Selecting feeding behavior would enhance feed efficiency in slow-growing free-range broiler

Background In the free-range poultry industry, commercial birds have access to outdoor environments where climatic conditions are variable and uncontrolled. However, breeding animals must be selected under confined and controlled conditions for sanitary reasons. These environmental differences can reduce selection efficiency due to genotype-by-environment (G×E) interactions, particularly affecting feed efficiency—a key trait for sustainability that impacts both economic performance and environmental footprint. Therefore, selection based solely on performance in confined conditions may not be optimal for free-range systems. To enhance sustainability in such systems, it is important to identify indoor traits genetically correlated with outdoor feed conversion ratios. This study investigates the potential for integrating feeding behavior traits into selection schemes for slow-growing broiler chickens raised outdoors during summer. Results We analyzed various feeding behaviors expressed indoors, including the number of meals, the timing and distribution of intake, and their variability both within and across days, as well as the correlation between these feeding behaviors and temperature. Many traits exhibited moderate to high heritability and were genetically correlated with outdoor feed conversion ratios. However, the correlations between traits were not always favorable: some behavioral traits were positively correlated with feed conversion ratio but they were negatively correlated with each other. To address these interactions, we summarized the G matrix of genetic variances and covariances among traits by its first two eigen vectors, g max and g 2 , accounting for 74% of the genetic variance in feeding behaviors. The g max axis included meal frequency, day-to-day variation in early-day intake, and meal number, while g 2 axis was mainly correlated with traits related to early-day feeding behaviors and their variation. Selection on the gmax axis could enhance feed efficiency and meat quality without compromising carcass yield, unlike selection on g 2 , which would improve feed efficiency at the expense of meat quality and carcass composition. Conclusions These findings suggest that incorporating feeding behavior traits, particularly those aligned with gmax, into indoor selection schemes could contribute to enhancing feed efficiency in outdoor environments.