than the DB pigs, suggesting that the TN group was more affected similar to, or higher than, the pre-weaning level, and the effects of by the disappearance of the sow and abrupt dietary shift. weaning method, genetic hybrid, or both might have been stronger. Conversely, from days 1 to 2, the TN pigs showed a steeper For instance, Camerlink et al. (2021) reported a lower level of social increase in duration of LOC compared to the DB pigs, suggesting play in weaner pigs during the first 48 h and a subsequent increase a faster recovery from this management practice. Additionally, we from day 3 relative to weaning. Moreover, Donaldson et al. (2002) found that the STAY pigs also displayed a steeper increase in time and Brown et al. (2018) observed higher levels of locomotor play in spent in LOC from days 1 to 2 and a higher engagement in SOC on pigs at days 3 and 5 post-weaning compared with the pre-weaning day 2 than the MOVE pigs. These findings suggest that weaning period. Together with these studies, our results suggest that the pigs in the farrowing pen for loose-housed sows and, hence, adverse effects of weaning on pigs are relatively short-lived given the avoiding social mixing in a new environment can reduce weaning tendency to increase time spent performing locomotor play. stress. Social mixing requires establishment of new dominance However, Donaldson et al. (2002) and Brown et al. (2018) could relationships, which consequently involves aggression commonly not tease apart post-weaning effects from increased pen space leading to body injuries (Meese and Ewbank, 1973), and availability after weaning, or from a possible age effect, given that physiological and psychological stress (Campbell et al., 2013; locomotor play has been observed to peak at approximately 4 to 5 Peden et al., 2018). Previous studies found increased levels of weeks of age (Newberry et al., 1988). Hence, it would also be salivary cortisol, indicating stress, in pigs moved to a new complex to disentangle the effects of the genetic hybrid and weaning environment or socially mixed, or both, after weaning (e.g., method from these other aspects (i.e., space and age). Merlot et al., 2004; Colson et al., 2012). Accordingly, the The weaning method of keeping pigs in the farrowing pen and, activation of stress pathways may result in intestinal and hence, keeping litters intact across weaning may support occurrence immunological dysfunction compromising feed intake, growth, of play and prevent or, at least, limit aggressive behavior (Puppe and resistance against pathogens (Le Dividich and Sève, 2000; Pié et al., 1997). However, social mixing is unavoidable during the et al., 2004; Moeser et al., 2012). Altogether, these potential negative production cycle and later rather than early social mixing can consequences related to social mixing in a novel environment as increase the duration of fights and severity of injuries during the part of conventional weaning could have constituted a welfare establishment of a new hierarchy (e.g., Pitts et al., 2000; D’Eath, threat to the pigs (Held and Špinka, 2011) and have reduced their 2005). D’Eath (2005) stated that between 5 and 12 days of age is the motivation to engage in play. Meanwhile, weaning pigs in the most suitable time to mix pigs as they are minimally prone to fight or farrowing pen and, hence, avoiding placement in an unfamiliar bully other pigs. Accordingly, Weary et al. (2008) recommend social environment and social disruption by mixing litters may have mixing of pigs at this age due to the lower susceptibility to aggression limited the negative effects of this management procedure. Other and fear compared to later life stages (Hillmann et al., 2003). As stressors were still present, such as maternal separation and abrupt reviewed by Weary et al. (2008), increasing the weaning age, making dietary shift (e.g., Hillmann et al., 2003; Martin et al., 2015), likely the solid diet more attractive, or reducing pigs’ access to the sow for explaining the reduction in time spent playing on day 1 in pigs some hours daily prior to sow removal, so that pigs may achieve subjected to this treatment. partial nutritional independence before separation from the sow, are In this study, the level of play recorded was low compared to additional strategies for reducing weaning stress. We encourage previous studies examining play in (weaner) pigs (e.g., Hötzel et al., future studies using play behavior as an indicator of how well pigs 2010; Camerlink et al., 2021). Methodological differences may recover from weaning depending on weaning practices. provide a partial explanation. For instance, our SOC definition required two or more pigs to mutually engage in social play, while the above-mentioned studies did not require the social play to be Conclusions mutual. Furthermore, there can be circadian differences in play behavior (Ocepek et al., 2020) that may explain differences in results Our study contributes to the understanding of effects of weaning between studies. Thus, for a more accurate interpretation and on pigs and suggests welfare-oriented improvements. The use of a comparison of results within and between studies, it may be genetic hybrid with characteristics related to increased viability advisable to express time spent playing relative to the time active resulted in an increased performance of LOC before weaning. during the observation periods in future studies. Furthermore, due Weaning pigs in the farrowing pen and, hence, avoiding moving to to low occurrence, we were unable to analyze the SOC durations an unfamiliar environment and social disruption by mixing litters and had to analyze the data as a binary response per 8-h period. resultedinanincreasedoccurrenceofSOCplayonday2afterweaning. Using one-zero sampling with short intervals over more time Additionally, pigs belonging to the genetic hybrid with higher weaning periods per day could have been more efficient for capturing weight and pigs kept in the farrowing pen displayed a steeper recovery treatment differences in play behavior. in performance of LOC from days 1 to 2 relative to weaning. Frontiers in Animal Science 10 frontiersin.org Franchi et al. 10.3389/fanim.2022.909038 Our experimental pigs reduced their time spent playing funding acquisition and resources. ML and LP: project when several sources of stress were introduced, such as administration. MJ and LP: supervision. GF, ML, JW, MJ, and removal of the sow, sudden dietary change, social mixing, and LP: writing, review, and editing. All authors contributed to the placement in a novel environment, supporting the idea that play article and approved the submitted version. behavior is typically suppressed by welfare threats (Burgdorf and Panksepp, 2007; Held and Š pinka, 2011). Earlier studies similarly found a drop in play behavior in adverse situations Funding (e.g., disbudding in dairy calves, Mintline et al., 2013; low milk allowance in dairy calves, Jensen et al., 2015; poor weight gain in This manuscript was developed as part of project ClearFarm weaner pigs, Brown et al., 2015). That play behavior is reduced that has received funding from the European Union’s Horizon by such welfare threats means that suppression of play behavior 2020 research and innovation programme under grant can indicate the absence of positive affective states and thus poor agreement No. 862919. The data collection and the writing of animal welfare. However, it is the association with positive this manuscript were further funded by the Ministry of Food, experiences (e.g., that animals seek out opportunities to play Agriculture and Fisheries of Denmark under the program and solicit play) that qualifies play behavior as a positive welfare “Veterinary agreement III” (Veterinærforlig III, AU6), and by indicator. Keeping animals in environments that allow and the Marie Skłodowska-Curie grant agreement No. 842555 stimulate play behavior thus promotes positive welfare because (project AutoPlayPig). it gives the animals the opportunity for positive experiences. Evidence of play behavior increasing when favorable conditions become even more favorable would support play behavior as a positive welfare indicator (Ahloy-Dallaire et al., 2018). Hence, Acknowledgments we encourage future studies to examine the validity of play behavior as a positive welfare indicator by use of physiological The authors acknowledge the help from the staff at the Pig Research Unit, Aarhus University (AU), Denmark. We also (e.g., plasma oxytocin), behavioral (e.g., body postures), and cognitive measures (e.g., cognitive bias tests, motor/sensorial thank Carsten Kjærulff Christensen and Dines Thøger Bolt (both AU) for assistance with pig handling. Special thanks to lateralization) (as reviewed by Ahloy-Dallaire et al., 2018 and Kremer et al., 2020). Such an examination would allow for Ida Højgaard Kristoffersen (AU) in assisting with video observations. stronger inference regarding affective states and animal welfare. Data availability statement Conflict of interest The original contributions presented in the study are We declare that the study was conducted in the absence of included in the article/supplementary material. Further any commercial or financial relationships that could be inquiries can be directed to the corresponding author. construed as a potential conflict of interest. Ethics statement Publisher’s note Ethical review and approval was not required for the animal All claims expressed in this article are solely those of the study because the animal experimental procedures and care of authors and do not necessarily represent those of their affiliated animal under study were carried out in accordance with the organizations, or those of the publisher, the editors and the Ministry of Food, Agriculture and Fisheries, The Danish reviewers. Any product that may be evaluated in this article, or Veterinary and Food Administration under act 474 of 15. May claim that may be made by its manufacturer, is not guaranteed 2014 and executive order 2028 of 14. December 2020, and under or endorsed by the publisher. consideration of the ARRIVE Guidelines (Du Sert et al., 2020). Author contributions Supplementary material GF, ML, JW, MJ, and LP: conceptualization, investigation, The Supplementary Material for this article can be found and methodology. GF and ML: data curation, visualization, and online at: https://www.frontiersin.org/articles/10.3389/ writing. GF: formal analysis and software. ML, MJ, and LP: fanim.2022.909038/full#supplementary-material Frontiers in Animal Science 11 frontiersin.org Franchi et al. 10.3389/fanim.2022.909038 References Ahloy-Dallaire, J., Espinosa, J., and Mason, G. (2018). Play and optimal welfare: Hansen, C. (2021) National average for productivity in pig production 2020 does play indicate the presence of positive affective states? Behav. Processes 156, 3– [Landsgennemsnit for produktivitet i produktionen af grise 2020]. Available at: 15. doi: 10.1016/j.beproc.2017.11.011 https://svineproduktion.dk/publikationer/kilder/notater/2021/2115 (Accessed 08 Bateson, M., and Martin, P. (2021). Measuring behaviour: an introductory guide. September 2021). 4th ed. (Cambridge, UK: Cambridge University Press). doi: 10.1017/ Held, S. D., and Š pinka, M. (2011). Animal play and animal welfare. Anim. 9781108776462 Behav. 81, 891–899. doi: 10.1016/j.anbehav.2011.01.007 Baxter, E. M., Jarvis, S., D’Eath, R. B., Ross, D. W., Robson, S. K., Farish, M., et al. Hillmann, E., von Hollen, F., Bunger, B., Todt, D., and Schrader, L. (2003). (2008). Investigating the behavioural and physiological indicators of neonatal Farrowing conditions affect the reactions of piglets towards novel environment and survival in pigs. Theriogenology 69, 773–783. doi: 10.1016/ social confrontation at weaning. Appl. Anim. Behav. Sci. 81, 99–109. doi: 10.1016/ j.theriogenology.2007.12.007 S0168-1591(02)00254-X Berridge, K. C., and Kringelbach, M. L. (2008). Affective neuroscience of Hinkle, D. E., Wiersma, W., and Jurs, S. G. (2003). Applied statistics for the pleasure: reward in humans and animals. Psychopharmacology 199, 457–480. behavioural sciences. 3rd edition (Boston, MA, USA: Houghton Mifflin). doi: 10.1007/s00213-008-1099-6 Horback, K. (2014). Nosing around: play in pigs. Anim. Behav. Cogn. 1, 186–196. Boissy, A., Manteuffel, G., Jensen, M. B., Moe, R. O., Spruijt, B., Keeling, L. J., doi: 10.12966/abc.05.08.2014 et al. (2007). Assessment of positive emotions in animals to improve their welfare. Hötzel, M. J., Machado Filho, L. C. P., Irgang, R., and Alexandre Filho, L. (2010). Physiol. Behav. 92, 375–397. doi: 10.1016/j.physbeh.2007.02.003 Short-term behavioural effects of weaning age in outdoor-reared piglets. Animal 4, Brooks, M. E., Kristensen, K., van Benthem, K. J., Magnusson, A., Berg, C. W., 102–107. doi: 10.1017/S1751731109990875 Nielsen, A., et al. (2017). glmmTMB balances speed and flexibility among packages Hötzel, M. J., Pinheiro Machado Filho, L. C., Wolf, F. M., and Dalla Costa, O. A. for zero-inflated generalized linear mixed modeling. R J. 9, 378–400. doi: 10.3929/ (2004). Behaviour of sows and piglets reared in intensive outdoor or indoor ethz-b-000240890 systems. Appl. Anim. Behav. Sci. 86, 27–39. doi: 10.1016/j.applanim.2003.11.014 Brown, S. M., Klaffenböck, M., Nevison, I. M., and Lawrence, A. B. (2015). Jensen, M. B., Duve, L. R., and Weary, D. M. (2015). Pair housing and enhanced Evidence for litter differences in play behaviour in pre-weaned pigs. Appl. Anim. milk allowance increase play behavior and improve performance in dairy calves. J. Behav. Sci. 172, 17–25. doi: 10.1016/j.applanim.2015.09.007 Dairy Sci. 98, 2568–2575. doi: 10.3168/jds.2014-8272 Brown, S. M., Peters, R., Nevison, I. M., and Lawrence, A. B. (2018). Playful pigs: Jensen, P., and Stangel, G. (1992). Behaviour of piglets during weaning in a evidence of consistency and change in play depending on litter and developmental seminatural enclosure. Appl. Anim. Behav. Sci. 33, 227–238. doi: 10.1016/S0168- stage. Appl. Anim. Behav. Sci. 198, 36–43. doi: 10.1016/j.applanim.2017.09.018 1591(05)80010-3 Burgdorf, J., and Panksepp, J. (2007). The neurobiology of positive emotions. Kammersgaard, T. S., Pedersen, L. J., and Jørgensen, E. (2011). Hypothermia in Neurosci. Biobehav. Rev. 30, 173–187. doi: 10.1016/j.neubiorev.2005.06.001 neonatal piglets: interactions and causes of individual differences. J. Anim. Sci. 89, Camerlink, I., Proßegger, C., Kubala, D., Galunder, K., and Rault, J. L. (2021). 2073–2085. doi: 10.2527/jas.2010-3022 Keeping littermates together instead of social mixing benefits pig social behaviour Kobek-Kjeldager, C., Moustsen, V. A., Theil, P. K., and Pedersen, L. J. (2020a). and growth post-weaning. Appl. Anim. Behav. Sci. 235, 105230. doi: 10.1016/ Effect of litter size, milk replacer and housing on behaviour and welfare related to j.applanim.2021.105230 sibling competition in litters from hyper-prolific sows. Appl. Anim. Behav. Sci. 230, Campbell, J. M., Crenshaw, J. D., and Polo, J. (2013). The biological stress of 105032. doi: 10.1016/j.applanim.2020.105032 early weaned piglets. J. Anim. Sci. Biotechnol. 4, 1–4. doi: 10.1186/2049-1891-4-19 Kobek-Kjeldager, C., Moustsen, V. A., Theil, P. K., and Pedersen, L. J. (2020b). Collins, C. L., Pluske, J. R., Morrison, R. S., McDonald, T. N., Smits, R. J., Effect of litter size, milk replacer and housing on production results of hyper- Henman, D. J., et al. (2017). Post-weaning and whole-of-life performance of pigs is prolific sows. Animal 14, 824–833. doi: 10.1017/S175173111900260X determined by live weight at weaning and the complexity of the diet fed after Krachun, C., Rushen, J., and de Passillé , A. M. (2010). Play behaviour in dairy weaning. Anim. Nutr. 3, 372–379. doi: 10.1016/j.aninu.2017.01.001 calves is reduced by weaning and by a low energy intake. Appl. Anim. Behav. Sci. Colson, V., Martin, E., Orgeur, P., and Prunier, A. (2012). Influence of housing 122, 71–76. doi: 10.1016/j.applanim.2009.12.002 and social changes on growth, behaviour and cortisol in piglets at weaning. Physiol. Kremer, L., Holkenborg, S. K., Reimert, I., Bolhuis, J. E., and Webb, L. E. (2020). Behav. 107, 59–64. doi: 10.1016/j.physbeh.2012.06.001 The nuts and bolts of animal emotion. Neurosci. Biobehav. Rev. 113, 273–286. Damm, B. I., Pedersen, L. J., Jessen, L. B., Thamsborg, S. M., Mejer, H., and doi: 10.1016/j.neubiorev.2020.01.028 Ersboll, A. K. (2003). The gradual weaning process in outdoor sows and piglets in Lawrence, I., and Lin, K. (1989). A concordance correlation coefficient to relation to nematode infections. Appl. Anim. Behav. Sci. 82, 101–120. doi: 10.1016/ evaluate reproducibility. Biometrics 45, 255–268. doi: 10.2307/2532051 S0168-1591(03)00052-2 Le Dividich, J., and Sève, B. (2000). Effects of underfeeding during the weaning D’Eath, R. B. (2005). Socialising piglets before weaning improves social period on growth, metabolism, and hormonal adjustments in the piglet. Domest. hierarchy formation when pigs are mixed postweaning. Appl. Anim. Behav. Sci. Anim. Endocrinol. 19, 63–74. doi: 10.1016/S0739-7240(00)00067-9 93, 199–211. doi: 10.1016/j.applanim.2004.11.019 Leeb, C., Hegelund, L., Edwards, S., Mejer, H., Roepstorff, A., Rousing, T., et al. Declerck, I., Dewulf, J., Sarrazin, S., and Maes, D. (2016). Long-term effects of (2014). Animal health, welfare and production problems in organic weaner pigs. colostrum intake in piglet mortality and performance. J. Anim. Sci. 94, 1–11. Org. Agric. 4, 123–133. doi: 10.1007/s13165-013-0054-y doi: 10.2527/jas.2015-9564 Lenth, R. V. (2021) Emmeans: estimated marginal means, aka least-squares Donaldson, T. M., Newberry, R. C., Š pinka, M., and Cloutier, S. (2002). Effects of means. r package version 1.6.3. Available at: https://CRAN.R-project.org/package= early play experience on play behaviour of piglets after weaning. Appl. Anim. emmeans (Accessed May 4, 2022). Behav. Sci. 79, 221–231. doi: 10.1016/S0168-1591(02)00138-7 Martin, J. E., Ison, S. H., and Baxter, E. M. (2015). The influence of neonatal Du Sert, N. P., Ahluwalia, A., Alam, S., Avey, M. T., Baker, M., Browne, W. J., environment on piglet play behaviour and post-weaning social and cognitive et al. (2020). Reporting animal research: explanation and elaboration for the development. Appl. Anim. Behav. Sci. 163, 69–79. doi: 10.1016/j.applanim.2014.11.022 ARRIVE guidelines 2.0. PloS Biol. 18, e3000411. doi: 10.1371/journal.pbio.3000411 Meese, G. B., and Ewbank, R. (1973). The establishment and nature of the Dybkjær, L. (1992). The identification of behavioural indicators of ‘stress’ in dominance hierarchy in the domesticated pig. Anim. Behav. 21, 326–334. early weaned piglets. Appl. Anim. Behav. Sci. 35, 135–147. doi: 10.1016/0168-1591 doi: 10.1016/S0003-3472(73)80074-0 (92)90004-U Merlot, E., Meunier-Salaun, M. C., and Prunier, A. (2004). Behavioural, Foisnet, A., Farmer, C., David, C., and Quesnel, H. (2010). Relationships endocrine and immune consequences of mixing in weaned piglets. Appl. Anim. between colostrum production by primiparous sows and sow physiology around Behav. Sci. 85, 247–257. doi: 10.1016/j.applanim.2003.11.002 parturition. J. Anim. Sci. 88, 1672–1683. doi: 10.2527/jas.2009-2562 Milligan, B. N., Fraser, D., and Kramer, D. L. (2001). Birth weight variation in Fraser, D. (1975). Vocalizations of isolated piglets. II some environmental the domestic pig: effects on offspring survival, weight gain and suckling behaviour. factors. Appl. Anim. Ethol. 2, 19–24. doi: 10.1016/0304-3762(75)90062-0 Appl. Anim. Behav. Sci. 73, 179–191. doi: 10.1016/S0168-1591(01)00136-8 Früh, B., Bochicchio, D., Edwards, S., Hegelund, L., Leeb, C., Sundrum, A., et al. Mintline, E. M., Stewart, M., Rogers, A. R., Cox, N. R., Verkerk, G. A., Stookey, J. M., (2014). Description of organic pig production in Europe. Org. Agric. 4, 83–92. et al. (2013). Play behavior as an indicator of animal welfare: disbudding in dairy calves. doi: 10.1007/s13165-013-0056-9 Appl. Anim. Behav. Sci. 144, 22–30. doi: 10.1016/j.applanim.2012.12.008 Frontiers in Animal Science 12 frontiersin.org Franchi et al. 10.3389/fanim.2022.909038 Moeser, A. J., Borst, L. B., and Pittman, J. S. (2012). Defects in small intestinal Puppe, B., Tuchscherer, M., and Tuchscherer, A. (1997). The effect of housing epithelial barrier function and morphology associated with peri-weaning failure to conditions and social environment immediately after weaning on the agonistic thrive syndrome (PFTS) in swine. Res. Vet. Sci 93, 975–982. doi: 10.1016/ behaviour, neutrophil/lymphocyte ratio, and plasma glucose level in pigs. Livest. j.rvsc.2012.01.003 Prod. Sci. 48, 157–164. doi: 10.1016/S0301-6226(97)00006-7 Moustsen, V. A., and Nielsen, M. B. (2017) Mammary glands and teats on Quiniou, N., Dagorn, J., and Gaudré , D. (2002). Variation of piglets’ birth weight Danish sows [in Danish: Mælkekirtler og patter på danske søer]. Available at: and consequences on subsequent performance. Livest. Prod. Sci. 78, 63–70. https://docplayer.dk/69475437-Maelkekirtler-og-patter-paa-danske-soeer.html doi: 10.1016/S0301-6226(02)00181-1 (Accessed 09 September 2021). R Core Team (2021) R: A language and environment for statistical computing Muller-Schwarze, D., Stagge, B., and Muller-Schwarze, C. (1982). Play behavior: (Vienna: R Foundation for Statistical Computing). Available at: https://www.R- persistence, decrease, and energetic compensation during food shortage in deer project.org/ (Accessed May 4, 2022). fawns. Science 215, 85–87. Rutherford, K. M. D., Baxter, E. M., D'Eath, R. B., Turner, S. P., Arnott, G., Newberry, R. C., Wood-Gush, D. G. M., and Hall, J. W. (1988). Playful Roehe, R., et al. (2013). The welfare implications of large litter size in the domestic behaviour of piglets. Behav. Processes 17, 205–216. doi: 10.1016/0376-6357(88) pig I: biological factors. Anim. Welf. 22, 199–218. doi: 10.7120/09627286.22.2.199 90004-6 Schild, S. L., Rangstrup-Christensen, L., Thorsen, C. K., Bonde, M. K., and Ocepek, M., Newberry, R. C., and Andersen, I. L. (2017). Trade-offs between Pedersen, L. J. (2019). The course of parturition in two sow genotypes and two hut litter size and offspring fitness in domestic pigs subjected to different genetic designs under free-range conditions. Appl. Anim. Behav. Sci. 213, 55–64. selection pressures. Appl. Anim. Behav. Sci. 193, 7–14. doi: 10.1016/ doi: 10.1016/j.applanim.2019.02.005 j.applanim.2017.03.008 Schild, S. L. A, Foldager, L., Rangstrup-Christensen, L., and Pedersen, L. J. Ocepek, M., Newberry, R. C., and Andersen, I. L. (2020). Which types of rooting (2020). Characteristics of piglets born by two highly prolific sow hybrids. Front. material give weaner pigs most pleasure? Appl. Anim. Behav. Sci. 231, 105070. Vet. Sci 7, 355. doi: 10.3389/fvets.2020.00355 doi: 10.1016/j.applanim.2020.105070 Siviy, S. M., and Panksepp, J. (1985). Energy balance and play in juvenile rats. Pajor, E. A., Weary, D. M., Caceres, C., Fraser, D., and Kramer, D. L. (2002). Physiol. Behav. 35, 435–441. doi: 10.1016/0031-9384(85)90320-8 Alternative housing for sows and litters: effects of diet quality for piglets on pre- Š pinka, M. (2017). Behaviour of pigs” in the ethology of domestic animals. 3rd and post-weaning behaviour and performance. Appl. Anim. Behav. Sci. 76, 267– Edition. Ed. P. Jensen (Wallingford, Oxfordshire, UK: CAB International 277. doi: 10.1016/S0168-1591(02)00010-2 Publishing), 214–227. Peden, R. S., Turner, S. P., Boyle, L. A., and Camerlink, I. (2018). The translation Š pinka, M., Newberry, R. C., and Bekoff, M. (2001). Mammalian play: training of animal welfare research into practice: the case of mixing aggression between pigs. for the unexpected. Q. Rev. Biol. 76, 141–168. doi: 10.1086/393866 Appl. Anim. Behav. Sci. 204, 1–9. doi: 10.1016/j.applanim.2018.03.003 Tybirk, P., Sloth, N. M., Kjeldsen, N., and Shooter, L. (2018). “Normer for Pedersen, L. J., Berg, P., Jorgensen, G., and Andersen, I. L. (2011). Neonatal næringsstoffer,” in SEGES svineproduktion Aarhus, Denmark: SEGES Gris. piglet traits of importance for survival in crates and indoor pens. J. Anim. Sci. 89, 1207–1218. doi: 10.2527/jas.2010-3248 Weary, D. M., Appleby, M. C., and Fraser, D. (1999). Responses of piglets to Pedersen, L. J., Schild, S. L. A., and Malmkvist, J. (2015). The influence of the early separation from the sow. Appl. Anim. Behav. Sci. 63, 289–300. doi: 10.1016/ thermal environment and other early life events on growth rate of piglets during S0168-1591(99)00021-0 lactation. Animal 9, 1529–1535. doi: 10.1017/S1751731115001007 Weary, D. M., Jasper, J., and Hötzel, M. J. (2008). Understanding weaning Pié , S., Lallè s, J. P., Blazy, F., Laffitte, J., Sève, B., and Oswald, I. P. (2004). distress. Appl. Anim. Behav. Sci. 110, 24–41. doi: 10.1016/j.applanim.2007.03.025 Weaning is associated with an upregulation of expression of inflammatory Wolter, B. F., and Ellis, M. (2001). The effects of weaning weight and rate of cytokines in the intestine of piglets. J. Nutr. 134, 641–647. doi: 10.1093/jn/ growth immediately after weaning on subsequent pig growth performance and 134.3.641 carcass characteristics. Can. J. Anim. Sci. 81, 363–369. doi: 10.4141/A00-100 Pitts, A. D., Weary, D. M., Pajor, E. A., and Fraser, D. (2000). Mixing at young Zupan, M., Rehn, T., De Oliveira, D., and Keeling, L. J. (2016). Promoting ages reduces fighting in unacquainted domestic pigs. Appl. Anim. Behav. Sci. 68, positive states: The effect of early human handling on play and exploratory 191–197. doi: 10.1016/S0168-1591(00)00104-0 behaviour in pigs. Animal 10, 135–141. doi: 10.1017/S1751731115001743 Frontiers in Animal Science 13 frontiersin.org