Dairy cows value access to pasture as highly as fresh feed


Source: von Keyserlingk, M. A. G. et al. Dairy cows value access to pasture as highly as fresh
feed. Sci. Rep. 7, 44953; doi: 10.1038/srep44953 (2017).
Many dairy cows in the developed world are now housed exclusively indoors with fewer than 5% of
the 10 million lactating cows in the United States having access to pasture during the grazing season.
Indoor housing systems are designed to meet biological needs for food, water, hygiene, and shelter,
but surveys of public and farmer opinion suggest that people think that pasture access is also important
for the well-being of dairy cows. To determine if pasture access is important to the cows themselves,
we investigated to what extent cows will work to access pasture (by pushing on a weighted gate), and
compared it to the motivation to access fresh feed. Cows worked at least as hard to access pasture as
they did to access the fresh feed and worked hardest for outdoor access in the evening hours. Echoing
public views on what allows for a good life for cattle, these results show that cows are highly motivated
for outdoor access.
Previous work has shown that dairy cows typically choose to remain indoors during the day, particularly when
temperature and humidity are high, but will spend most of their time on pasture at night if provided the
opportunit1. These results show a preference for outdoor access at night, but do not tell us the strength of this
preference. Some preferences may be trivial, making it difficult to draw strong inferences about the welfare
consequences of pasture access based on these results. Operant responses can be used to experimentally assess
motivation for access to a specific resource such as pasture. Once animals have learned to perform an operant task
to obtain access to a resource, the ‘work’ required for each access can be increased. Resources that are very
important to the animal show a relatively inelastic demand, a concept used frequently by economists and adapted by
Dawkins (1983) for the study of animal welfar2. Rewards are considered to have inelastic demand if animals are willing
to work to maintain a given level of the reward even when the costs (e.g. weight pushed, distance travelled, pain
endured) increase. Resources of less value, i.e. luxuries, have elastic demand: as the costs increase, animals stop
working to access the reward. Thus, the relative importance of a reward can be assessed with the elasticity of the
demand curve. The reservation price or maximum price paid, i.e., the highest effort that the animals complete, is
an additional way of estimating the value that the animal attaches to the resource 3.
By relying on decision making (the choice of working or not working for the resource), these measures of
motivation can distinguish between weak and fleeting preferences that may have little consequence for animal
welfare, and resources and experiences that the animal considers to be more valuable. For example, to investigate
feeding motivation, some authors measured the maximum distance walked to obtain food when cows were
exposed to different levels of feed deprivation, and found that cows were willing to walk longer distances to access
feed after longer periods of feed deprivation. Weighted doors (that animals push open to access a reward have
also been used to study motivation for access to specific resources 5, including access to a dust bathing are6
and a nest boby chickens, and access to a water bath by farmed mink 8. This experimental paradigm is based on the
concept that animals will be willing to push heavier weights to access more important resources.
The objective of this experiment was to assess cows’ motivation to access pasture and compare it to the
motivation to access fresh food immediately after milking, when cows are most motivated to eat 9
and thus a gold-standard for inelastic deman10. We hypothesized that cows would work to access pasture, but that the
motivation for pasture access would be less than that for fresh feed after milking. We also predicted that cows
would push more weight in the evening than during the day, consistent with the preference data showing stronger
preferences for outdoor access at night.
Animal Welfare Program, Faculty of Land and Food Systems, University of British Columbia, 2357 Main Mall,
Vancouver, BC, Canada.
Universidade Estadual de Londrina, Parana, CEP-86051-990, Brazil. Correspondence and
requests for materials should be addressed to M.A.G.v. K. (email: marina.vonkeyserlingk@ubc.ca)
Materials and Methods
We used 22 pregnant, late lactation Holstein cows averaging (mean ± SD) 221 ± 14.4 days in milk, producing on
average 30.0 ± 3.8 L of milk/day, 1.4 ± 0.7 parity, weighing 648 ± 93.3 kg, and body condition score of 3.0 ± 0.4
(scored from 1 to 5) 11. The study was approved by the University of British Columbias Animal Care Committee
(#A10-0162) and cared for according to the guidelines outlined by the Canadian Council of Animal Care 12.
Cows were tested in two phases. In Phase 1 we used two indoor pens fitted with lying stalls, a designated feed
ing area and fresh drinking water. Cows were milked twice daily at approximately 0630 and 1700 h. Cows were
offered fresh feed in the form of a total mixed ration (TMR) ad libitum, delivered daily at approximately 0630 h
and 1600 h. The TMR consisted of 21.8% corn silage, 18.6% grass silage, 10.3% alfalfa hay and 49.3% concentrate
mash, on a dry matter basis to ad libitum intake. The TMR was composed of 48.2% dry matter on average, with
18.5% crude protein, 31.6% neutral detergent fibre and 18.9% acid detergent fibre.
A weighted push-gate (see supplementary Video) was used to assess motivation to access fresh feed. We
trained cows twice daily to open the push-gate placed between two adjacent pens (one without feed) and the
other with fresh feed. To obtain access to the feed cows were required to open the push-gate in the adjacent pen.
Cows were considered trained when the gate was successfully pushed open in four consecutive training sessions.
Cows were allowed to access the gate starting 1.5 h after each milking. An additional 7 kg of weight was added
daily to the initial 7 kg training weight until cows did not perform the task for two consecutive sessions. Cows
were given 2.5 h to complete the task. The maximum weight pushed for access to the feed (i.e. maximum price
paid) was recorded for each individual cow. Behaviors were recorded using a surveillance camera (Panasonic,
CCTV, WV-BP334, Osaka, Japan) installed 3 m over the push gate.
The procedure for Phase 2 was identical except the push gate was positioned between the indoor pen and the
pasture. To ensure that cows were acclimated to the specific pasture used in this experiment, all cows were kept
on this pasture for 5 d before testing. The distance between the pasture and the barn entrance was 12 m. All cows
had access to the same fresh TMR described in Phase 1 inside the barn throughout the entire study. As in Phase 

1, the push gate was fitted with an intial weight of 7 kg and this weight was increased by 7 kg every 24 hours. The
test stopped when no cow within the group exited the gate for two consecutive days.
The majority of cows (59%) pushed just as hard or harder to access pasture as they did to access the TMR (bino
mial test P > 0.52). Modeling the number of cows who were willing to push at each weight with a survival anal
ysis allowed us to estimate a demand curve for the group, with steeper survival curves indicating more elastic
demand (less valuable resources). Survival analysis (log-rank: P > 0.42; Fig.1a) and reservation price (mean max
weight [SEM] in kg: pasture = 31.18 [3.82] and TMR = 36.91 [3.06]; paired t(18) = 1.58, P > 0.12) both failed
to detect a difference in motivation for the two resources. After the evening milking, cows pushed harder for
access to pasture than in the morning (binomial test P = 0.05), with survival analysis (log-rank test PM vs TMR:
P > 0.3; log-rank test AM vs TMR: P < 0.01; Fig.1b) and reservation price (mean max weight [SEM] in kg: after
noon = 29.27 [3.90] and morning = 23.55 [3.00]; paired t(18) = 2.08, P < 0.05) both showing higher motivation
for pasture in the evening.
Access to fresh feed was considered our gold standard given that cows are known to be highly motivated to access
fresh feed after milking 9; we expected that the weight pushed to access the feed would be close to the maximum
amount that cows were able to push. We had expected pasture to be valuable to the cows, but not as valuable as
access to the fresh feed, but instead found no difference in motivation to access the two resources.
Cows pushed harder to access pasture after the evening milking; this result is consistent with earlier results
showing that preference for access to pasture is stronger at nigh1,13, likely because solar heating can make cows
uncomfortably hot if they are outdoors during the heat of the da14. We expect that this greater motivation for
evening access would be less pronounced if cows were tested in the winter when radiant heating is less likely to
cause discomfort.
The results of the current study show that dairy cows are as motivated to access pasture as they are to eat fresh feed
two hours after milking. As cows had free access to fresh feed inside the barn when tested for motivation to access
pasture, our results suggest that motivation to access pasture was not driven by hunger, but rather motivation to
be outside (and to engage in behaviors associated with outdoor access, including grazing). Further research could
investigate the nature of this motivation by varying the outdoor conditions (e.g. providing grazing opportunities
vs. outdoor access only). Addressing public concerns over animal welfare will require better data on the animals
under our care—data that include not only the motivation for material outcomes such as food and shelter, but also
the motivation to engage in behavioral processes for their own sake.
We are grateful to The Emerging Leaders in the Americas Program (ELAP) for providing a scholarship for
Andressa Amorim Cestari and to the Killam Foundation for providing a postdoctoral fellowship to Becca Franks.
This research was funded by the Natural Sciences and Engineering Research Council of Canada (NSERC)
Industrial Research Chair program with industry contributions provided from the Dairy Farmers of Canada
(Ottawa, ON, Canada), British Columbia Dairy Association (Burnaby, BC Canada), Westgen Endowment Fund
(Milner, BC, Canada), Intervet Canada Corporation (Kirkland, QC, Canada), Zoetis (Kirkland, QC, Canada), BC
Cattle Industry Development Fund (Kamloops, BC, Canada), Alberta Milk (Edmonton, AB, Canada), Valacta (St.
Anne-de-Bellevue, QC, Canada), and CanWest DHI (Guelph, ON, Canada).
Author Contributions
Conceived and designed the experiment (M.A.G.v.K., D.M.W., A.A.C., J.A.F.). Performed the experiment (A.A.C.,
J.F.). Statistical analyses (B.F.). Contributed to reagents, materials and analytical tools (M.A.G.v.K., D.M.W.).
Contributed to the writing of the manuscript (M.A.G.v.K., B.F., D.M.W.).