Sonja Hillemacher

The increasing demand on adapting modern livestock farming to higher animal welfare standards requires a thorough understanding of a species’ cognitive abilities to determine their adaptability. With the chicken being the world’s most numerous birds in animal production, it is mandatory to identify its cognitive abilities and limitations in order to meet its needs. We investigated if chickens are able to use and understand the reflective properties of a mirror that is the correlation of reflections of food rewards and their real location. In total, 18 female chickens of two different breeds were tested in a mirror-mediated spatial location task. Eleven out of eighteen hens solved the task successfully and thus, possibly exploited the correlation between the reflection in the mirror and the real food reward. We found differences on a breed and on the individual level, with different amounts of time needed learning the association of reward and mirror image. The results imply sophisticated cognitive abilities in chickens, assuming they may be able to understand how mirror images represent objects in the real environment, and to make use of it during foraging. The chicken’s cognitive ability might lead to a new understanding and provision of animal welfare-compliant production environments.

Touching a mark on the own body when seeing this mark in a mirror is regarded as a correlate of self-awareness and seems confined to great apes and a few further species. However, this paradigm often produces false-negative results and possibly dichotomizes a gradual evolutionary transition of self-recognition. We hypothesized that this ability is more widespread if ecologically tested and developed such a procedure for a most unlikely candidate: chickens (Gallus gallus domesticus). Roosters warn conspecifics when seeing an aerial predator, but not when alone. Exploiting this natural behavior, we tested individual roosters alone, with another male, or with a mirror while a hawk’s silhouette flew above them. Roosters mainly emitted alarm calls in the presence of another individual but not when alone or seeing themselves in the mirror. In contrast, our birds failed the classic mirror test. Thus, chickens possibly recognize their reflection as their own, strikingly showing how much cognition is ecologically embedded.

Improving animal welfare is a prerequisite for the societal acceptance of poultry production. Support for improvements requires practical tools to quantify animal welfare and identify predispositions at the individual level, where possible. In this study, the activities and behavior of dual-purpose chickens (N = 245) and commercial breeders (N = 224) were analyzed. The general locomotor activity (GLA) data were collected using an RFID system over five days with 9-to-14-week-old animals. The results show that the animals of comparable age and stocking density differed from each other in their activity (p ≤ 0.001) according to breed, but no sex differences were observed (p = 0.159). No correlations were found between GLA and plumage condition (p > 0.05). The individual variations within the breeds are presented and discussed on an animal-by-animal level, providing new insights into the individual behavioral variability of chickens. The RFID systems can reliably generate GLA data that help to understand the potential interplay between behavior and animal welfare. The technology is also suitable for creating individual (personality) profiles that can be used for breeding. With a better understanding of the role of activity, husbandry and management practices can be adapted to improve animal welfare.

Fear and exploration are crucial traits determining how animals behave in novel situations, and thus, they influence animal welfare. The aim of this study was the characterization of these behavioral traits among different strains to identify interesting alternatives for future poultry production. Whereas the Novel Object Test (NOT) focuses on fear and exploration of novel objects, the Avoidance Distance Test (ADT) addresses this in the context of humans. Here, a commercial hybrid line, a dual-purpose hybrid and a local adapted strain were tested. For the differences between strains and development of fear, Lohmann Brown (n = 714), Lohmann Dual (n = 844) and Rhinelander (n = 458) were observed weekly until maturity. Results show that fear and exploration towards unknown objects and humans are breed-specific (all p < 0.01). Additionally, development of fear in NOT and ADT differed between all three strains (both p < 0.01). The expressions of fear of humans or objects should be regarded as characteristics adapted for different husbandry systems and breeding goals, e.g., high exploratory behavior in aviary or high avoidance of predators in free-ranging husbandry or at least a balanced ratio between fear and exploration. Characterization of behavioral traits among different strains, understanding diversity and integrating these behaviors into future breeding and husbandry systems might reflect the need to preserve local strains and the potential to improve animal welfare.

Chickens are the world’s most widely used farm animal and have a significant genetic diversity. In the current study, we investigated three strains for their suitability as dual-purpose chickens, with a focus on the fattening ability and welfare of the cockerels: 1. layer cockerels (Lohmann Brown, LB, n = 714); 2. cockerels of a dual-purpose hybrid (Lohmann Dual, LD, n = 844); and 3. cockerels of a native breed (Rhinelander, RL, n = 458). Chicks were raised under identical conditions and marked individually to compare focus and random sampling methods for weighing birds weekly. Because chicks of dual-purpose origins are usually raised mixed-sex, cockerels and pullets were weighed and observed together until sexes the were identifiable at week 10 of their life. During the 10th to 20th week of life, investigations were continued on 100 cockerels per genotype. Key figures for growth performance, such as feed conversion ratio (FCR) and European production efficiency factor (EPEF), were also calculated at weekly intervals. LD cockerels showed considerable growth performance (p < 0.001 compared to LB, RL, 2 kg at 9 weeks), whereas LB reached a live weight of 2 kg at 13 weeks and RL at 15 weeks of age. Genotype-dependent differences were also evident, with favorable FCR and EPEF for LD, intermediate for LB, and unfavorable for RL (all p < 0.001). The results of the FCR and EPEF suggest that cockerels should be slaughtered around week 8 of life, although only the carcass of the LD might be marketable. Thus, the optimal time of slaughter based on production parameters such as FCR and EPEF is different from the time when the animal reaches a marketable 2 kg live weight. Animal-based welfare indicators revealed that the RL are not adapted to production environments, including those that are extensive. Further research aimed at adapted feed management, including better FCR, and animals adapted to the respective production environments is necessary to improve alternative poultry production in the future.