- Email: [email protected]
Laterality of lying behaviour in dairy cattle
Applied Animal Behaviour Science 120 (2009) 125–131
Contents lists available at ScienceDirect
Applied Animal Behaviour Science journal homepage: www.elsevier.com/locate/applanim
Laterality of lying behaviour in dairy cattle Cassandra Blaine Tucker a,*, Neil Ralph Cox b, Daniel Martin Weary c, Marek Sˇpinka d a
Department of Animal Science, University of California, Davis, USA Bioinformatics, Mathematics & Statistics, AgResearch Ltd., Hamilton, New Zealand c Animal Welfare Program, University of British Columbia, Vancouver, Canada d Department of Ethology, Institute of Animal Science, Prague, Czech Republic b
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 29 May 2009
Dairy cattle spend, on average, between 8 and 15 h/d lying down. Our objective was to describe the laterality of lying behaviour and assess several internal and external factors that may affect laterality. Internal factors included time spent and time since eating or lying before choosing to lie down again. External factors included the slope and the amount of bedding on the of the lying surface. The dependent variables were the proportion of time spent lying on left versus right sides (as affected by eating and by the slope of stalls) and the probability of switching sides between two consecutive lying bouts (as influenced by previous lying bouts and the amount of bedding). The proportion of time on the left and right sides matched the mixed pattern in the literature; some groups of cows (n = 35, nonlactating, freestall housed) spent more time (56 3.0%, P = 0.042) on their left side, while other groups (n = 151, housed either in a freestall barn or at pasture) showed no preference for lying on one side versus the other (50 1.2% on left side, P 0.308). Laterality while lying was not influenced by eating behaviour or by the slope of the lying surface. Overall, cattle switched sides in 64% of consecutive lying events, more than 50% expected by chance (SE = 0.8%, P < 0.001). Switching sides was influenced by previous lying behaviour: cattle were more likely to switch sides if the previous lying bout was either long or recent (1.5 0.33 and 1.1 0.21% change in probability of switch for every 10 min in the last lying bout and for every 10 min since previous lying event, respectively, P < 0.001). Cows were more likely to switch sides when housed on mattresses with more bedding (switched sides in 68, 77, and 97 10.8% lying events for 0, 1 and 7.5 kg of sawdust bedding, mean SE, P = 0.042), possibly because cows had more lying events with shorter intervals between these events when the stall surface was well bedded. The probability of terminating a lying bout was also influenced by the duration of the bout. For example, during the first 10 min of a lying bout, the probability of standing up was only 5% but climbed to approximately 25% when the bout lasted 80 min or more. In conclusion, overall laterality in lying behaviour is shifted to the left in some groups but not others. Eating behaviour has little effect on time spent lying on either side. Cows switched sides between consecutive lying bouts and switching was more likely if the previous bout was either recent or long. Finally, continuous lying may become uncomfortable when bouts are longer than 80 min, and cows may switch sides to alleviate this discomfort. ß 2009 Elsevier B.V. All rights reserved.
Keywords: Behaviour Bedding Cow comfort Eating Laterality Lying
1. Introduction * Corresponding author at: Department of Animal Science, University of California, Davis, 1 Shields Avenue, CA 95616, USA. Tel.: +1 530 754 5750. E-mail address: [email protected] (C.B. Tucker). 0168-1591/$ – see front matter ß 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.applanim.2009.05.010
The laterality of lying behaviour in cattle has been documented several times in the last century. Some studies have concluded that more time is spent lying on the left side, and others have reported no difference in
126
C.B. Tucker et al. / Applied Animal Behaviour Science 120 (2009) 125–131
Table 1 Summary of literature about laterality of lying behaviour in adult dairy cattle. Reference
Time on left side
Statistically different?
Based on 24-h observation?
Individual animals identified?
Arave and Walters (1980) Bao and Giller (1991) Bao and Giller (1991) Bryan and Taylor (1938) Ewbank (1966) Forsberg et al. (Study 1, 2008) Forsberg et al. (Study 2, 2008) Forsberg et al. (Study 3, 2008) Jackson (1905) Phillips et al. (2003) Uhrbrock (1969) Wagnon and Rollins (1972) Wagnon and Rollins (1972)
62–64% 59% 48% 48% 51% 50% 50% 43–61% 53–61% 55% 52% 50% 56%
Yes Yes (cows pregnant with twins) No (cows pregnant with singletons) No Not tested No No Yes, in 9th month of pregnancy Yes No No No (AM only) Yes (PM only)
Yes Yes Yes ? Yes No Yes No No No No No No
Yes Yes Yes Yes Yes Yes Yes Yes No Yes ? Yes Yes
laterality (Table 1). In some of the studies that report differences in laterality (e.g. Jackson, 1905; Wagnon and Rollins, 1972), individual cows were not identified or followed over a 24-h period; studies with these limitations may not provide a representative sample of lying events. In addition, only a small sub-sample of studies describe the pattern of laterality. For example, Wagnon and Rollins (1972) anecdotally report that cows often switch sides between lying events and Bao and Giller (1991) report that cows switched sides between 50 and 60% of lying events. In the studies that report differences in laterality while lying, both internal and external factors seem to play a role. Internal factors include rumen fill, comfort, rumination, and pregnancy. Wagnon and Rollins (1972) speculate that cows spend more time on the left side after eating (56% of time on left side in afternoon observations occurring after a meal) to balance the additional weight in the rumen, situated on the left side of the body. Unfortunately, these authors did not record time spent feeding or feed intake. Comfort may also influence the side that cows lie on. Ruminally cannulated dairy cattle spend less time on their left side (30%) compared to intact cows (53% on left side, Grant et al., 1990), possibly because of discomfort associated with the cannula. Pregnancy also influences laterality. Cows in later stages of pregnancy are more likely to lie on their left side (Arave and Walters, 1980; Forsberg et al., 2008) and cows carrying twins are more likely to lie on their left side, compared to cows pregnant with singletons (Bao and Giller, 1991). These differences may also be explained by discomfort: Bao and Giller suggest that the enlarged uterus pushes the rumen further to the left (1991). External factors also influence laterality of lying behaviour. For example, Arave and Walters (1980) suggest that cows are more likely to lie with their dorsal side uphill. Others have found that laterality differs within the barn for reasons that are not clear (more time on left side in north end of barn, Gwynn et al., 1993). Over a number of years, we collected information about the laterality of lying behaviour. Experiments were performed on four farms in two countries (Canada and New Zealand), generating a large dataset to explore patterns in laterality. Our objective was to use these datasets to describe the effect of housing condition (pasture vs. freestall) and stage of lactation (dry vs. mid- and late lactation) on the
laterality of lying behaviour of dairy cattle. We predicted that cows would spend more time on the left side after eating, particularly if the eating bout was long or recent. In addition, we aimed to describe and the pattern of switching sides between lying events and predicted that cows would be more likely to switch sides if the previous lying bout was long or recent. Our goal was to use patterns in lying bout structure to help understand why cows may switch sides between consecutive lying events. Finally, we predicted that cows would be more likely to switch sides when lying on an uncomfortable surface (e.g. less bedding). 2. Materials and methods 2.1. Time spent on each side 2.1.1. Cows and treatments Data were pooled from six experiments and information from 186 cows was broken into four groups based on stage of lactation and housing conditions: (1) pregnant, non-lactating cows kept in a freestall barn (n = 35 from Tucker and Weary, 2004; Tucker et al., 2005, 2006), (2) mid-lactation cows kept in freestall barn (n = 27 from Experiment 2 in Tucker et al., 2004a), (3) mid-lactation cows kept at pasture (n = 60 from Experiment 2 in Tucker et al., 2007), and (4) late-lactation and non-lactating cows kept at pasture (n = 64 from Tucker et al., 2009). We specifically tested the effect of freestall slope (3% slope in alleyways and base of freestalls) with cows in Group 1. Thirteen of these cows were housed in stalls with a 3% slope to the left and 22 cows were housed with stalls with a 3% slope to the right (total n = 35). 2.1.2. Measurements Lying behaviour was measured by continuous observation from video for Groups 1 and 2 and by live observation with 10-min instantaneous scan samples for Groups 3 and 4. At least two 24-h periods were used from each cow (range: 2–10 d of lying information per cow). A sub-set of these data (n = 136 cows; 12 cows from Group 1 and all of Groups 3 and 4) was used to examine if eating behaviour influenced the choice to lie on the right or left side. We considered all lying events in relation to eating and the first lying event after eating. Information
C.B. Tucker et al. / Applied Animal Behaviour Science 120 (2009) 125–131
from multiple days was summed (range 2–10 d of information per cow). The eating behaviour of 12 dry, pregnant cows housed in a freestall barn was continuously recorded from video tape (from Tucker et al., 2006). Eating was defined as the head in the feeder and this criterion was used to calculate duration of eating bouts. The eating behaviour of all cows kept at pasture (n = 124) was recorded with live observation using 10-min scan samples. Eating was defined as active ingestion of grass or the characteristic side-to-side movements of the head associated with grazing. In this case, an eating bout was defined as at least one scan of eating behaviour. 2.1.3. Statistical analysis The percentage of time spent lying on the left side was calculated on a daily basis and a single, average value calculated for each cow. A one-sample t-test (PROC UNIVARIATE in SAS, 1999) was used to compare the difference between percentage of time spent lying on the left side and the expected 50%. A total of five one-sample ttests were used: one for the overall dataset (n = 186) and one for each of the four groups. A two-sample t-test (PROC GLM in SAS, 1999) was used to assess the effect of the direction of the stall slope on the percentage time spent lying on the left side (using cows from Group 1, above, n = 35), with direction of the slope of the floor (1 df; left or right) tested against the error term (33 df). The effect of the duration of eating bouts and time since eating on the choice to lie on either the left or right side was assessed with a logistic regression in GenStat (Payne et al., 2005). We considered both the first lying event after eating and all lying events. For both of these categories, we examined choice to lie either left or right with either the duration of eating bouts or time since eating as a fitted term. The interaction between the cow term and either the duration of eating bouts or time spent eating were used as the error term (df = 135). 2.2. Switching between sides 2.2.1. Cows, treatments and measurements All cows kept in freestall barns (Groups 1 and 2, n = 62) were used to evaluate switching sides between consecutive lying events. Firstly, the lying behaviour of 51 cows (24 from Tucker et al., 2005, 2006 and 27 from Tucker et al., 2004a) were used to describe the overall frequency of switching between consecutive lying events and how this switching behaviour was influenced by both the duration of and time since the previous lying bout. A minimum of two 24-h periods were used per cow (range: 2–8 d). All lying behaviour was recorded continuously from video. Secondly, we examined if the amount of bedding on the lying surface influenced the amount of switching sides between lying events in 11 non-lactating, pregnant cows (from Tucker and Weary, 2004). Each stall was fitted with a Pasture Mat geotextile mattress (Promat Ltd., Seaforth, Ontario, Canada), and bedded with one of three levels of kiln-dried sawdust: 0, 1, and 7.5 kg. Bedding was removed and reapplied and pens cleaned twice each day during the morning and afternoon feedings (8:00 and 15:00) in order
127
to maintain the appropriate amount of bedding on the surface, as there was no bedding retainer. All stalls were identical except for bedding levels. Cows were allowed access to only a single stall at a time, each for a 3-d period (2 d were video recorded), and the order of access to each treatment was assigned randomly without replacement and balanced across cows. 2.2.2. Statistical analysis The effect of the duration of the previous lying event or time since the previous lying event on the likelihood of a switch was assessed using logistic regression in GenStat (Payne et al., 2005). The interaction between the cow term and either the duration of the previous lying event or time since the previous lying event were used as the error term (error df = 50). The effect of the amount of bedding was tested using the model described by Tucker and Weary (2004). The percentage of lying events in which cows switched sides from the previous lying event was modified with arc sine transformation and analyzed using a mixed model (SAS, 1999). This model included order of exposure to treatments as a random factor, and used a first order autoregressive covariance structure. The repeated statement included a term for order and the individual animals were treated as the subject. The linear and quadratic effects of treatment were tested with a contrast statement in this mixed model. As the treatments were not evenly spaced, the coefficients used in these contrast statements were generated (Proc IML, SAS, 1999). The quadratic effect was not significant and is not discussed further. The P-value reported in the results section is for the linear contrast. 3. Results 3.1. Time spent on each side Across all four groups, cows averaged (mean SE) 51 0.8% of their lying time on the left side; this value did not differ from the expected 50% (P = 0.251). Some individuals showed clear preferences for one side (Fig. 1). Inter-individual variability was most marked in the non-lactating, pregnant, barn-housed cows. When tested separately, three of the four groups showed no overall preference: cows kept at pasture spent 49 0.9 and 50 1.0% during mid-lactation and latelactation/early dry period of their lying time on the left side, respectively (P 0.308), and cows kept in a freestall barn at mid-lactation spent 51 1.7% of their lying time on the left side (P = 0.600). Only one group differed from the expected 50%: the non-lactating, pregnant cows housed in a freestall barn spent 56 3.0% lying on their left side (P = 0.042). The direction of the slope in the stalls did not explain the variation in laterality of lying behaviour in this group (stall sloped left: 57 0.5% time lying on left side, stall sloped to the right: 56 0.4% time lying on left side, mean SE, P = 0.850). 3.1.1. Effect of eating Cows (n = 136) showed no preference for the left or right side in their first lying events after eating (50% of occasions on left, out of total of 4752 choices, P = 0.646). The duration of the eating bout and time since eating also had no influence on this response (P = 0.453 and P = 0.261, respectively).
128
C.B. Tucker et al. / Applied Animal Behaviour Science 120 (2009) 125–131
Fig. 2. Number of lying events on left (closed squares) and right (open circles) sides in relationship to (A) length of eating bout and (B) time since eating. Cows (n = 136) were either housed in a freestall barn or kept out at pasture.
A similar result was obtained when every lying event was considered. Cows showed no preference for the left side (50% of occasions on left, total of 9683 choices, P = 0.719). The duration of the eating bout (Fig. 2A) and time since eating (Fig. 2B) had no influence on this response (P = 0.730 and P = 0.180, respectively). 3.2. Switching between sides
Fig. 1. Percentage difference in time cows spent lying on the left versus the right from the expected 50%. Each bar represents an individual cow. Bars on the right show a preference for lying on the right side and vice versa. Data are shown separately for groups of cows (1) that were in the dry period and housed in a freestall barn (n = 35), (2) mid-lactation and
Lying bouts lasted, on average, 72 42.6 min/bout (mean SD), and 53% of bouts were 72 min (Fig. 3A). There was an exponential decline in the number of lying bouts longer than 80 min, and this decline corresponds with a constant probability of ending a bout (Fig. 3B). For bouts shorter than 80 min, the probability of a cow terminating a lying event increased with the duration of the lying bout (Fig. 3B). For example, during the first 10 min of lying, the probability was only 5% but it climbed to about 25% per 10 min once the cow had been lying for 80 min or more (Fig. 3C). The probability levels off around 80 min and then becomes more variable after 200 min, likely because there were relatively few long lying bouts. housed in a freestall barn (n = 27), (3) mid-lactation and kept at pasture (n = 60) or (4) late-lactation and early dry period and kept at pasture (n = 64).
C.B. Tucker et al. / Applied Animal Behaviour Science 120 (2009) 125–131
129
Fig. 4. The probability of switching sides (left to right or right to left) between consecutive lying events in relation to the (A) length of the previous lying bout and (B) time since the previous lying bout. The dashed line represents the fitted regression line. Cows (n = 51) were housed in a freestall barn.
Fig. 3. Statistics on the duration of 3656 lying bouts from 51 dairy cows. Bout duration is plotted on the x-axis and each tick mark represents 10 min. (A) Distribution of percentage of bouts in relation to the duration of the lying bout. (B) Log survivor plot of the number of bouts longer than the duration displayed on x-axis. The straight line for duration >80 min indicates a constant probability of ending a bout, corresponding to the exponential decline in number of bouts in panel A. (C) The probability that a lying bout will be terminated within the given 10-min period.
Of the 3656 bouts (n = 51), cows switched sides between 64% of consecutive lying bouts, and this percentage was great than the 50% expected by chance (SE = 0.8%, P < 0.001). 3.2.1. Effect of previous lying bout Longer previous lying bouts increased (P < 0.001, Fig. 4A) the likelihood of switching sides (1.5% increase in probability of switch for every 10 min in previous lying bout, SE = 0.33%). The time since the previous lying bout also influenced switching sides: cows lying down very soon after a lying bout were more likely to switch. The probability of switching sides decreased by 1.1% for
every 10 min between lying bouts (SE = 0.21%, P < 0.001, Fig. 4B). Approximately 3 h between lying bouts was required for the influence of the previous bouts to disappear. 3.2.2. Effect of lying surface Cows were more likely to switch sides on well-bedded surfaces (P = 0.042; means SE: 68, 77, and 97 10.8% lying events for 0, 1, 7.5 kg, respectively). The likelihood of switching sides in relation to the duration of and time since the previous lying bout was similar between bedding treatments (time by treatment interaction, P = 0.633; duration by treatment interaction, P = 0.743). 4. Discussion In three of the four groups, cows spent equal time on left and right sides. One group of non-lactating pregnant cows housed in freestalls spent more time on their left side than expected by chance. Our findings are consistent with the pattern within the literature: some studies report that cows spend more time on the left side, while others report
130
C.B. Tucker et al. / Applied Animal Behaviour Science 120 (2009) 125–131
no difference (Table 1). In addition, previous work has demonstrated that stage of pregnancy can influence laterality in lying behaviour. Cows in later stages of pregnancy are more likely to lie on their left side (Arave and Walters, 1980; Forsberg et al., 2008) and cows carrying twins are more likely to lie on their left side, compared to cows pregnant with singletons (Bao and Giller, 1991). In the current study, the group of cows that spent the most time on their left sides were also the group closest to calving. Discomfort associated with pregnancy or the size of the calf may explain differences in laterality of lying behaviour. For example, Grant et al. (1990) found that cannulated cattle spend less time lying on the side with the cannula (30% on left side, 1990). In addition, Grubmu¨eller et al. (personal communication) have found that cows spend less time lying on one side when wearing a logger on the metatarsus (40% on side with logger, 60% on side without, compared to 50:50% in cows without a logger). Marked individual differences also support the idea that discomfort may play a role in laterality of lying behaviour. Other studies have also documented such individual differences (Bryan and Taylor, 1938; Ewbank, 1964; King, 1967; Kikkers et al., 2006; Forsberg et al., 2008). For example, Bryan and Taylor (1938) reported that 23% of cows spent more than 60% of time on the left side. In the current study, the individual differences were most apparent for the non-lactating cows kept in freestalls. Others have also found that laterality in lying behaviour differs between farms: cows on a pasture-based dairy in Brazil had a normal distribution of left and right laterality in lying behaviour, while dairy cattle in an intensive British dairy had much more marked individual differences in laterality while lying (both to right and left, Phillips et al., 2003). These results suggest that when cows are more restricted in their choice of lying behaviour, as in a freestall barn, individual laterality becomes more pronounced. This individual asymmetry in lying may also be related to other challenges such as the health status of the legs and claws. These parameters were not investigated in the current study but the link between lying asymmetry and other aspects of individual well being, such as health, is worth investigating. Cattle also show laterality and individual differences in other behaviours including the preferred side of milking parlour (e.g. Hopster et al., 1998; Paranhos da Costa and Broom, 2001) and choices in a y-maze (Hosoi et al., 1995). It is unclear if individual asymmetry in lying is related to other lateral asymmetries in behaviour and if so, what mechanisms unpin these relationships. Cows may spend more time on the left side after eating in order to balance a full rumen, which is situated on the left side of the body. To test this idea, we examined how time spent eating and the time since eating influenced laterality and found no differences. We examined both the first lying event after eating and all lying events. Cows were equally likely to lie on the left or right side, regardless of the amount of time spent or since eating. We, however, recorded time spent eating, not feed intake. These two measures do not always correspond (Nielsen, 1999), and time spent eating is not the best test of our prediction about rumen fill. Additional information about feed intake and rumen size
would be useful to better understand any possible relationship between rumen position and laterality. Cattle switched sides between 64% of consecutive lying events. These results are in agreement with the anecdotal report that cows often switch sides between lying events (Wagnon and Rollins, 1972) and Bao and Giller’s (1991) finding that cows switched sides between 50 and 60% of lying events. Switching sides between lying events was influenced by previous behaviour. We predicted and found that the probability of cows changing sides increased with the duration of the previous lying bout and decreased as the time since the last bout elapsed. One explanation of both trends is that switching sides may alleviate discomfort associated with prolonged lying in one position. The probability of changing sides was between 50 and 55% if the previous lying bout was less than 60 min in duration, but this figure increased to over 70% if the previous bout was 80 min or longer. Somewhat surprisingly, the active interval between two lying bouts needed to last around 3 h before the effect of the previous lying bout no longer influenced which side the cow would lie on next. These patterns show that cows avoid repeatedly lying on the same side for long periods. Lying bouts averaged 72 min, a value similar to other studies (reviewed in Tucker et al., 2004b). Interestingly, the probability of a cow standing up was much smaller during the first 60 min into the bout (5–12% per 10 min) than when she had been already lying for 80 min or longer (about 25% per 10 min). Taken together with the effect of duration of lying on switching sides, we speculate that continuous lying is comfortable for about an hour, but may become uncomfortable when a bout lasts more than 80 min. We had also predicted that cows would be more likely to switch sides when lying on less comfortable surfaces. In a previous experiment, we found that the amount of bedding on the lying surface affected cow comfort (Tucker and Weary, 2004). Cows preferred and spent more time lying on mattresses bedded with more sawdust. However, in disagreement with our prediction, cows were more likely to switch sides when housed on a well-bedded mattress. The increased switching may be explained by longer lying times and more lying bouts in this treatment, resulting in shorter intervals between bouts. We speculate that the duration of lying bouts and the interval between successive bouts have more effect on the decision to switch sides than does the comfort of the lying surface. Switching sides does not seem to be a promising indicator of freestall comfort, but differences in laterality in lying behaviour may be useful for assessing the welfare of dairy cattle. To date, the evidence for this idea is piecemeal. Differences in laterality are more marked late in gestation and in response to cannulation, perhaps because cows are less comfortable in these situations. In addition, there is evidence of individual differences in laterality, and these differences seem more marked in freestall housing than on pasture. 5. Conclusions Laterality in lying is shifted to the left in some groups but not others. Eating behaviour has little effect on overall time
C.B. Tucker et al. / Applied Animal Behaviour Science 120 (2009) 125–131
spent lying on either side. Cows switch sides on 64% of consecutive lying bouts and the likelihood of switching increases if the previous bout has been either recent or long. Cows are also much more likely to terminate long lying bouts and we speculate that continuous lying may become uncomfortable when bouts are longer than 80 min, indicating that cows may switch sides to alleviate this discomfort.
Acknowledgements We thank the faculty, staff and students at University of British Columbia’s Dairy Education and Research Centre, the UBC Animal Welfare Program and the staff at AgResearch and DairyNZ for their support and help managing the cows during these experiments. Data collection were funded as part of other experiments financially supported by the Dairy Farmers of Canada, the Natural Sciences and Engineering Research Council of Canada (NSERC), Dairy InSight, and the New Zealand Ministry of Agriculture and Forestry. Marek Sˇpinka was supported by the grant MZE0002701404 from the Czech Ministry of Agriculture. References Arave, C.W., Walters, J.L., 1980. Factors affecting lying behaviour and stall utilization of dairy cattle. Appl. Anim. Ethol. 6, 369–376. Bao, J., Giller, P.S., 1991. Observations on the changes in behavioural activities of dairy cows prior to and after parturition. Ir. Vet. J. 44, 4–6. Bryan, C.S., Taylor, G.E., 1938. The relation of certain physical factors to infection with streptococcic mastitis. N. Am. Vet. 19, 26–30. Ewbank, R., 1964. Observations on the suckling habits of twin lambs. Anim. Behav. 12, 34–37. Ewbank, R., 1966. A possible correlation, in one herd, between certain aspects of the lying behaviour of tied-up dairy cows and the distribution of subclinical mastitis among the quarters of their udders. Vet. Rec. 78, 299–303. Forsberg, A.-M., Pettersson, G., Ljungberg, T., Svennersten-Sjaunja, K., 2008. A brief note about cow lying behaviour—do cows choose left and right lying side equally? Appl. Anim. Behav. Sci. 114, 32–36. Grant, R.J., Colenbrander, V.F., Albright, J.L., 1990. Effect of particle size of forage and rumen cannulation upon chewing activity and laterality in dairy cows. J. Dairy Sci. 73, 3158–3164.
131
Gwynn, P.E.J., Wilkinson, R., Thomas, T.P.E.C.-E., Boon, C., 1993. Factors influencing dairy cows in their choice of modified prefabricated timber cubicles. Proc. Livest. Environ. IV, Coventry UK, pp. 306–313. Hopster, H., van der Werf, J.T.N., Blokhuis, H.J., 1998. Side preference of dairy cows in the milking parlour and its effects on behaviour and heart rate during milking. Appl. Anim. Behav. Sci. 55, 213–229. Hosoi, E., Rittenhouse, L.R., Swift, D.M., Richards, R.W., 1995. Foraging strategies of cattle in a Y-maze—influence of food availability. Appl. Anim. Behav. Sci. 43, 189–196. Jackson, J., 1905. Ambidexterity. Kegan Paul, Trench and Trubner & Co., London, United Kingdom. ´ zsva´ri, L., Van Eerdenburg, F., Bajcsy, A.C., Szenci, O., 2006. Kikkers, B.H., O The influence of laterality on mastitis incidence in dairy cattle— preliminary study. Acta Vet. Hung. 54, 161–171. King, J.O.L., 1967. Influence of side on which a cow lies on milk yield and composition. Br. Vet. J. 123, 358–362. Nielsen, B.L., 1999. On the interpretation of feeding behaviour measures and the use of feeding rate as an indicator of social constraint. Appl. Anim. Behav. Sci. 63, 79–91. Paranhos da Costa, M.J.R., Broom, D.M., 2001. Consistency of side choice in the milking parlour by Holstein-Friesian cows and its relationship with their reactivity and milk yield. Appl. Anim. Behav. Sci. 70, 177–186. Payne, R., Murray, D., Harding, S., Baird, D., Soutar, D., 2005. GenStat for WindowsTM, Version 8 edition. VSN International Ltd., Herts, UK. Phillips, C.J.C., Llewellyn, S., Claudia, A., 2003. Laterality in bovine behavior in an extensive partially suckled herd and an intensive dairy herd. J. Dairy Sci. 86, 3167–3173. SAS, 1999. User’s Guide: Statistics, Version 9.1 edition. SAS Inst., Inc., Cary, NC, USA. Tucker, C.B., Weary, D.M., 2004. Bedding on geotextile mattresses: how much is needed to improve cow comfort? J. Dairy Sci. 87, 2889–2895. Tucker, C.B., Weary, D.M., Fraser, D., 2004a. Free-stall dimensions: effects on preference and stall usage. J. Dairy Sci. 87, 1208–1216. Tucker, C.B., Weary, D.M., Fraser, D., 2005. Influence of neck-rail placement on free-stall preference, use, and cleanliness. J. Dairy Sci. 88, 2730–2737. Tucker, C.B., Lacy-Hulbert, S.J., Webster, J.R., 2009. Effect of milking frequency and feeding level before and after dry off on dairy cattle behavior and udder parameters. J. Dairy Sci. 92, 3194–3203. Tucker, C.B., Weary, D.M., Rushen, J., de Passille´, A.M., 2004b. Designing better environments for cattle to rest. In: Kennelly, J. (Ed.), Adv. Dairy Technol. Univ. Alberta, Red Deer, Alberta, pp. 39–53. Tucker, C.B., Dalley, D.E., Burke, J.K., Clark, D.A., 2007. Milking cows once daily influences behavior and udder firmness at peak and mid-lactation. J. Dairy Sci. 90, 1692–1703. Tucker, C.B., Weary, D.M., De Passille, A.M., Campbell, B., Rushen, J., 2006. Flooring in front of the feed bunk affects feeding behavior and use of freestalls by dairy cows. J. Dairy Sci. 89, 2065–2071. Uhrbrock, R.S., 1969. Bovine laterality. J. Clin. Psychol. 115, 77–79. Wagnon, K.A., Rollins, W.C., 1972. Bovine laterality. J. Anim. Sci. 35, 486–488.
Contents lists available at ScienceDirect
Applied Animal Behaviour Science journal homepage: www.elsevier.com/locate/applanim
Laterality of lying behaviour in dairy cattle Cassandra Blaine Tucker a,*, Neil Ralph Cox b, Daniel Martin Weary c, Marek Sˇpinka d a
Department of Animal Science, University of California, Davis, USA Bioinformatics, Mathematics & Statistics, AgResearch Ltd., Hamilton, New Zealand c Animal Welfare Program, University of British Columbia, Vancouver, Canada d Department of Ethology, Institute of Animal Science, Prague, Czech Republic b
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 29 May 2009
Dairy cattle spend, on average, between 8 and 15 h/d lying down. Our objective was to describe the laterality of lying behaviour and assess several internal and external factors that may affect laterality. Internal factors included time spent and time since eating or lying before choosing to lie down again. External factors included the slope and the amount of bedding on the of the lying surface. The dependent variables were the proportion of time spent lying on left versus right sides (as affected by eating and by the slope of stalls) and the probability of switching sides between two consecutive lying bouts (as influenced by previous lying bouts and the amount of bedding). The proportion of time on the left and right sides matched the mixed pattern in the literature; some groups of cows (n = 35, nonlactating, freestall housed) spent more time (56 3.0%, P = 0.042) on their left side, while other groups (n = 151, housed either in a freestall barn or at pasture) showed no preference for lying on one side versus the other (50 1.2% on left side, P 0.308). Laterality while lying was not influenced by eating behaviour or by the slope of the lying surface. Overall, cattle switched sides in 64% of consecutive lying events, more than 50% expected by chance (SE = 0.8%, P < 0.001). Switching sides was influenced by previous lying behaviour: cattle were more likely to switch sides if the previous lying bout was either long or recent (1.5 0.33 and 1.1 0.21% change in probability of switch for every 10 min in the last lying bout and for every 10 min since previous lying event, respectively, P < 0.001). Cows were more likely to switch sides when housed on mattresses with more bedding (switched sides in 68, 77, and 97 10.8% lying events for 0, 1 and 7.5 kg of sawdust bedding, mean SE, P = 0.042), possibly because cows had more lying events with shorter intervals between these events when the stall surface was well bedded. The probability of terminating a lying bout was also influenced by the duration of the bout. For example, during the first 10 min of a lying bout, the probability of standing up was only 5% but climbed to approximately 25% when the bout lasted 80 min or more. In conclusion, overall laterality in lying behaviour is shifted to the left in some groups but not others. Eating behaviour has little effect on time spent lying on either side. Cows switched sides between consecutive lying bouts and switching was more likely if the previous bout was either recent or long. Finally, continuous lying may become uncomfortable when bouts are longer than 80 min, and cows may switch sides to alleviate this discomfort. ß 2009 Elsevier B.V. All rights reserved.
Keywords: Behaviour Bedding Cow comfort Eating Laterality Lying
1. Introduction * Corresponding author at: Department of Animal Science, University of California, Davis, 1 Shields Avenue, CA 95616, USA. Tel.: +1 530 754 5750. E-mail address: [email protected] (C.B. Tucker). 0168-1591/$ – see front matter ß 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.applanim.2009.05.010
The laterality of lying behaviour in cattle has been documented several times in the last century. Some studies have concluded that more time is spent lying on the left side, and others have reported no difference in
126
C.B. Tucker et al. / Applied Animal Behaviour Science 120 (2009) 125–131
Table 1 Summary of literature about laterality of lying behaviour in adult dairy cattle. Reference
Time on left side
Statistically different?
Based on 24-h observation?
Individual animals identified?
Arave and Walters (1980) Bao and Giller (1991) Bao and Giller (1991) Bryan and Taylor (1938) Ewbank (1966) Forsberg et al. (Study 1, 2008) Forsberg et al. (Study 2, 2008) Forsberg et al. (Study 3, 2008) Jackson (1905) Phillips et al. (2003) Uhrbrock (1969) Wagnon and Rollins (1972) Wagnon and Rollins (1972)
62–64% 59% 48% 48% 51% 50% 50% 43–61% 53–61% 55% 52% 50% 56%
Yes Yes (cows pregnant with twins) No (cows pregnant with singletons) No Not tested No No Yes, in 9th month of pregnancy Yes No No No (AM only) Yes (PM only)
Yes Yes Yes ? Yes No Yes No No No No No No
Yes Yes Yes Yes Yes Yes Yes Yes No Yes ? Yes Yes
laterality (Table 1). In some of the studies that report differences in laterality (e.g. Jackson, 1905; Wagnon and Rollins, 1972), individual cows were not identified or followed over a 24-h period; studies with these limitations may not provide a representative sample of lying events. In addition, only a small sub-sample of studies describe the pattern of laterality. For example, Wagnon and Rollins (1972) anecdotally report that cows often switch sides between lying events and Bao and Giller (1991) report that cows switched sides between 50 and 60% of lying events. In the studies that report differences in laterality while lying, both internal and external factors seem to play a role. Internal factors include rumen fill, comfort, rumination, and pregnancy. Wagnon and Rollins (1972) speculate that cows spend more time on the left side after eating (56% of time on left side in afternoon observations occurring after a meal) to balance the additional weight in the rumen, situated on the left side of the body. Unfortunately, these authors did not record time spent feeding or feed intake. Comfort may also influence the side that cows lie on. Ruminally cannulated dairy cattle spend less time on their left side (30%) compared to intact cows (53% on left side, Grant et al., 1990), possibly because of discomfort associated with the cannula. Pregnancy also influences laterality. Cows in later stages of pregnancy are more likely to lie on their left side (Arave and Walters, 1980; Forsberg et al., 2008) and cows carrying twins are more likely to lie on their left side, compared to cows pregnant with singletons (Bao and Giller, 1991). These differences may also be explained by discomfort: Bao and Giller suggest that the enlarged uterus pushes the rumen further to the left (1991). External factors also influence laterality of lying behaviour. For example, Arave and Walters (1980) suggest that cows are more likely to lie with their dorsal side uphill. Others have found that laterality differs within the barn for reasons that are not clear (more time on left side in north end of barn, Gwynn et al., 1993). Over a number of years, we collected information about the laterality of lying behaviour. Experiments were performed on four farms in two countries (Canada and New Zealand), generating a large dataset to explore patterns in laterality. Our objective was to use these datasets to describe the effect of housing condition (pasture vs. freestall) and stage of lactation (dry vs. mid- and late lactation) on the
laterality of lying behaviour of dairy cattle. We predicted that cows would spend more time on the left side after eating, particularly if the eating bout was long or recent. In addition, we aimed to describe and the pattern of switching sides between lying events and predicted that cows would be more likely to switch sides if the previous lying bout was long or recent. Our goal was to use patterns in lying bout structure to help understand why cows may switch sides between consecutive lying events. Finally, we predicted that cows would be more likely to switch sides when lying on an uncomfortable surface (e.g. less bedding). 2. Materials and methods 2.1. Time spent on each side 2.1.1. Cows and treatments Data were pooled from six experiments and information from 186 cows was broken into four groups based on stage of lactation and housing conditions: (1) pregnant, non-lactating cows kept in a freestall barn (n = 35 from Tucker and Weary, 2004; Tucker et al., 2005, 2006), (2) mid-lactation cows kept in freestall barn (n = 27 from Experiment 2 in Tucker et al., 2004a), (3) mid-lactation cows kept at pasture (n = 60 from Experiment 2 in Tucker et al., 2007), and (4) late-lactation and non-lactating cows kept at pasture (n = 64 from Tucker et al., 2009). We specifically tested the effect of freestall slope (3% slope in alleyways and base of freestalls) with cows in Group 1. Thirteen of these cows were housed in stalls with a 3% slope to the left and 22 cows were housed with stalls with a 3% slope to the right (total n = 35). 2.1.2. Measurements Lying behaviour was measured by continuous observation from video for Groups 1 and 2 and by live observation with 10-min instantaneous scan samples for Groups 3 and 4. At least two 24-h periods were used from each cow (range: 2–10 d of lying information per cow). A sub-set of these data (n = 136 cows; 12 cows from Group 1 and all of Groups 3 and 4) was used to examine if eating behaviour influenced the choice to lie on the right or left side. We considered all lying events in relation to eating and the first lying event after eating. Information
C.B. Tucker et al. / Applied Animal Behaviour Science 120 (2009) 125–131
from multiple days was summed (range 2–10 d of information per cow). The eating behaviour of 12 dry, pregnant cows housed in a freestall barn was continuously recorded from video tape (from Tucker et al., 2006). Eating was defined as the head in the feeder and this criterion was used to calculate duration of eating bouts. The eating behaviour of all cows kept at pasture (n = 124) was recorded with live observation using 10-min scan samples. Eating was defined as active ingestion of grass or the characteristic side-to-side movements of the head associated with grazing. In this case, an eating bout was defined as at least one scan of eating behaviour. 2.1.3. Statistical analysis The percentage of time spent lying on the left side was calculated on a daily basis and a single, average value calculated for each cow. A one-sample t-test (PROC UNIVARIATE in SAS, 1999) was used to compare the difference between percentage of time spent lying on the left side and the expected 50%. A total of five one-sample ttests were used: one for the overall dataset (n = 186) and one for each of the four groups. A two-sample t-test (PROC GLM in SAS, 1999) was used to assess the effect of the direction of the stall slope on the percentage time spent lying on the left side (using cows from Group 1, above, n = 35), with direction of the slope of the floor (1 df; left or right) tested against the error term (33 df). The effect of the duration of eating bouts and time since eating on the choice to lie on either the left or right side was assessed with a logistic regression in GenStat (Payne et al., 2005). We considered both the first lying event after eating and all lying events. For both of these categories, we examined choice to lie either left or right with either the duration of eating bouts or time since eating as a fitted term. The interaction between the cow term and either the duration of eating bouts or time spent eating were used as the error term (df = 135). 2.2. Switching between sides 2.2.1. Cows, treatments and measurements All cows kept in freestall barns (Groups 1 and 2, n = 62) were used to evaluate switching sides between consecutive lying events. Firstly, the lying behaviour of 51 cows (24 from Tucker et al., 2005, 2006 and 27 from Tucker et al., 2004a) were used to describe the overall frequency of switching between consecutive lying events and how this switching behaviour was influenced by both the duration of and time since the previous lying bout. A minimum of two 24-h periods were used per cow (range: 2–8 d). All lying behaviour was recorded continuously from video. Secondly, we examined if the amount of bedding on the lying surface influenced the amount of switching sides between lying events in 11 non-lactating, pregnant cows (from Tucker and Weary, 2004). Each stall was fitted with a Pasture Mat geotextile mattress (Promat Ltd., Seaforth, Ontario, Canada), and bedded with one of three levels of kiln-dried sawdust: 0, 1, and 7.5 kg. Bedding was removed and reapplied and pens cleaned twice each day during the morning and afternoon feedings (8:00 and 15:00) in order
127
to maintain the appropriate amount of bedding on the surface, as there was no bedding retainer. All stalls were identical except for bedding levels. Cows were allowed access to only a single stall at a time, each for a 3-d period (2 d were video recorded), and the order of access to each treatment was assigned randomly without replacement and balanced across cows. 2.2.2. Statistical analysis The effect of the duration of the previous lying event or time since the previous lying event on the likelihood of a switch was assessed using logistic regression in GenStat (Payne et al., 2005). The interaction between the cow term and either the duration of the previous lying event or time since the previous lying event were used as the error term (error df = 50). The effect of the amount of bedding was tested using the model described by Tucker and Weary (2004). The percentage of lying events in which cows switched sides from the previous lying event was modified with arc sine transformation and analyzed using a mixed model (SAS, 1999). This model included order of exposure to treatments as a random factor, and used a first order autoregressive covariance structure. The repeated statement included a term for order and the individual animals were treated as the subject. The linear and quadratic effects of treatment were tested with a contrast statement in this mixed model. As the treatments were not evenly spaced, the coefficients used in these contrast statements were generated (Proc IML, SAS, 1999). The quadratic effect was not significant and is not discussed further. The P-value reported in the results section is for the linear contrast. 3. Results 3.1. Time spent on each side Across all four groups, cows averaged (mean SE) 51 0.8% of their lying time on the left side; this value did not differ from the expected 50% (P = 0.251). Some individuals showed clear preferences for one side (Fig. 1). Inter-individual variability was most marked in the non-lactating, pregnant, barn-housed cows. When tested separately, three of the four groups showed no overall preference: cows kept at pasture spent 49 0.9 and 50 1.0% during mid-lactation and latelactation/early dry period of their lying time on the left side, respectively (P 0.308), and cows kept in a freestall barn at mid-lactation spent 51 1.7% of their lying time on the left side (P = 0.600). Only one group differed from the expected 50%: the non-lactating, pregnant cows housed in a freestall barn spent 56 3.0% lying on their left side (P = 0.042). The direction of the slope in the stalls did not explain the variation in laterality of lying behaviour in this group (stall sloped left: 57 0.5% time lying on left side, stall sloped to the right: 56 0.4% time lying on left side, mean SE, P = 0.850). 3.1.1. Effect of eating Cows (n = 136) showed no preference for the left or right side in their first lying events after eating (50% of occasions on left, out of total of 4752 choices, P = 0.646). The duration of the eating bout and time since eating also had no influence on this response (P = 0.453 and P = 0.261, respectively).
128
C.B. Tucker et al. / Applied Animal Behaviour Science 120 (2009) 125–131
Fig. 2. Number of lying events on left (closed squares) and right (open circles) sides in relationship to (A) length of eating bout and (B) time since eating. Cows (n = 136) were either housed in a freestall barn or kept out at pasture.
A similar result was obtained when every lying event was considered. Cows showed no preference for the left side (50% of occasions on left, total of 9683 choices, P = 0.719). The duration of the eating bout (Fig. 2A) and time since eating (Fig. 2B) had no influence on this response (P = 0.730 and P = 0.180, respectively). 3.2. Switching between sides
Fig. 1. Percentage difference in time cows spent lying on the left versus the right from the expected 50%. Each bar represents an individual cow. Bars on the right show a preference for lying on the right side and vice versa. Data are shown separately for groups of cows (1) that were in the dry period and housed in a freestall barn (n = 35), (2) mid-lactation and
Lying bouts lasted, on average, 72 42.6 min/bout (mean SD), and 53% of bouts were 72 min (Fig. 3A). There was an exponential decline in the number of lying bouts longer than 80 min, and this decline corresponds with a constant probability of ending a bout (Fig. 3B). For bouts shorter than 80 min, the probability of a cow terminating a lying event increased with the duration of the lying bout (Fig. 3B). For example, during the first 10 min of lying, the probability was only 5% but it climbed to about 25% per 10 min once the cow had been lying for 80 min or more (Fig. 3C). The probability levels off around 80 min and then becomes more variable after 200 min, likely because there were relatively few long lying bouts. housed in a freestall barn (n = 27), (3) mid-lactation and kept at pasture (n = 60) or (4) late-lactation and early dry period and kept at pasture (n = 64).
C.B. Tucker et al. / Applied Animal Behaviour Science 120 (2009) 125–131
129
Fig. 4. The probability of switching sides (left to right or right to left) between consecutive lying events in relation to the (A) length of the previous lying bout and (B) time since the previous lying bout. The dashed line represents the fitted regression line. Cows (n = 51) were housed in a freestall barn.
Fig. 3. Statistics on the duration of 3656 lying bouts from 51 dairy cows. Bout duration is plotted on the x-axis and each tick mark represents 10 min. (A) Distribution of percentage of bouts in relation to the duration of the lying bout. (B) Log survivor plot of the number of bouts longer than the duration displayed on x-axis. The straight line for duration >80 min indicates a constant probability of ending a bout, corresponding to the exponential decline in number of bouts in panel A. (C) The probability that a lying bout will be terminated within the given 10-min period.
Of the 3656 bouts (n = 51), cows switched sides between 64% of consecutive lying bouts, and this percentage was great than the 50% expected by chance (SE = 0.8%, P < 0.001). 3.2.1. Effect of previous lying bout Longer previous lying bouts increased (P < 0.001, Fig. 4A) the likelihood of switching sides (1.5% increase in probability of switch for every 10 min in previous lying bout, SE = 0.33%). The time since the previous lying bout also influenced switching sides: cows lying down very soon after a lying bout were more likely to switch. The probability of switching sides decreased by 1.1% for
every 10 min between lying bouts (SE = 0.21%, P < 0.001, Fig. 4B). Approximately 3 h between lying bouts was required for the influence of the previous bouts to disappear. 3.2.2. Effect of lying surface Cows were more likely to switch sides on well-bedded surfaces (P = 0.042; means SE: 68, 77, and 97 10.8% lying events for 0, 1, 7.5 kg, respectively). The likelihood of switching sides in relation to the duration of and time since the previous lying bout was similar between bedding treatments (time by treatment interaction, P = 0.633; duration by treatment interaction, P = 0.743). 4. Discussion In three of the four groups, cows spent equal time on left and right sides. One group of non-lactating pregnant cows housed in freestalls spent more time on their left side than expected by chance. Our findings are consistent with the pattern within the literature: some studies report that cows spend more time on the left side, while others report
130
C.B. Tucker et al. / Applied Animal Behaviour Science 120 (2009) 125–131
no difference (Table 1). In addition, previous work has demonstrated that stage of pregnancy can influence laterality in lying behaviour. Cows in later stages of pregnancy are more likely to lie on their left side (Arave and Walters, 1980; Forsberg et al., 2008) and cows carrying twins are more likely to lie on their left side, compared to cows pregnant with singletons (Bao and Giller, 1991). In the current study, the group of cows that spent the most time on their left sides were also the group closest to calving. Discomfort associated with pregnancy or the size of the calf may explain differences in laterality of lying behaviour. For example, Grant et al. (1990) found that cannulated cattle spend less time lying on the side with the cannula (30% on left side, 1990). In addition, Grubmu¨eller et al. (personal communication) have found that cows spend less time lying on one side when wearing a logger on the metatarsus (40% on side with logger, 60% on side without, compared to 50:50% in cows without a logger). Marked individual differences also support the idea that discomfort may play a role in laterality of lying behaviour. Other studies have also documented such individual differences (Bryan and Taylor, 1938; Ewbank, 1964; King, 1967; Kikkers et al., 2006; Forsberg et al., 2008). For example, Bryan and Taylor (1938) reported that 23% of cows spent more than 60% of time on the left side. In the current study, the individual differences were most apparent for the non-lactating cows kept in freestalls. Others have also found that laterality in lying behaviour differs between farms: cows on a pasture-based dairy in Brazil had a normal distribution of left and right laterality in lying behaviour, while dairy cattle in an intensive British dairy had much more marked individual differences in laterality while lying (both to right and left, Phillips et al., 2003). These results suggest that when cows are more restricted in their choice of lying behaviour, as in a freestall barn, individual laterality becomes more pronounced. This individual asymmetry in lying may also be related to other challenges such as the health status of the legs and claws. These parameters were not investigated in the current study but the link between lying asymmetry and other aspects of individual well being, such as health, is worth investigating. Cattle also show laterality and individual differences in other behaviours including the preferred side of milking parlour (e.g. Hopster et al., 1998; Paranhos da Costa and Broom, 2001) and choices in a y-maze (Hosoi et al., 1995). It is unclear if individual asymmetry in lying is related to other lateral asymmetries in behaviour and if so, what mechanisms unpin these relationships. Cows may spend more time on the left side after eating in order to balance a full rumen, which is situated on the left side of the body. To test this idea, we examined how time spent eating and the time since eating influenced laterality and found no differences. We examined both the first lying event after eating and all lying events. Cows were equally likely to lie on the left or right side, regardless of the amount of time spent or since eating. We, however, recorded time spent eating, not feed intake. These two measures do not always correspond (Nielsen, 1999), and time spent eating is not the best test of our prediction about rumen fill. Additional information about feed intake and rumen size
would be useful to better understand any possible relationship between rumen position and laterality. Cattle switched sides between 64% of consecutive lying events. These results are in agreement with the anecdotal report that cows often switch sides between lying events (Wagnon and Rollins, 1972) and Bao and Giller’s (1991) finding that cows switched sides between 50 and 60% of lying events. Switching sides between lying events was influenced by previous behaviour. We predicted and found that the probability of cows changing sides increased with the duration of the previous lying bout and decreased as the time since the last bout elapsed. One explanation of both trends is that switching sides may alleviate discomfort associated with prolonged lying in one position. The probability of changing sides was between 50 and 55% if the previous lying bout was less than 60 min in duration, but this figure increased to over 70% if the previous bout was 80 min or longer. Somewhat surprisingly, the active interval between two lying bouts needed to last around 3 h before the effect of the previous lying bout no longer influenced which side the cow would lie on next. These patterns show that cows avoid repeatedly lying on the same side for long periods. Lying bouts averaged 72 min, a value similar to other studies (reviewed in Tucker et al., 2004b). Interestingly, the probability of a cow standing up was much smaller during the first 60 min into the bout (5–12% per 10 min) than when she had been already lying for 80 min or longer (about 25% per 10 min). Taken together with the effect of duration of lying on switching sides, we speculate that continuous lying is comfortable for about an hour, but may become uncomfortable when a bout lasts more than 80 min. We had also predicted that cows would be more likely to switch sides when lying on less comfortable surfaces. In a previous experiment, we found that the amount of bedding on the lying surface affected cow comfort (Tucker and Weary, 2004). Cows preferred and spent more time lying on mattresses bedded with more sawdust. However, in disagreement with our prediction, cows were more likely to switch sides when housed on a well-bedded mattress. The increased switching may be explained by longer lying times and more lying bouts in this treatment, resulting in shorter intervals between bouts. We speculate that the duration of lying bouts and the interval between successive bouts have more effect on the decision to switch sides than does the comfort of the lying surface. Switching sides does not seem to be a promising indicator of freestall comfort, but differences in laterality in lying behaviour may be useful for assessing the welfare of dairy cattle. To date, the evidence for this idea is piecemeal. Differences in laterality are more marked late in gestation and in response to cannulation, perhaps because cows are less comfortable in these situations. In addition, there is evidence of individual differences in laterality, and these differences seem more marked in freestall housing than on pasture. 5. Conclusions Laterality in lying is shifted to the left in some groups but not others. Eating behaviour has little effect on overall time
C.B. Tucker et al. / Applied Animal Behaviour Science 120 (2009) 125–131
spent lying on either side. Cows switch sides on 64% of consecutive lying bouts and the likelihood of switching increases if the previous bout has been either recent or long. Cows are also much more likely to terminate long lying bouts and we speculate that continuous lying may become uncomfortable when bouts are longer than 80 min, indicating that cows may switch sides to alleviate this discomfort.
Acknowledgements We thank the faculty, staff and students at University of British Columbia’s Dairy Education and Research Centre, the UBC Animal Welfare Program and the staff at AgResearch and DairyNZ for their support and help managing the cows during these experiments. Data collection were funded as part of other experiments financially supported by the Dairy Farmers of Canada, the Natural Sciences and Engineering Research Council of Canada (NSERC), Dairy InSight, and the New Zealand Ministry of Agriculture and Forestry. Marek Sˇpinka was supported by the grant MZE0002701404 from the Czech Ministry of Agriculture. References Arave, C.W., Walters, J.L., 1980. Factors affecting lying behaviour and stall utilization of dairy cattle. Appl. Anim. Ethol. 6, 369–376. Bao, J., Giller, P.S., 1991. Observations on the changes in behavioural activities of dairy cows prior to and after parturition. Ir. Vet. J. 44, 4–6. Bryan, C.S., Taylor, G.E., 1938. The relation of certain physical factors to infection with streptococcic mastitis. N. Am. Vet. 19, 26–30. Ewbank, R., 1964. Observations on the suckling habits of twin lambs. Anim. Behav. 12, 34–37. Ewbank, R., 1966. A possible correlation, in one herd, between certain aspects of the lying behaviour of tied-up dairy cows and the distribution of subclinical mastitis among the quarters of their udders. Vet. Rec. 78, 299–303. Forsberg, A.-M., Pettersson, G., Ljungberg, T., Svennersten-Sjaunja, K., 2008. A brief note about cow lying behaviour—do cows choose left and right lying side equally? Appl. Anim. Behav. Sci. 114, 32–36. Grant, R.J., Colenbrander, V.F., Albright, J.L., 1990. Effect of particle size of forage and rumen cannulation upon chewing activity and laterality in dairy cows. J. Dairy Sci. 73, 3158–3164.
131
Gwynn, P.E.J., Wilkinson, R., Thomas, T.P.E.C.-E., Boon, C., 1993. Factors influencing dairy cows in their choice of modified prefabricated timber cubicles. Proc. Livest. Environ. IV, Coventry UK, pp. 306–313. Hopster, H., van der Werf, J.T.N., Blokhuis, H.J., 1998. Side preference of dairy cows in the milking parlour and its effects on behaviour and heart rate during milking. Appl. Anim. Behav. Sci. 55, 213–229. Hosoi, E., Rittenhouse, L.R., Swift, D.M., Richards, R.W., 1995. Foraging strategies of cattle in a Y-maze—influence of food availability. Appl. Anim. Behav. Sci. 43, 189–196. Jackson, J., 1905. Ambidexterity. Kegan Paul, Trench and Trubner & Co., London, United Kingdom. ´ zsva´ri, L., Van Eerdenburg, F., Bajcsy, A.C., Szenci, O., 2006. Kikkers, B.H., O The influence of laterality on mastitis incidence in dairy cattle— preliminary study. Acta Vet. Hung. 54, 161–171. King, J.O.L., 1967. Influence of side on which a cow lies on milk yield and composition. Br. Vet. J. 123, 358–362. Nielsen, B.L., 1999. On the interpretation of feeding behaviour measures and the use of feeding rate as an indicator of social constraint. Appl. Anim. Behav. Sci. 63, 79–91. Paranhos da Costa, M.J.R., Broom, D.M., 2001. Consistency of side choice in the milking parlour by Holstein-Friesian cows and its relationship with their reactivity and milk yield. Appl. Anim. Behav. Sci. 70, 177–186. Payne, R., Murray, D., Harding, S., Baird, D., Soutar, D., 2005. GenStat for WindowsTM, Version 8 edition. VSN International Ltd., Herts, UK. Phillips, C.J.C., Llewellyn, S., Claudia, A., 2003. Laterality in bovine behavior in an extensive partially suckled herd and an intensive dairy herd. J. Dairy Sci. 86, 3167–3173. SAS, 1999. User’s Guide: Statistics, Version 9.1 edition. SAS Inst., Inc., Cary, NC, USA. Tucker, C.B., Weary, D.M., 2004. Bedding on geotextile mattresses: how much is needed to improve cow comfort? J. Dairy Sci. 87, 2889–2895. Tucker, C.B., Weary, D.M., Fraser, D., 2004a. Free-stall dimensions: effects on preference and stall usage. J. Dairy Sci. 87, 1208–1216. Tucker, C.B., Weary, D.M., Fraser, D., 2005. Influence of neck-rail placement on free-stall preference, use, and cleanliness. J. Dairy Sci. 88, 2730–2737. Tucker, C.B., Lacy-Hulbert, S.J., Webster, J.R., 2009. Effect of milking frequency and feeding level before and after dry off on dairy cattle behavior and udder parameters. J. Dairy Sci. 92, 3194–3203. Tucker, C.B., Weary, D.M., Rushen, J., de Passille´, A.M., 2004b. Designing better environments for cattle to rest. In: Kennelly, J. (Ed.), Adv. Dairy Technol. Univ. Alberta, Red Deer, Alberta, pp. 39–53. Tucker, C.B., Dalley, D.E., Burke, J.K., Clark, D.A., 2007. Milking cows once daily influences behavior and udder firmness at peak and mid-lactation. J. Dairy Sci. 90, 1692–1703. Tucker, C.B., Weary, D.M., De Passille, A.M., Campbell, B., Rushen, J., 2006. Flooring in front of the feed bunk affects feeding behavior and use of freestalls by dairy cows. J. Dairy Sci. 89, 2065–2071. Uhrbrock, R.S., 1969. Bovine laterality. J. Clin. Psychol. 115, 77–79. Wagnon, K.A., Rollins, W.C., 1972. Bovine laterality. J. Anim. Sci. 35, 486–488.