Individual pig detection and tracking is an important requirement in many video-based pig monitoring applications. However, it still remains a challenging task in complex scenes, due to problems of light fluctuation, similar appearances of pigs, shape deformations, and occlusions. In order to tackle these problems, we propose a robust on-line multiple pig detection and tracking method which does not require manual marking or physical identification of the pigs and works under both daylight and infrared (nighttime) light conditions. Our method couples a CNN-based detector and a correlation filter-based tracker via a novel hierarchical data association algorithm. In our method, the detector gains the best accuracy/speed trade-off by using the features derived from multiple layers at different scales in a one-stage prediction network. We define a tag-box for each pig as the tracking target, from which features with a more local scope are extracted for learning, and the multiple object tracking is conducted in a key-points tracking manner using learned correlation filters. Under challenging conditions, the tracking failures are modelled based on the relations between responses of the detector and tracker, and the data association algorithm allows the detection hypotheses to be refined; meanwhile the drifted tracks can be corrected by probing the tracking failures followed by the re-initialization of tracking. As a result, the optimal tracklets can sequentially grow with on-line refined detections, and tracking fragments are correctly integrated into respective tracks while keeping the original identifications. Experiments with a dataset captured from a commercial farm show that our method can robustly detect and track multiple pigs under challenging conditions. The promising performance of the proposed method also demonstrates the feasibility of long-term individual pig tracking in a complex environment and thus promises commercial potential.

Obtaining the correct balance of nutrients requires that the brain integrates information about the body's nutritional state with sensory information from food to guide feeding behaviour. Learning is a mechanism that allows animals to identify cues associated with nutrients so that they can be located quickly when required. Feedback about nutritional state is essential for nutrient balancing and could influence learning. How specific this feedback is to individual nutrients has not often been examined. Here, we tested how the honeybee's nutritional state influenced the likelihood it would feed on and learn sucrose solutions containing single amino acids. Nutritional state was manipulated by pre-feeding bees with either 1M sucrose or 1M sucrose containing 100mM of isoleucine, proline, phenylalanine, or methionine 24h prior to olfactory conditioning of the proboscis extension response. We found that bees pre-fed sucrose solution consumed less of solutions containing amino acids and were also less likely to learn to associate amino acid solutions with odours. Unexpectedly, bees pre-fed solutions containing an amino acid were also less likely to learn to associate odours with sucrose the next day. Furthermore, they consumed more of and were more likely to learn when rewarded with an amino acid solution if they were pre-fed isoleucine and proline. Our data indicate that single amino acids at relatively high concentrations inhibit feeding on sucrose solutions containing them, and they can act as appetitive reinforcers during learning. Our data also suggest that select amino acids interact with mechanisms that signal nutritional sufficiency to reduce hunger. Based on these experiments, we predict that nutrient balancing for essential amino acids during learning requires integration of information about several amino acids experienced simultaneously.

One of the most important tasks of the brain is to learn and remember information associated with food. Studies in mice and Drosophila have shown that sugar rewards must be metabolisable to form lasting memories, but few other animals have been studied. Here, we trained adult, worker honeybees (Apis mellifera) in two olfactory tasks (massed and spaced conditioning) known to affect memory formation to test how the schedule of reinforcement and the nature of a sugar reward affected learning and memory. The antennae and mouthparts of honeybees were most sensitive to sucrose but glucose and fructose were equally phagostimulatory. Whether or not bees could learn the tasks depended on sugar identity and concentration. However, only bees rewarded with glucose or sucrose formed robust long-term memory. This was true for bees trained in both the massed and spaced conditioning tasks. Honeybees fed with glucose or fructose exhibited a surge in haemolymph sugar of greater than 120mM within 30s that remained elevated for as long as 20min after a single feeding event. For bees fed with sucrose, this change in haemolymph glucose and fructose occurred with a 30s delay. Our data showed that olfactory learning in honeybees was affected by sugar identity and concentration, but that olfactory memory was most strongly affected by sugar identity. Taken together, these data suggest that the neural mechanisms involved in memory formation sense rapid changes in haemolymph glucose that occur during and after conditioning.

Rhesus macaques are an important model in behavioral neuroscience due to their advanced cognitive abilities. To motivate animals to engage in complex tasks, fluid rewards, in conjunction with fluid control protocols, are often used. The impact of these protocols on animal welfare is controversial. We compared two fluid control protocols against a protocol providing free access to water and evaluated the impacts on physiological states of hydration, behavioral measures of welfare, and scientific output. Blood physiology did not significantly differ between any of the protocols, and urine measures were indicative of well functioning, healthy kidneys. Changes in behaviors were limited, the main one being an increase in motivation to drink on the stricter fluid control protocol, and improved task performance early in the week. Overall, fluid control protocols had little measurable impact on the welfare of rhesus macaques while ensuring that scientific data of high quality could be obtained.

For farmed species, good health and welfare is a win-win situation: both the animals and producers can benefit. In recent years, animal welfare scientists have embraced cognitive sciences to rise to the challenge of determining an animal's internal state in order to better understand its welfare needs and by extension, the needs of larger groups of animals. A wide range of cognitive tests have been developed that can be applied in farmed species to assess a range of cognitive traits. However, this has also presented challenges. Whilst it may be expected to see cognitive variation at the species level, differences in cognitive ability between and within individuals of the same species have frequently been noted but left largely unexplained. Not accounting for individual variation may result in misleading conclusions when the results are applied both at an individual level and at higher levels of scale. This has implications both for our fundamental understanding of an individual's welfare needs, but also more broadly for experimental design and the justification for sample sizes in studies using animals. We urgently need to address this issue. In this review, we will consider the latest developments on the causes of individual variation in cognitive outcomes, such as the choice of cognitive test, sex, breed, age, early life environment, rearing conditions, personality, diet, and the animal's microbiome. We discuss the impact of each of these factors specifically in relation to recent work in farmed species, and explore the future directions for cognitive research in this field, particularly in relation to experimental design and analytical techniques that allow individual variation to be accounted for appropriately.