Recipient Organization
UNIV OF PENNSYLVANIA
(N/A)
PHILADELPHIA,PA 19104
Performing Department
(N/A)
Non Technical Summary
The welfare of pregnant pigs has been a public concern in this country for over 20 years. However, the experience of a female pig reared under conventional husbandry practices is not well understood. This is due to the lack of appropriate assessment tools. The goal of this work is to use novel objective approaches to study what an individual female pig wants. We will examine her choices and motivations for different types of housing. Choice tests record an animal's decision when presented with options. It provides a simple way to assess the relative value an animal attaches to available resources. Motivation tests quantify the work animals will do to express a choice. It gives insight into how important the decision is to the animal. Innovative technologies will be used to speed up data collection. The study findings promise new insight into a sow's experience and her welfare under different housing conditions. Our project aims to provide stakeholders with robust, animal-based data to optimize sow housing practices. This work will ensure that welfare expectations are scientifically based and impactful on-farm.
Animal Health Component
50%
Research Effort Categories
Basic
50%
Applied
50%
Developmental
(N/A)
Goals / Objectives
The overarching goal of this research is to apply novel, non-invasive, automated and objective measures of sow welfare to help define optimal solutions for sow welfare and well-being. Our supporting objectives address the dearth of appropriate experimental tools to study the affective state of conventionally housed animals by applying new technologies to study the preference and motivation of sows related to their housing. Preference tests record an animal's decision when presented with a choice and provides a simple way to assess the relative value an animal attaches to available resources. Motivation tests quantify the work animals will do to express that choice and give insight into how important the decision is to the animal. Taken together our novel technologies promise to provide unprecedented insight into a sow's experience, and hence her welfare, as a contributing factor in optimizing her housing. The application of computer vision and a unique force measurement device to automate the capture and analysis of objective data allows for the non-invasive study of a large number of animals and experimental conditions. Thus, our project will provide stakeholders with robust, animal-based data to inform sow-housing practices and ensure that welfare expectations are scientifically substantiated and effectively implemented on farm.
Project Methods
The overarching goal of this research is to apply novel, non-invasive, automated and objective measures of sow welfare to help define optimal solution for sow welfare and well-being. Two specific objectives will be investigated.Objective 1 - To employ computer vision to study what factors impact a sow's housing preference for a crate versus a pen. Preference tests record an animal's decision when presented with a choice and provides a simple way to assess the relative value an animal attaches to available resources. Video imaging combined with deep learning algorithms can identify the location and posture of animals without human intervention and allow for the ready study of large numbers of animals under a variety of experimental conditions including intrinsic (parity, social status, stage of gestation) and extrinsic (group dynamics and confinement duration) factors.Objective 2 - To use a gestation stall modified to measure force and examine what factors influence the work an animal will do to exit conventional housing.The incorporation of a force transducer records pressure applied by the sow to the stall as a proxy of her motivation to change housing. Motivation tests are an important complement to preference tests as quantify the work animals will do to express a choice and give insight into how important that decision is to the animal.Force is an objective measure that easily is digitized and analyzed with appropriate software to allow increased experimental throughput to study the role of parity, stage of gestation, confinement duration and the animal's agency over housing type.