25
Proceedings of the Benelux ISAE conference 2013 10 th October, Sterksel, The Netherlands Conference committee: Liesbeth Bolhuis Stephanie Buijs Merel Verhoeven
Proceedings of the Benelux ISAE conference 2013 10th October, Sterksel, The Netherlands
Conference committee:
Liesbeth Bolhuis
Stephanie Buijs
Merel Verhoeven
2
3
About the ISAE
The International Society for Applied Ethology (ISAE) is the world’s leading professional organisation for anyone interested in the behaviour and welfare of confined or domesticated animals, including companion, farm, zoo and managed wild animal species. Our members include the world’s most renowned authorities in the sciences of applied animal behaviour and animal welfare, many of whom serve as expert advisors to national and international committees. Every year the ISAE organises a major international meeting as well as several regional meetings, where members present and discuss their most recent scientific findings and their implications.
Members save on registration fees for meetings and subscription rates to Applied Animal Behaviour Science, the official, refereed journal of the ISAE. To stay informed about ISAE news, meetings, job opportunities and more, society members receive a biannual ISAE Newsletter. Members also have privileged access to the official ISAE website containing teaching resources and other useful material.
Membership is open to anyone with an academic interest in agriculture, veterinary medicine, zoology and other animal sciences. The annual fee is just £25 (£15 with proof of student or retired status). Fill in the online form to apply for membership (http://www.applied-ethology.org/members_signup.html). Each application should be supported by a proposer and a seconder (people who can confirm your professional interest in ethology, at least one of them a current ISAE member). For more information on membership, or if you need help with finding a seconder, please contact one of the secretaries of the Benelux region:
[email protected] [email protected]
ISAE Benelux
Today’s conference is the restart of regional ISAE activities in the Benelux. We are currently rethinking the type and frequency of such activities. A brainstorm session is included in today’s programme to get your opinion on what activities should be organized and how often. Of course, it is always possible to contact the regional secretaries on this topic afterwards.
4
Programme
9.30 Opening and introduction of the ISAE (Benelux)
Session 1: Development and validation of welfare measures Chair: Hans Spoolder
9.45 The ability to detect lameness in sows using a motivation test Bos EJ, Nalon E, Van Riet MMJ, Millet S, Janssens GPJ, Maes D, Tuyttens FAM
10.00 Validation and application of the voluntary water consumption test for the on-farm measurement of thirst in broiler chickens Tuyttens FAM, Vanderhasselt RF, Federici JF, Sans ECO, Molento CFM, Goethals K, Buijs S, Duchateau L
10.15 Assessing the EEG in pigs anaesthesized with propofol Verhoeven M, Gerritzen M, Peuscher J, Hellebrekers L, Kemp B
10.30 Challenges in standardization and generalization of research in animal welfare Nordquist RE, Arndt SS, Van der Staay FJ
10.45 Coffee break
Session 2: Cognition, emotion, personality and fear Chair: Bas Kemp
11.15 Time budget, social interactions, and fearfulness of laying hens characterized as fearful and non-fearful in groups with different compositions De Haas EN, Bolhuis JE, Henskens M, Grimberg C, Rodenburg TB
11.30 Differences in giving and receiving social support in pigs with different coping styles Reimert I, Bolhuis JE, Kemp B, Rodenburg TB
11.45 Individual differences affect response to a cognitive bias task in laying hens Rodenburg TB, Hernandez CE, Van der Kroef JW, Naguib M, Lee C
12.00 The effect of very low birth weight in piglets on cognitive holeboard performance Schoonderwoerd A, Antonides S, Nordquist RE, Van der Staay FJ
12.15 Effect of the Stableizer® on behaviour and heart rate in Thoroughbred horses Bonner L, Ödberg FO, Verschueren K, Moons CPH
12.30 Lunch
5
Session 3: Housing, management and social behaviour Chair: Frank Tuyttens
13.30 Lying behaviour of dairy cows in cubicles Van Erp - Van der Kooij LE, Mijdam ECJ, Almalik OO, Van Eerdenburg FJCM
13.45 Effects of dietary fibre on feeding patterns, behaviour and growth in pigs Souza da Silva CS, Bosch G, Bolhuis JE, Stappers LJN, Van Hees HMJ, Gerrits WJJ, Kemp B
14.00 Revolutionary equine group housing system with automatic roughage feeding system moving in between a group: the effect of increasing from 3 to 6 feeding runs Van Dierendock M, Vogel - van Vreeswijk T
14.15 Effect of environmental enrichment and territory on aggression in group-housed rabbit does Rommers JM, Reuvekamp B, Gunnink H, De Jong IC
14.30 Spatial integration in pigs after regrouping Camerlink I, Turner SP, Bolhuis JE
Brainstorm session 14.45 What can the Benelux ISAE do for you?
15.00 Coffee break
Session 4: Automation and modelling Chair: Marc Naguib
15.30 Automated measurement of pig behavioural activity at pen level using imaging processing techniques Ott S, Moons CPH, Kashiha M, Vandermeulen J, Bahr C, Berckmans D, Niewold T, Tuyttens FAM
15.45 Measuring intra-rater and inter-rates reliability using the spatial holeboard discrimination task for pigs Melendez D, Roelofs S, Nordquist RE, Van der Staay FJ
16.00 Modelling tail biting behaviour in pigs Boumans IJMM, Hofstede GJ, De Boer IJM, Bokkers EAM
16.15 Wintertime use of natural versus artificial shelter by cattle in nature reserves Van laer E, Moons CPH, Sonck B and Tuyttens FAM
16.30 Closing
16.35 Drinks & student presentation award announcement
17.30 Dinner
6
The ability to detect lameness in sows using a motivation test Bos, Emilie-Julie1,2, Nalon, Elena2, Van Riet, Miriam M.J.1,2, Millet, Sam1, Janssens, Geert P.J.2, Maes, Dominiek2 and Tuyttens, Frank A.M.1,2 1 Institute for Agricultural and Fisheries Research (ILVO), Animal Sciences Unit, Melle, Belgium 2 Ghent University, Faculty of Veterinary Medicine, Merelbeke, Belgium [email protected]
Even though group housing implies higher welfare standards than gestation crates, it sometimes impairs welfare of sows. Lameness is one of the main problems. Motivation tests are used to measure animals’ willingness to work for a reward, e.g. feed. Perhaps motivation tests could also be used to detect lameness if the type of work for the reward involves locomotion. The objective of this study was to investigate the impact of lameness on the ability to cover distances for food rewards. Twenty-nine hybrid sows were used. They were fed restrictedly (2.6 kg commercial gestation diet, daily). Sows were habituated and trained for 5 days in an experimental pen with two feeding locations. They were trained daily for 15 minutes, after scoring locomotion and classifying them as non-lame, mildly-lame or lame. Both a light and sound cue was used to indicate an available reward. Sows had to walk around a Y-shaped fence separating both feeders in order to obtain successive rewards. Training was considered successful when sows obtained at least four food rewards. After completion of training, sows were tested individually, once per day on three non-consecutive days. The effect of lameness on willingness to work for rewards was analysed using a Poisson model with gestation stage, parity and body weight as fixed effects. The willingness to cover distances for food rewards differed (P=0.006) between lame sows (mean=4.07, SEM=0.41) vs. non/mildly-lame sows, but not (P=0.479) between non-lame (mean=10.76, SEM=0.98) and mildly-lame sows (mean=11.75, SEM=1.65). These findings suggest that either the locomotion ability is not strongly reduced in sows with a mild degree of lameness, and/or that the present motivation test, although sufficiently sensitive to detect truly lame sows, may not be sensitive enough to detect mildly-lame sows. Sensitivity could possibly be improved by increasing the workload or reducing the attractiveness of reward.
7
Validation and application of the voluntary water consumption test for the on-farm assessment of thirst in broiler chickens Tuyttens, Frank A.M.1,2, Vanderhasselt, Roselien F.1, Federici, Juliana F.3, Sans, Elaine C.O.3, Molento, Carla F.M.3, Goethals, Klara2, Buijs, Stephanie1 and Duchateau, Luc2 1 Institute for Agricultural and Fisheries Research (ILVO), Animal Sciences Unit, Melle, Belgium 2 Ghent University, Faculty of Veterinary Medicine, Merelbeke, Belgium [email protected]
Animal welfare monitoring schemes use poorly validated resource-based measures, such as the number of animals per drinker, for assessing thirst. We performed a novel animal-based test of thirst on 10 Belgian and 10 Brazilian broiler farms to (1) test its validity and sensitivity; (2) identify confounding variables; and (3) compare with the Welfare Quality® resource-based assessment of thirst. Per flock 4 groups of 5 chickens from 4 locations (2 near walls, 2 central) were enclosed and the amount of water consumed from an unfamiliar open drinker during 90 min was measured. After this first test, the groups of chickens remained enclosed for 6h before the test was repeated. Half of the groups had access to water during this interval, the other groups had not. The mixed model indicated that water consumption in the first test was higher in Brazil than Belgium (P<0.001), but was not affected by location. The Welfare Quality® thirst scores were not correlated with the water consumption of Belgian (r=-0.10, P=0.55) or Brazilian (r=0.20, P=0.23) birds. In the 2nd test, the deprived chickens drank more than the control chickens, particularly in Brazil (country x treatment: P<0.001). Brazilian control birds consumed more water with decreasing relative humidity (P=0.020) and increasing temperature (P=0.025). ROC-curves showed that the test poorly detects a 6h-dehydration period in Belgium, but in Brazil 90% of the 6h dehydrated broilers would be detected if 15% false positives are allowed. We conclude that the voluntary water consumption test (1) does not correlate with the Welfare Quality® resource-based thirst score, (2) is sufficiently sensitive to discriminate between non-deprived and 6h deprived flocks in Brazil, but not in Belgium, and 3) ought to take into account indoor climatic variables and possibly genotype when interpreting the test outcome in terms of the thirst level experienced by broilers.
8
Assessing the EEG in pigs anaesthetised with propofol Verhoeven, Merel1,2, Gerritzen, Marien1, Peuscher, Jan3, Hellebrekers, Ludo4 and Kemp, Bas2
1Wageningen UR Livestock Research, Lelystad, The Netherlands 2 Wageningen University, Adaptation Physiology Group , Wageningen, The Netherlands 3Twente Medical Systems International, Enschede, The Netherlands 4Utrecht University, Faculty of Veterinary Medicine, Utrecht, The Netherlands [email protected]
The current study is part of a research project which aims at developing a tool in which brain activity can be assessed immediately after stunning at the slaughter plant to ascertain that an animal has been stunned correctly. The objective of this pilot study was to obtain good quality electroencephalogram (EEG) data and analyse the relationship between reflexes, response to a pain stimulus and derivatives from the EEG in pigs anaesthetised with propofol. Its results will be used in the search for an instant EEG parameter that can be used to assess unconsciousness. Two Topigs20 pigs (26.3 and 27.1 kg) were anaesthetised twice, with a 1 week interval, for 25 minutes. Anaesthesia was induced with an intravenous bolus of 100 mg propofol and thereafter maintained with an increasing propofol infusion rate of 100 to 800 mg hr-1. Cornea reflex, Palpebral reflex and the response to a pain stimulus, applied by pinching the ear, were tested every 30 s and classified as present, doubtful or absent. The EEG was recorded continuously and EEG parameters that are already analysed consisted of the Total Power (Ptot) and power in the different frequency bands. Only 3% of the raw data had to be deleted for further analyses, due to artefacts. Pigs lost consciousness 27 ± 4 s after induction with propofol, based on visual assessment of the EEG. Simultaneously, amplitudes of all EEG waves increased and the Ptot increased from 57 ± 7 to 105 ± 12 µV. Individual variation between both pigs made it difficult to assess general patterns in the EEG derivatives and responses to the reflexes and pain stimulus. In conclusion, good quality EEG was obtained, but further research should indicate what EEG derivative can represent (levels of) unconsciousness best.
9
Challenges in standardization and generalization of research in animal welfare Nordquist, Rebecca E.1, Arndt, Saskia S.2 and Van der Staay, Franz Josef1
1 Emotion & Cognition Group, Department of Farm Animal Health, Faculty of Veterinary Medicine, Utrecht University, The Netherlands 2 Emotion & Cognition Group, Department of Animals in Science and Society, Division of Animal Welfare & Laboratory Animal Science, Faculty of Veterinary Medicine, Utrecht University, The Netherlands [email protected]
Standardization of experimental testing is a manner to improve replicability of results, one of the pillars of empirical research. Within biomedical research, a heated debate has emerged on the use and necessity of standardization in experimental testing using animal subjects. This discussion must also take place within animal welfare research. The difficulties emerge when trying to reconcile standardization and generalization: the more highly specific the testing procedures are, the less one can generalize to more naturalistic situations, including on-farm situations. If the immediate goal of a study is to generalize to a larger population, the balance sways toward generalization. It is important to remain aware, however, of situations in which false negatives can be counterproductive, such as when the goal is to elucidate a physiological mechanism, or when expected effect sizes are small. In such cases, experiments based on standardization, generally conducted in a laboratory setting, may provide more useful information. From there, studies can be designed to test generalizability to larger populations, including on-farm studies. We pose that it is essential that the goal of the specific experiment conducted is clearly defined, and that the decision to balance between standardization and generalization must be made based on the specific needs to meet the intended goal. In this light, we discuss a multi-tiered approach to animal experimentation, in which standardization and generalizability are each given precedence during different phases of research, depending upon the goal of the experiment.
10
Time budget and fearfulness of laying hens characterized as fearful and non-fearful in groups with different compositions De Haas, Elske N.1, Bolhuis, J. Elizabeth1, Henskens, Marlijn2, Grimberg, Charlotte2 and Rodenburg, T. Bas2
1 Adaptation Physiology Group, Animal Science, Wageningen University, The Netherlands 2 Behavioural Ecology Group, Animal Science, Wageningen University, The Netherlands [email protected]
Birds with different levels of fearfulness may behave differently within a group. And the interactions of birds with different levels of fearfulness may influence others within the same group. Here we studied the differences between fearful and non-fearful laying hens in groups with different group compositions (0%, 25% and 50% fearful birds vs. non-fearful birds). We assessed the fear-response of 5-week old chicks in an Open Field, whereby we characterized them as fearful (by only freezing response) or non-fearful (by any locomotion or vocalizations). At 20 weeks of age we conducted a Tonic Immobility (TI) test, to assess their fearfulness. At 25 weeks a Novel Object (NO) test was conducted on group level to assess group fear level. Home pen observations were conducted once a week throughout the study. Fearful hens had a longer TI (F1,168=7.9, P=0.005; fearful: 136.1±14.3 s; non-fearful: 98.4±6.8 s) and needed fewer inductions to get into TI compared to non-fearful hens (F1,168=2.7, P=0.009; fearful: 1.10±0.04; non-fearful: 1.24±0.05). Groups with 50% fearful birds were more active (20%±0.1 vs. 12%±0.08; F2,26=5.3, P=0.02), more vigilant (2.5% vs. 1.9%: P=0.04), had higher initial fear for the NO (0 birds approached vs. > 1±2 birds that approached: F2,26=3.1, P=0.05) compared to the 0% group. Non-fearful hens did not become more fearful in the presence of fearful hens (Treatment*Fearfulness*TI-duration: F1,158=0.5,ns; Treatment*Fearfulness: F1,158=0.4, ns). This study shows that the proportion of fearful and hens in a group can affect the groups’ time budget and influence the groups’ fear level.
11
Differences in giving and receiving social support in pigs with different coping styles Reimert, Inonge1, Bolhuis, J. Elizabeth1, Kemp, Bas1 and Rodenburg, T. Bas2
1 Adaptation Physiology Group, Animal Science, Wageningen University, The Netherlands 2 Behavioural Ecology Group, Animal Science, Wageningen University, The Netherlands [email protected]
Studies have shown that pigs can benefit from social support, i.e. the ability of a partner to reduce the stress response of an individual during a stressful experience. In this study, we investigated whether giving and receiving social support is dependent on a pig’s coping style. Hereto, 72 pigs were subjected to a backtest at 10 days of age and classified as high-resister (HR) or low-resister (LR). At 11 weeks of age, 12 gilts of each coping style were subjected to a 15 min restraint test in a weighing cage without a pen mate present and 12 others gilts of each coping style were subjected to this test with a pen mate, a boar with the same coping style, present. Behaviours were scored from video using focal sampling and continuous recording. Salivary cortisol was measured at -30 min, just before, and 15, 30, 45 and 60 min after the start of the test. Data were analysed with mixed linear models. With the pen mate present, LR gilts showed less standing alert behaviour (5.3±1.0 vs. 12.1±3.4 % of time, P<0.05) and had their ears back less often (21.1±5.2 vs. 45.3±9.1 % of time, P<0.05) than when alone. The behaviour of the HR gilts was not affected by treatment, but they did vocalize more than LR gilts (376.9±29.3 vs. 281.7±30.0 times, P<0.01). Salivary cortisol levels were highest at 30 min (P<0.05), but were not affected by treatment or coping style. The behaviour and salivary cortisol levels of the boars were also not affected by coping style. Although few, the results of this study do suggest that pigs are able to benefit from social support, but this seemed to be dependent on the coping style of the pigs, because effects were only clear for the LR pigs.
12
Individual differences affect learning of a cognitive bias task in laying hens Rodenburg, T. Bas1, Hernandez, Carlos E.2,3, Van der Kroef, Jan-Willem1, Naguib, Marc1 and Lee, Caroline2 1 Behavioural Ecology Group, Wageningen University, The Netherlands 2 CSIRO Animal, Food and Health Sciences, FD McMaster Laboratory, Armidale, Australia 3 School of Environmental and Rural Sciences, University of New England, Armidale, Australia [email protected]
Animals differ in the way they perceive and respond to ambiguous cues in the environment. Animals that experience poor or stressful conditions tend to respond more pessimistically to ambiguous cues than animals that experience good conditions. This is known as cognitive bias, which could be used as a welfare indicator. It is not known, however, how individual (or personality) differences affect performance in a cognitive bias task. Therefore, the aim of this study was to investigate how individual differences affected laying hens’ performance in a cognitive bias task. Twenty White Leghorn laying hens were tested in an open-field test, a tonic immobility test and a manual restraint test between 18 and 20 weeks of age. Birds were then trained for 14 days for a cognitive bias task (two-choice test), where they had to move to the side of the test arena where the correct symbol and background colour were displayed in order to obtain a reward. Further, we tested cognitive bias by presenting ambiguous cues to test if these were interpreted as highly positive (optimistic) or less positive (pessimistic). Birds were tested for 10 subsequent test trials per day and rewards were balanced over the left and right side. We found that 8 of the 20 birds reached success rates of 70% and higher at the end of training. These individuals also tended to display a lower open-field activity (86 versus 52 % sitting; P<0.10) and a longer latency to vocalize during manual restraint (230 versus 140 s; P<0.10) than the individuals with lower success rates, indicating high levels of fearfulness. These results indicate that individual differences indeed can play an important role in cognitive bias training and testing. Hence these differences should be taken into account when using cognitive bias tests as a welfare indicator.
13
The effect of very low birth weight in piglets on cognitive holeboard performance Schoonderwoerd, Anne1,2, Antonides, Sanja2, Nordquist, Rebecca E.2 and Van der Staay, Franz Josef 2
1 Master Neuroscience & Cognition, Graduate School of Life Sciences, Utrecht University, The Netherlands
2 Emotion and Cognition Group, Department of Farm Animal Health, Faculty of Veterinary Medicine, Utrecht University, The Netherlands [email protected]
Low birth weight (LBW) is a serious public health problem, which can be caused by intrauterine growth restriction (IUGR). LBW in humans can lead to various cognitive deficits later in life. In the pig industry, the number of LBW piglets is increasing, due to the increased litter size as a result of selective breeding. In addition to posing a potential welfare problem for pigs on farms, LBW piglets may be used as model for IUGR in humans because of the similarities in brain structure and physiology. In previous studies, we have shown deficits in working memory in reversal in LBW piglets. To further investigate the effect of birth weight on cognition, we tested very LBW piglets and their normal birth weight (NBW) siblings in a spatial holeboard task. This is a square arena with a 4x4 matrix of holes in the floor, in which the piglets have to find 4 hidden rewards. Effects of birth weight category were analyzed using repeated measures ANOVA. Both groups acquired the task and improved their performance over time. However, vLBW piglets had a higher reference memory score (0.47 ± 0.03) than NBW piglets (0.39 ± 0.02) in the acquisition (P=0.004) and reversal phase (P=0.008). The vLBW group fell back less in working memory score (∆=0.13) than the NBW group (∆=0.27) starting the reversal (P<0.001). Whereas the faster learning of the reference component of the task supported earlier (albeit statistically non-significant) results we did not replicate the earlier effects on reversal learning. The criteria for selecting LBW piglets were stricter than in our earlier study, and the animals were tested at a younger age. In the present study, a higher motivation of the vLBW animals may be one of the factors responsible for the better cognitive performance of the LBW piglets.
14
Effect of the Stableizer on behaviour and heart rate in Thoroughbred horses Bonner, Lara1, Ödberg, Frank O.1, Verschueren, Katrien2 and Moons, Christel P.H1.
1 Faculty of Veterinary Medicine, Ghent University, Merelbeke, Belgium 2 Living Statistics, Gent, Belgium [email protected]
The Stableizer® is a device designed to induce sedation through peri-auricular and upper gum pressure. In this study, its effect on heart rate (BPM) and behaviour in 27 thoroughbreds (6 mares, 2 stallions, 19 geldings; mean age±SD: 4.5±1.53 years) was investigated. Responses to a sudden visual stimulus (10 s umbrella opening/closing) were also included. Successive conditions were (10 min each): No Stableizer® basal (NS-B), Stableizer® loose-basal (SL-B), Stableizer® loose-stimulus (SL-S), Stableizer® tightened-basal (ST-B), Stableizer® tightened-stimulus (ST-S). For half of the horses, the order of conditions SL-B/SL-S and ST-B/ST-S was reversed. PROC MIXED (SAS v9.3) was used to analyse heart rate (maximum, minimum, average) and behaviour, for the first two minutes and for the entire duration of a condition. Average heart rate during the first two minutes was lower for ST-S versus SL-S (lsmeans ± SEM: 66.97±7.77 versus 74.76±7.77 BPM; P=0.047). Similarly, maximal heart rate during 10 min was lower for ST-S versus SL-S (124.85±12.00 versus 143.16±12.00 BPM; P=0.0097). Minimal heart rate was never affected (NS). In the SL-S versus the ST-S condition, horses startled more frequently with displacement of body (0.67±0.18 versus 0.30±0.12; P=0.0346) and tended to startle more in general (8.04±0.53 versus 7.17±0.50; P=0.0830). There appeared to be an effect of the order in which the SL an ST conditions were applied, as the maximal heart rate was higher if the SL conditions occurred before the ST (74.39±4.33 versus 55.57±4.02 BPM, P=0.0466). Also, more behaviours indicative of excitation occurred following the visual stimulus in an SL period when this preceded an ST period. Conversely, more passive behaviour was seen in an SL period when this was preceded by an ST period. These findings could indicate an effect of the Stableizer® after it was loosened. Future research should include measurements of hormones and neurotransmitters to elucidate the potential sedating effects.
15
Lying behaviour of dairy cows in cubicles Van Erp - van der Kooij, Lenny E.1, Mijdam, Elco C.J., Almalik, Osama O.1 and Van Eerdenburg, Frank J.C.M.2
1 Department of Applied Biology, HAS Den Bosch University of Applied Sciences, The Netherlands 2 Department of Farm Animal Health, Utrecht University, The Netherlands [email protected] In dairy barns, cows are restricted in their lying behaviour. A better designed lying area may ensure better welfare and higher production levels. Standing and lying times are important indicators for cow welfare. In this study, we have recorded lying behaviour for 24 h on 12 farms using webcams. Data of the light period were used, because cows were not discernible in the dark. Lying behaviour was analysed using the Observer 10.5 (Noldus Information Technology) using continuous sampling, resulting in duration and frequencies of lying bouts, time used for lying down, for standing up, and time spent standing in the cubicle (‘standing idle’). Effects of bedding material and partition type on lying times, standing time and time used for standing up and lying down were analysed using a Multivariate ANOVA (MANOVA) model. Data were analysed from 127 cubicles on 12 dairy farms. Results of the video analyses showed that cows spent 6.09 ± 3.08 hours of daytime lying down. The average time ‘standing idle’ was 8.59 ± 6.35 minutes with an average duration of 1.57 ± 1.44 hours. Mean time needed for standing up was 8.31 ± 11.39 seconds, for lying down 6.98 ± 7.94 seconds. Soft bedding material (deep straw, deep sawdust or sand) significantly decreased the time needed for lying down (5.89 ± 2.17 s versus 8.67± 12.39 s, P=0.015), lying time (5.55± 3.21 h versus 7.01±3.14 h, P=0.001) and standing time (1.11± 1.08 h versus 2.27 ± 1.19 h, P=0.012) compared to harder bedding (rubber mat with sawdust). Rubber mats with Rbox partitions increased time needed for lying down (15.21 ± 21.66 s) compared to other combinations (range 4.63-6.89 s) (P=0.028). Shorter lying bouts were seen on rubber mats with Rbox partitions (51 ± 15 min) and on soft bedding with Combibox partitions (45± 7 min) than for other combinations (range 1.00-1.18 h) (P=0.025).
16
Effects of dietary fibre on feeding patterns, behaviour and growth in pigs Souza da Silva, Carol1,2, Bosch, Guido2, Bolhuis, J. Elizabeth1, Stappers, Lian J.N.1, Van Hees, Hubert M.J.3, Gerrits, Walter J.J.2 and Kemp, Bas1 1 Adaptation Physiology Group, Wageningen University, The Netherlands 2 Animal Nutrition Group, Wageningen University, The Netherlands 3 Nutreco Research and Development, Boxmeer, The Netherlands [email protected]
Fibres’ satiating properties may reduce hunger and improve welfare of restrictedly-fed pigs; however, reduced energy intake from fibre diets may reduce pigs’ growth performances. We assessed long-term effects of gelling fibre (alginate, ALG) and fermentable fibre (resistant starch, RS) on feeding patterns, behaviour and growth in ad libitum-fed growing pigs during 12 weeks. A control diet (ALG-RS-) containing 40% digestible starch was formulated, from which ALG+RS-, ALG-RS+ and ALG+RS+ diets were formulated by exchanging fibres for starch in a 2×2 factorial arrangement. In total 240 pigs (6 pigs/pen) were used. From all visits to an electronic feeder, cumulative feed intake (cFI) and feeding patterns were calculated. Pigs’ postures and behaviours were scored for 6h using 4-min scan sampling in week 12. Data were analysed with a mixed model including ALG (yes/no), RS (yes/no), their interaction and batch as fixed, and pen within (ALG RS batch) as random effects. ALG increased cFI (yes: 185±2 kg, no: 178±2 kg; P<0.05). RS increased meal size (yes: 148±5 g, no: 131±5 g; P<0.05), meal duration (yes: 6.7±0.2 min, no: 6.1±0.2 min; P<0.05) and inter-meal interval (yes: 89±3 min, no: 81±3 min; P=0.05), and decreased daily meals (yes: 15.7±0.5, no: 18.2±0.5; P<0.01). ALG alone increased standing and walking (28±2 vs. 22±2 %), aggressive (1.1±0.2 vs. 0.4±0.2 %), feeder-directed (11±1 vs. 7±1 %) and drinking (3.3±0.3 vs. 2.5±0.3 %) behaviours compared with the combination diet (ALG×RS interaction, P<0.05). Growth was similar between diets. In conclusion, ALG-fed pigs compensated for the reduced dietary energy content by increasing cFI, achieving similar growth as control-fed pigs, despite increased behavioural activity. RS-fed pigs changed feeding patterns, but did not increase cFI, achieving similar growth as control-fed pigs. These results show that the lower energy content of fibre diets and their influence on satiety do not necessarily reduce pigs’ growth performances.
17
Revolutionary equine group housing system with automatic roughage feeding system moving in between a group: the effect of increasing from 3 to 6 feeding runs Van Dierendonck, Machteld1,2 and Vogel – van Vreeswijk, Tessa2 1 Faculty of Veterinary Medicine, Ghent University, Merelbeke, Belgium 2 Faculty of Science, Dept. Animal Ecology, Utrecht University, The Netherlands [email protected]; [email protected]
Group housing of horses may lead to an increased risk of insufficient feeding and decreased resting time for lower-ranked horses and a higher chance on interaction-induced injuries. This study aimed to assess the welfare effects of increasing roughage-feeding frequency from 3 (3T) to 6 (6T) times per day in group-housed horses. In a revolutionary FreeMovingEquine Stable® a large automatic feeding system distributed roughage over large feeding troughs. Subjects were 47 riding school horses/ponies. According to a pre-planned balanced schedule, real time scoring by means of scan-sampling and all occurrence-sampling was performed: threat and aggression frequency (AF); time budget (TB); spatial relationships (SR) and avoidance behaviour for at least 8 days per trial. BodyConditionScores (BCS) were assessed before and after each trial. The same observations at the old location served as control. The AF was scored using video recordings before, during and after the feeding runs. Data were standardized; checked for normality; the appropriate statistical tests were performed depending on the type of data. Mean bite frequency (3T: 0.0090±0.00329 6T: 0.0029±0.00065 (n/horse/min±s.e.m.) and total aggression frequency (3T: 0.0928±0.01304 6T: 0.0658±0.00495 (n/horse/min±s.e.m.) before feeding decreased significantly in 6T compared with 3T (MWU: bite p=0.003; AF p=0.03). The AF during the day did not differ between trials and was relatively low and comparable with other studies. Horses spent more time resting (42 vs. 22%; t=-6.7, p<0.001) and less time feeding (60 vs. 39%; t=5.7, p=0.001) in 6T compared with 3T. Change in TB of low ranking horses was similar to the middle and high-ranking horses (ANOVA: p>0.05). Average BCS did not differ between ranking groups and remained the same during both trials (t-test: α=0.008: t=20.3, p=0.03). The SR and rank-orders did not change. Combined results indicated that increasing feeding frequency gave the horses more control over their environment and improved their welfare.
18
Effect of environmental enrichment and territory on aggression in group-housed rabbit does Rommers, Jorine M., Reuvekamp, Berry, Gunnink, Henk and De Jong, Ingrid C. Wageningen UR Livestock Research, Lelystad, The Netherlands [email protected]
Aggression between rabbit does can lead to severe injuries and impedes the application of group-housing on commercial farms. We studied the effect of environmental enrichment and territory on aggression in group-housed rabbit does. The group housing system consisted of four adjacent individual cages that were transformed into one group-housing system for four multiparous lactating hybrid does by taking away three side walls at d 18 of lactation. Until d 18 of lactation, does were housed individually in the cages of the group pen (territory (T)) or in other individual cages (no territory (NT)). Does were housed in groups until weaning of the kits at 35d of lactation. Within T and NT the following enrichments were provided (split-plot design): hiding places (HP), straw (S), hiding places and straw (HP+S) or no enrichment (NE)). Sixteen group pens were used for five subsequent rounds (n=10/treatment). Behavioural observations were performed and skin injuries and technical performance were registered. Treatments did not differ in frequency of agonistic interactions at 1 and 3 days after placement in the groups (average number of agonistic interactions for all treatments, day 1 148±24; day 3 51±27). Overall 52.2% of the animals had injuries (range 13 - 39%). Treatments with hiding places did have the lowest percentages of wounds and injuries numerically (% does with wounds: HP 13%; HP+S 27%; NE 31%; S 39%; average T 24%, NT 31%) but no significant treatment effects were found. There was a tendency for less does culled in HP and HP+S (0.25 and 0.3) compared with NE (0.8) and S (0.65) (P<0.10). To conclude, does do not defend a territory within a group pen and neither of the enrichments had positive effect on aggressive behaviour or injuries. The high level of aggression in all treatments merits further study.
19
Spatial integration in pigs after regrouping Camerlink, Irene1,2, Turner, Simon P.3 and Bolhuis, J. Elizabeth1 1 Adaptation Physiology Group, Wageningen University, The Netherlands 2 Animal Breeding and Genomics Centre, Wageningen University, The Netherlands 3 SRUC, Edinburgh, UK [email protected]
When unfamiliar animals meet, a high level of aggression can be observed, and the distance between animals may be informative of the level of social tension. In pigs, skin lesions may reflect aggressive interactions, but might not reflect social tension. The objective was to examine spatial integration between pigs upon encountering unfamiliar pigs during a regrouping test, and to relate the spacing patterns to skin lesions and weight. Pigs (n=385) of 8w of age were housed with six per group (3 males, 3 females), in conventional barren pens or in straw-enriched pens (both ~1m2/pig). Pairs of pigs (1 male, 1 female) were relocated to a pen with two other unfamiliar pig pairs. Distances between lying pigs were determined from video footage on 14 occasions over 24h. After 24h, skin lesions were counted and pigs were weighed. Spatial integration was determined as no difference in distance between familiar and unfamiliar pigs. Results showed that over the whole 24h, pigs kept more distance to unfamiliar pigs than to the familiar pig (P<0.001). Neither the distance between pigs, nor the time to reach spatial integration after first encounter was related to the amount of skin lesions (average 36±30 lesions) or to the presence of pigs with many skin lesions (80±29 lesions). Pigs with low body weight (26.6±2kg) stayed 0.4m closer to their familiar pig (P<0.001), needed ~90 minutes more to spatially integrate (P=0.02), and had 30% less skin lesions (P<0.01) than heavier pigs (34.5±3kg). Lying down in body contact occurred more often between familiar pigs (P<0.05). The results indicate that spatial integration is largely unrelated to skin lesions, but that social tension might especially be present in the smaller pigs that attempt to avoid fights. Spatial integration thereby may provide important information on subtle social patterns that go beyond aggression.
20
Automated measurement of pig behavioural activity at pen level using image processing techniques Ott, Sanne1,2, Moons, Christel P.H.1, Kashiha, Mohammadamin3, Vandermeulen, Joris3, Bahr, Claudia3, Berckmans, Daniel3, Niewold, Theo2 and Tuyttens, Frank A.M.1,4
1 Laboratory for Ethology, Faculty of Veterinary Sciences, Ghent University, Merelbeke, Belgium 2 Division of Livestock-Nutrition-Quality, Faculty of Bioscience Engineering, KU Leuven, Belgium 3 M3-BIORES - Measure, Model & Manage Bioresponses, KU Leuven, Belgium 4 Institute for Agricultural and Fisheries Research (ILVO), Animal Sciences Unit, Melle, Belgium [email protected]
Image processing techniques show great promise for the automated monitoring of animal behaviour. Such techniques could assist farmers in early detection of welfare and health problems in their livestock. Before applying such technology on-farm, however, the method ought to be well validated. The present study investigated for the first time the correspondence between automated measures of pig activity by image analysis and the behavioural activity as measured by a trained observer in a farm setting. Pigs were housed in 4 pens of 10 pigs and were video recorded by a top-view camera that covered an entire pen floor area. Pig behaviour was recorded during six days. On each of these days, four sessions of 30-min video recordings, two in the morning and two in the afternoon, were used for the comparison between the automated and human labelled data. A trained observer labelled pig activity in each session using 2-min instantaneous scan sampling. At each sampling point, each pig was scored as being behaviourally active when it was in locomotion (walking, running) and/or performing another activity (e.g. feeding, drinking, manipulating pen mate or pen fixtures, interacting socially). The behavioural activity scores of all individuals of a pen were averaged per session (6 days x 4 pens x 4 sessions = 96 data points). Automated pig activity was calculated by the relative number of moving pixels between two consecutive image frames (1 frame/second) and expressed as the average image activity index per session per pen. The automated activity measures were correlated to the human observations of pig behavioural activity. Automated activity measures were strongly correlated to the human observations (N=96, rs = 0.92, P<0.0001). This result seems promising for the use of automated activity measures as a cost-effective tool to measure pig behavioural activity at pen level.
21
Measuring intra-rater and inter-rater reliability using the spatial Holeboard discrimination task for pigs Melendez, Daniela, Roelofs, Sanne, Nordquist, Rebecca E. and Van der Staay, Franz Josef Emotion & Cognition Group, Department of Farm Animal Health, Faculty of Veterinary Medicine, Utrecht University, The Netherlands [email protected]
Automated tests can reduce the variation in results compared to those acquired from manual evaluation. Even so, the reliability of the programs used to score animal behaviour must be tested to guarantee that the results obtained by the computer programs are reliable. We have built an automated holeboard task for pigs, in which animals search an arena for rewards hidden in food bowls under a ball. We have automated this setup to measure when a ball is lifted using a magnet in the ball, a sensor in the bowl, and an interface (LabJack Data Acquisition Device, LabJack Corporation, USA) connected to a laptop running custom made software (Blinq Systems, Delft, The Netherlands). The holeboard task can yield numerous variables, spatial working memory (short-term memory, calculated as: number of rewarded visits divided through the number of visits to the baited set of holes) and reference memory (long-term memory calculated as: number of visits to the baited set of holes divided through the number of visits to all holes) being two of the most important. The purpose of this study was to measure the intra-rater reliability with repeated scoring by hand from a monitor (recorded using JWatcher Version 1.0) and the inter-rater reliability between scoring by hand and our custom automation. The result of the intra-rater reliability was a kappa coefficient of 0.8. The inter-rater reliability of working and reference memory had a Pearson correlations ranging from 0.928 to 0.995 (P-value <0.05). The intra- and inter-rater reliabilities obtained support the notion that automation of scoring visits to the bowls of the holeboard task yields reliable data.
22
Modelling tail biting behaviour in pigs Boumans, Iris J.M.M.1, Hofstede, Gert Jan2, De Boer, Imke J.M.1 and Bokkers, Eddie A.M.1
1 Animal Production Systems Group, Wageningen University, The Netherlands 2 Information Technology Group, Wageningen University, The Netherlands [email protected]
Tail biting in pigs is an abnormal behaviour frequently observed in pig husbandry. The cause of tail biting can be multi-factorial. Housing systems often do not provide enough stimuli to fulfil pigs’ needs to forage and explore, and also internal factors can cause tail biting, such as genetics or sex. The aim of this study was to investigate the suitability of an agent-based approach for modelling behaviour of pigs and understanding mechanisms underlying tail biting behaviour. Based on empirical data and hypotheses in literature, a spatially explicit agent-based computer model was constructed in Netlogo to simulate the behaviour of group housed growing pigs. Key behaviours of the agents (pigs) in the model were determined by internal states of agents and influenced by environmental stimuli. If the stress level of an agent exceeds a threshold, the agent will perform tail biting behaviour. The effects of enrichment and group density on internal states, behavioural time budget, the frequency of tail bites, and pigs changing into victims, biters, or both biter and victim of tail biting were studied. Results showed that at a high group density and in a barren environment the frequency of tail biting behaviour on group level could increase exponentially, which corresponds to effects shown in empirical pig behaviour studies. The model allows to further study behaviour of pigs in different environmental settings and the effect of farmers’ interventions. We, therefore, believe that agent-based modelling is a suitable technique to gain insight in emergent pig behaviour and to contribute to evaluation of novel housing systems that better meet pigs’ needs.
23
Wintertime use of natural versus artificial shelter by cattle in nature reserves Van laer, Eva1, Moons, Christel P.H.2, Sonck, Bart1 and Tuyttens, Frank A.M.1,2 1 Institute for Agricultural and Fisheries Research (ILVO), Animal Sciences Unit, Melle, Belgium 2 Faculty of Veterinary Sciences, Ghent University, Merelbeke, Belgium [email protected]
Cattle in nature reserves are exposed to aversive weather. Whether artificial shelter (AS) (man-made) should be provided in addition to natural shelter (NS) (vegetation) is a topic of debate. It’s also unknown which of both cattle prefer. In seven reserves with varying degrees of NS and one AS, we fitted a GPS-collar onto one cow to monitor terrain use during one (n=3) or two (n=4) winters. Weather stations registered local open-field climatic parameters that allowed calculation of the Wind Chill Index (WCI) and Comprehensive Climate Index (CCI). Animal positions were plotted onto digital maps of the reserves, to determine when different shelter types were used. Here we present exploratory analyses. In most reserves AS and its immediate surroundings were used little (mean±SE: 2.8±1.1% of recordings/winter). We found no clear diurnal pattern in use of different shelter types, though in most reserves open area was used less at night, when mostly NS was used. We explored - per reserve - whether air-temperature, black-globe-temperature, wind-speed, WCI or CCI differed when cows used open area and different shelter types. In most reserves no such differences were found. In one reserve, though, WCI differed between shelter types (P=0.043): WCI was slightly higher when the cow used NS (5.7±0.2) vs. 5m around it (5.6±0.2) or open area (5.6±0.2). In another reserve, black-globe-temperature differed between shelter types (P=0.028): it was slightly lower when the cow used AS (4.3±0.3°C) vs. <5m around AS (4.4±0.3°C), NS (4.4±0.3°C), <5m around NS (4.5±0.3°C) or open area (4.4±0.3°C). It’s hard to make generalizing statements on use of/preference for NS vs. AS in relation to climatic conditions based on these preliminary results. Cows’ sheltering behaviour is reserve-dependent and likely influenced by availability of NS. Further analyses will be conducted before advising legislators and reserve managers concerning wintertime protection for cattle kept outdoors.
24
Participants
Name Organization E-mail Fleur Bartels Wageningen University [email protected] Eddie Bokkers Wageningen University [email protected] Liesbeth Bolhuis Wageningen University [email protected]
Lara Bonner Ghent University [email protected]
Emilie-Julie Bos Institute for Agricultural and Fisheries Research [email protected]
Iris Boumans Wageningen University [email protected]
Petra Briene* [email protected] Stephanie Buijs Institute for Agricultural and Fisheries Research [email protected]
Elodie Bullens* Wageningen University [email protected]
Irene Camerlink Wageningen University [email protected]
Sophie de Graaf Institute for Agricultural and Fisheries Research [email protected]
Elske de Haas Wageningen University [email protected]
Ingrid de Jong Wageningen UR Livestock Research [email protected]
Jasper Heerkens Institute for Agricultural and Fisheries Research [email protected]
Leonie Jacobs Institute for Agricultural and Fisheries Research [email protected]
Manon Kestens [email protected]
Bas Kemp Wageningen University [email protected]
Daniela Melendez Suarez* Utrecht University [email protected]
Jenny Middelkoop* Wageningen University [email protected]
Elise Montfort FPW asbl (Filière Porcine Wallonne) [email protected]
Christel Moons Ghent University [email protected]
Marc Naguib Wageningen University [email protected]
Rebecca Nordquist Utrecht University [email protected]
Monique Ooms Wageningen University [email protected]
Sanne Ott Institute for Agricultural and Fisheries Research [email protected]
Marsha Reijgwart Utrecht University [email protected]
Inonge Reimert Wageningen University [email protected] Bas Rodenburg Wageningen University [email protected]
Sanne Roelofs Utrecht University [email protected]
Anne Schoonderwoerd Utrecht University [email protected] Anne-Marieke Smid* Wageningen University [email protected]
Carol Souza da Silva Wageningen University [email protected]
Hans Spoolder Wageningen UR Livestock Research [email protected]
Lisanne Stadig Institute for Agricultural and Fisheries Research [email protected]
Frank Tuyttens Institute for Agricultural and Fisheries Research [email protected]
Sofie van Amerongen Wageningen University [email protected]
Franz Josef van der Staay Utrecht University [email protected]
Yorike van der Weijden* Wageningen University [email protected]
Machteld van Dierendonck Ghent University [email protected]
Frank van Eerdenburg Utrecht University [email protected]
Lenny van Erp – van der Kooij HAS Hogeschool Den Bosch [email protected]
Eva Van laer Institute for Agricultural and Fisheries Research [email protected]
Renee van Toor* Utrecht University [email protected]
Wim Vanhove Hogeschool VIVES campus Roeselare [email protected]
Merel Verhoeven Wageningen UR Livestock Research [email protected]
Tessa Vogel - van Vreeswijk Utrecht University [email protected]
Elly Zeinstra Utrecht University [email protected] *UFAW sponsored attendants
25
Acknowledgement
We would like to acknowledge the financial support of:
International Society for Applied Ethology
Universities Federation for Animal Welfare
