AI Helps Researchers Understand Mating Behaviour and Fertility in Broiler Breeders

Fertility is influenced by many factors, but it ultimately depends on a series of brief behaviors that occur during mating. Researchers from the University of Tennessee recently used artificial intelligence and digital phenotyping to better understand these interactions in broiler breeders.

A Smarter Way to Study Breeder Mating

The study, partially funded by the Cobb Research Initiative, developed a digital phenotyping system based on the YOLOv8 deep learning model to automatically detect mating behavior in broiler breeders.

Pens containing Cobb 500 females and MX males were monitored continuously for approximately 4.5 months using side-view and overhead cameras. The goal was to automatically identify the mounting phase of mating and better understand how often it occurs, how long it lasts, and what factors influence successful mating behavior.

The deep learning model was highly accurate, achieving approximately 91% accuracy for detecting mounting behavior and 90% accuracy for identifying non-mating behavior.

After training, the system was applied to a large video dataset and identified more than 7,200 mounting events. This demonstrated that automated monitoring can reliably capture mating activity that would be extremely labor-intensive to observe manually.

Tail Movements Offer Clues to Fertility

The researchers reported that breeder matings are both brief and relatively infrequent, averaging about 13 mating attempts per male per day and lasting just over five seconds each.

One of the most interesting findings was the role of tail movements during mounting. About 77% of mountings included observable rooster tail movements, and these events lasted slightly longer than mountings without tail movement.

When both birds’ tails could be observed, approximately 68% of mountings involved synchronized tail movements by both the rooster and the hen, suggesting a greater likelihood of successful cloacal contact and semen transfer.

The study also revealed differences in mating motivation between males and females over time. During the first part of the breeding period, hens often showed more tail-movement activity than roosters, indicating strong receptivity.

However, this trend reversed later in the flock cycle, with hens showing reduced participation as they aged. The authors suggest this may help explain declining fertility in older flocks and supports current industry practices such as spiking to maintain reproductive performance.

Healthy Birds, Better Reproductive Outcomes

From an animal health perspective, one of the most important findings was the relationship between mobility and mating activity. Birds with poorer gait scores were less likely to participate in or sustain mating behavior.

Heavier roosters also showed reduced mating activity. These results reinforce the close connection between leg health, body weight management, and reproductive performance.

Breeders that can move comfortably and maintain good physical condition are better equipped for successful fertility.

The researchers also observed that mating activity was highest during the early morning and late afternoon hours and occurred more frequently in drier areas of the pen.

This suggests that environmental factors, including litter quality and bird comfort, may influence reproductive behavior. Maintaining dry litter not only supports footpad health and welfare but may also create conditions that encourage normal mating activity.

Practical Implications for Breeder Flocks

The main outcome of the study clearly shows that good animal health supports good reproductive performance.

Healthy legs, correct body weight, and good housing conditions all contribute to a breeder flock’s ability to complete matings. By quantifying mating behavior in greater detail, these technologies may help identify management factors that influence fertility before declines become evident in flock performance.