Under semi-natural conditions calves experience a complex social environment, engaging in social interactions not only with their dam but also with other conspecifics of different ages and sex (Kiley-Worthington and de la Plain, Reference Kiley-Worthington and de la Plain1983). Today, dairy calves are usually separated from their mother shortly after birth and reared artificially, housed singly or in groups homogeneous with regard to age and sex. Systems allowing calves access not only to peers but also to their dam and other cows (cow-calf contact, CCC) offer a more complex social environment resembling the natural social structure better. The early social environment affects welfare, social behaviour and challenge response both in the short and long-term (Wagner et al., Reference Wagner, Barth, Palme, Futschik and Waiblinger2012, Reference Wagner, Barth, Hillmann, Palme, Futschik and Waiblinger2013, Reference Wagner, Seitner, Barth, Palme, Futschik and Waiblinger2015; Buchli et al., Reference Buchli, Raselli, Bruckmaier and Hillmann2017). However, play as a potential welfare indicator (Held and Spinka, Reference Held and Spinka2011) and social interactions of calves were not investigated in CCC systems so far. Thus, we observed these behaviours in calves reared artificially or with CCC at three times during the first three months of life. We hypothesized that calves with CCC would both show more play behaviour and encounter more social interactions.
Methods
The experiments were conducted in accordance with Good Scientific Practice guidelines and national legislation.
Animals, housing and management
The study was conducted at the Thünen-Institute of Organic Farming, Germany. Two dairy herds of different breeds (Black-and-White German Holstein, GH and German Red Pied, GRP) each with 45–50 cows were kept in two identical cubicle loose housing systems. One calf area per cow herd was placed adjacent to the cow barn. Due to the dynamic group management the total individual space allowance in the calf area ranged from 3.5–16.8 m2 per calf (mean ± sd: 6.4 ± 3.2 m2) depending on the actual group size.
Calves were randomly allocated to two rearing groups: Artificial (n = 20) and CCC (Contact, n = 19). Cows gave birth in individual calving pens where they stayed together with their calf for less than 1 d (Artificial) or for 5 d after calving (Contact). Afterwards, calves were moved to the respective calf pen, with deep-littered resting area and a running area, where they were kept until weaning. According to treatment calves had transponder-controlled access either to the cow barn and thus their dam and the cow herd via selections gates (Contact) or to the milk feeder (Artificial, milk amount increased from 4 up to 16 kg/d/animal). All calves had access to a transponder-controlled concentrate feeder, and to silage, hay and water ad libitum.
Behavioural observations
Calves were observed at an age of about 30 (25–36), 44 (39–49) and 84 (74–88) days, for 4 h each day (14 : 00 to 18 : 00). Observations were performed directly in the cow barn and from video recording in the calf area by continuous focal animal sampling. Duration of play behaviour (object play, locomotor play and social play) and frequency of agonistic social interactions (threats, head butts, avoiding, submission) and affiliative social interactions (social licking, licking solicitation) were recorded as well as initiator and receiver of an interaction.
Statistical analyses
The program R 2.13.2 (R Core Team, 2011) was used for statistical analysis. Play behaviour and affiliative interactions were analysed using generalized linear mixed models with the fixed effects of treatment, day of observation and their interaction, breed, sex and space allowance in the calf area, and calf identity as random effect. To fulfil model assumptions (checked graphically) data were square root transformed. Due to data loss of three Contact calves, sample size was reduced to n = 16. For comparing play according to location (cow barn/calf area) in Contact calves, data were calculated per hour to correct for the different duration of presence in the two locations. Agonistic social interactions were rarely observed and thus were dichotomized as ‘observed within the 3 d’, yes or no.
Results
Play behaviour
Contact calves showed longer duration of solitary play than Artificial calves (F = 4.561, P < 0.05, Fig. 1b) and this solitary play was performed mainly in the cow barn (F = 13.743, P < 0.001, Fig. 1c,d). Most of the solitary play was locomotor play (locomotor play in percent of solitary play: median, 25–75 quartile: Artificial, 100%, 60–100; Contact: 90%, 38–100). There was no difference between treatments regarding social play (F = 0.066, P > 0.05). The other variables had no effect on solitary or social play except that GRP showed longer duration of social play than GH (F = 4.533, P < 0.05).
Fig. 1. Duration of social play (a) or solitary play (b) per 4 h observation for artificially reared animals (Artificial n = 57 reduced number of data points due to a loss of observations in one of the 3 d in some animals) or animals with cow contact (Contact n = 56). C and D depict the duration of solitary play for Contact animals only, according to location cow barn or calf area either as duration per 4 h observation (c, n = 53) or as duration per 1 h presence in the respective location (d, n = 45, reduced number of data points as some animals did not stay in the calf pen at one or several days).
Social interactions
Contact calves initiated agonistic interactions more often (Z = 3.473, P < 0.001 Fig. 2a) and were receivers of interactions more often (Z = 3.170, P < 0.01, Fig. 2b) than Artificial calves that were almost never involved in such interactions (in total: only one interaction as initiator and two as receiver). Most of the agonistic interactions (about 75%) received by Contact calves were initiated by cows other than the dam (Fig. 2c). The number of affiliative social interactions was not affected by treatment (Fig. 2d).
Fig. 2. Frequency of social interactions of artificially reared calves (Artificial) and calves with cow contact (Contact). Initiated (a) and received (b, c) agonistic interaction are shown per 12 h i.e. the total number observed per animal during the three observations days; affiliative social interactions (d) are shown per 4 h observation. Graph C depicts the number of agonistic interactions with Contact calves being receiver depending on whether the actor is a cow or another calf.
Discussion
The longer duration of solitary play in calves with CCC is likely caused by more available space. In addition to the calf area, Contact calves had free access to the cow barn, where they had much more space and long alleys at their disposal (more than 25 × 3 m). More space leads to higher level of play behaviour, and calves show more locomotor play (running) in long-shaped/elongated areas compared to quadratic ones with the same supply of space (Jensen et al., Reference Jensen, Vestergaard and Krohn1998; Mintline et al., Reference Mintline, Wood, de Passillé, Rushen and Tucker2012). In line with this, we observed solitary play mostly in the cow barn. Although the average space allowance in the calf area was large in comparison to legal requirements or earlier studies (e.g. Jensen et al. Reference Jensen, Vestergaard and Krohn1998), it seems to have been insufficient to fulfil the Artificial calves' need for locomotion. Accordingly, the same Artificial calves showed more locomotor play as compared to the Contact calves in a test situation, which can be interpreted as a sign of deprived locomotor motivation (Jensen et al., Reference Jensen, Vestergaard, Krohn and Munksgaard1997; Wagner et al., Reference Wagner, Barth, Hillmann, Palme, Futschik and Waiblinger2013). Reduced play indicates impaired welfare of calves due to factors such as social deprivation or pain (Jensen et al., Reference Jensen, Vestergaard, Krohn and Munksgaard1997, Reference Jensen, Vestergaard and Krohn1998; Mintline et al., Reference Mintline, Stewart, Rogers, Cox, Verkerk, Stookey, Webster and Tucker2013) and play may elicit positive emotions (Held and Spinka, Reference Held and Spinka2011). More solitary play thus indicates higher welfare in Contact calves. Long-term beneficial effects of play on mental and physical development are also suggested (Held and Spinka, Reference Held and Spinka2011).
Treatments did not differ with respect to social play, which again is in line with Wagner et al. (Reference Wagner, Barth, Hillmann, Palme, Futschik and Waiblinger2013). Calves were available as play partners in both treatments, explaining the lack of a difference as calves primarily play with other calves (Kiley-Worthington and de la Plain, Reference Kiley-Worthington and de la Plain1983).
As expected, Contact calves experienced more agonistic interactions than Artificial due to their contact to cows. These early experiences of being in a subordinate position, may be important for development of social competence with both respect to learning of subtle social signals and accepting social rules. Evidence for higher social competence of animals reared with cow contact were found in calves confronted with an unfamiliar cow in a test pen (Buchli et al., Reference Buchli, Raselli, Bruckmaier and Hillmann2017), and in heifers when being integrated into the cow herd (Wagner et al., Reference Wagner, Barth, Palme, Futschik and Waiblinger2012), where Contact animals seem to follow strategies to avoid aggression (keeping more distance from cows, showing more submissive gestures). It would be interesting in future studies to investigate the ability of calves from different rearing experiences to recognize social signals and to accept social rules in more detail by trying to disentangle these two aspects, for instance by testing their ability to recognize different social signals in cognitive discrimination tests. The rare agonistic interactions in Artificial calves are in line with previous observations that interactions between calves are mostly non-agonistic up to an age of two months (Bouissou et al., Reference Bouissou, Boissy, Le Neindre, Veissier, Keeling and Gonyou2001). Because social licking is an important part of maternal behaviour (Bouissou et al., Reference Bouissou, Boissy, Le Neindre, Veissier, Keeling and Gonyou2001) more received affiliative behaviour in Contact calves was expected, but may have taken place at other times of the day.
In conclusion, and based on our findings, we suggest that the common space allowance in group housed calves still does not fulfil their need for locomotor activity. Enhanced level of social experience in a socially and spatially more complex environment may cause higher social competence later in life and potentially higher abilities to cope with social challenges in calves with cow contact.
Acknowledgements
The publication of this article was facilitated by COST Action FA1308 DairyCare, supported by COST (European Cooperation in Science and Technology, www.cost.eu). COST is a funding agency for research and innovation networks. COST Actions help connect research initiatives across Europe and enable scientists to grow their ideas by sharing them with their peers. This boosts their research, career and innovation. We acknowledge Marina Distl, Julia Johns, Kerstin Wilke, Kerstin Hofmann and Jaqueline Felix for help in data acquisition, and Thomas Dvorak, Michael Kossmeier and Andreas Futschik for statistical advice and calculations. This study was funded by the PhD Initiative Program BIOREC of the University of Veterinary Medicine Vienna.
Open access
