Sleeping beauties

A time budget for young pigs

By Tawni Williams-Stroud, Kaitlyn Sommer, Ryan N. Dilger and Angela Green-MillerBehavior observations can provide valuable insight into an animal’s overall health and welfare status (Wilson et al., 2019; Matthews et al., 2017). The pre-wean period of development (less than 21 days of age in the United States) is difficult to manage and can have lasting impacts on swine development throughout the production stages (Tuchscherer et al., 2006; Muns et al., 2013; Sommer et al., 2022). Pre-wean pigs are exposed to many challenges. These challenges can include exposure to pathogens, changes within the thermal environment, painful procedures, nutritional demands, social encounters and more. A daily time budget can be a valuable tool to create a baseline for behavior expression and be used to determine changes in behavior when a young pig is exposed to a challenge (Williams-Stroud, 2023). Behavior observations can be used as a tool for animal caregivers to give timely and efficient interventions when necessary. This article outlines an ethogram- and spatial-based time budget to determine where and how young pigs are spending their time.Two approaches were utilized to create a daily time budget for young pigs housed in an individual research setting. The dataset included continuously collected video for 21 pigs over a 24-hour period within the Piglet Nutrition and Cognition Laboratory (PNCL) at the University of Illinois at Urbana-Champaign. The individual pens are paired and stacked, with two pigs neighboring one another separated by a clear plexiglass divider. The pigs can see, hear and smell one another but they do not have physical contact. Figure 1 demonstrates the individual research pen configuration at PNCL.

Figure 1. Individual pens for piglets housed at the Piglet Nutrition and Cognition Laboratory at the University of Illinois at Urbana-Champaign.

The first approach yielded an ethogram-based time budget, with behavior measured directly from the video for pigs at 6-days of age. A validated behavior ethogram was applied to the dataset, consisting of five main categories (inactive, consumption, exploratory, social and other) and 23 total behavior labels. The full dataset was labeled using the behavior ethogram and a validated sampling strategy, as detailed within Williams-Stroud, 2023. The resulting ethogram-based time budget revealed that the 6-day-old pigs spend much of their time at rest, accounting for 98.60% of their total time. Within this inactive category, 85.43% of time is spent in a lying position and 13.14% is spent standing, sitting, or kneeling at rest. The most performed behavior overall is lying socially at 38.48% of the daily time budget. Within this environment, lying socially is defined as: “The pig is lying (sternal or lateral) touching the plexiglass barrier. The pig in the adjacent pen is lying directly on the other side of the barrier, and the pigs would be touching each other if the barrier was not in place” (Williams-Stroud, 2023). The pigs are housed individually, but they are seeking the companionship of the neighboring pig by lying next to them across from the plexiglass barrier. Figure 2 demonstrates the percentage of time spent performing each posture within the overall categories and inactive behavior category.

Figure 2. A daily time budget, presented as a pie-chart, for the percentage of the observation period that each behavior category was observed for individually housed 6-day old pigs. The pie chart call-out displays the percentage of observation for each of the behavior labels within the behavior category “Resting.”

The second approach utilized computer vision to identify and summarize the location of the pig in the pen throughout a 10-day period, pigs from 3-days to 13-days-old. The pig’s location within their pen was tracked utilizing a model that identifies the pig’s center of mass or approximately the middle of the pig’s back. This corresponds with an (x,y) coordinate for the current position of the pig which is associated with a specific quadrant in the pen. Figure 3 demonstrates how each individual pen is broken into 4 quadrants and the time spent within each quadrant.

Figure 3. Individual pen broken down into the 4 quadrants utilized to assign piglets’ spatial time budget and the percentage of time spent within each quadrant. Quadrants 1 and 2 allow pigs to socialize through a clear, plexiglass divider. Quadrant 3 is where the animals typically defecate. Quadrant 4 houses the feeding bowl.

The resulting tracking-based time budget revealed that pigs from 3 to 13 days of age spend much of their time in quadrants 1 or 2, accounting for 87.5% of their total time. Within these two quadrants, pigs spent 38% within quadrant 1 and 49.5% within quadrant 2. These quadrants are located near the clear, perforated divider where pigs can hear, smell and see one another, and where pigs are able to lie socially. Overall, pigs spent the least amount of their time within quadrants 3 and 4, representing 3.6% and 8.8% of their time, respectively. Currently, there are few time budgets available within literature for young pigs, specifically for young pigs that are individually housed. The time budgets presented here may be used as a baseline for behavior comparison over time and during exposure to challenges within this specific laboratory environment. The observations that lying socially accounts for 38.48% of the daily time budget, and pigs spend 87.5% of their time within the quadrants located near the clear divider likely indicates the significance of social resting for young pigs. Further daily time budgets should be completed throughout the pre-wean growth period at various ages to determine if the prevalence of social resting and location within pen changes with time or interventions. As behavior analysis tools, such as behavior ethograms, time budgets, and computer vision models become more readily available, they can be utilized to better understand behavioral expression over time to closely monitor the health and welfare status of swine. Further application of a time budget for young pigs can include the use of a computer vision model for automated detection of animals or defined characteristics. The use of computer vision could increase the number of daily behavior observations and provide the ability to improve the animal welfare of young pigs in agricultural and biomedical research. References cited
Wilson, V., Guenther, A., Øverli, Ø., Seltmann, M. W., Altschul, D. 2019. Future Directions for Personality Research: Contributing New Insights to the Understanding of Animal Behavior. Animals, 9(5), 240.Matthews, S.G., Miller, A.L., PlÖtz, T., & Kyriazakis, I.. 2017. Automated tracking to measure behavioural changes in pigs for health and welfare monitoring. Scientific Reports (7).Muns, R., Manzanilla, E.G., Sol, X., Manteca, X., & Gasa, J. 2013. Piglet behavior as a measure of vitality and its influence on piglet survival and growth during lactation. Journal of Animal Science, 91(4), 1838-1843.Tuchscherer, M., Kanitz, E., Puppe, B., & Tuchscherer A. (2006). Early social isolation alters behavioral and physiological responses to endotoxin challenge in piglets. Hormones and Behavior, 50, 753-761.Sommer, K.M., Jespersen, J.C., Sutkus, L.T., Lee, Y., Donovan, S.M., & Dilger, R.N. 2022. Oral gamma-cyclodextrin-encapsulated tributyrin supplementation in young pigs with experimentally induced colitis. Journal of Animal Science, 100(11). Williams-Stroud, T. 2023. Toward automated digital technologies for piglet behavior analysis. [Master’s thesis, University of Illinois at Urbana-Champaign]. Illinois Digital Environment for Access to Learning and Scholarship, IDEALS.

Williams-Stroud is a Ph.D. student in swine welfare and behavior, Sommer is a research assistant, Dilger is a professor in animal sciences, and Green-Miller is an associate professor in agricultural and biological engineering, all with the University of Illinois at Urbana-Champaign.