Search Results (21)

Automated monitoring of welfare-relevant behavior in laying hens (Gallus gallus domesticus) has advanced with developments in computer vision, deep learning, and precision livestock farming (PLF). This narrative review integrates the ethological basis of welfare indicators with the development and readiness of monitoring technologies. It distinguishes routinely expressed diagnostic behaviors, including preening, locomotion, dustbathing, feeding, drinking, and nesting, from high-priority welfare risks such as aggression, piling, feather pecking, and inactivity or prostration. The review traces the progression from manual ethograms to semi-automated tools and recent artificial intelligence (AI) applications. These include You Only Look Once (YOLO)-based detection, multi-object tracking with BoT-SORT (a robust association-based tracking algorithm), pose estimation, and multimodal sensor fusion. Application-specific Technology Readiness Levels (TRL 1–9) indicate that most behavior-analysis systems remain at TRL 4–6. Several technologies extend into TRL 6–8, whereas few established systems reach TRL 9. Persistent barriers include domain shift under production conditions, annotation costs, inconsistent validation protocols, and limited economic accessibility. By linking welfare relevance, evaluation level, and deployment evidence, this review identifies priorities for scalable and actionable monitoring in commercial laying-hen production. Full article

Long-duration animal trajectories are central to computational ethology, yet constructing large rodent cohorts remains costly, time-intensive, and constrained by animal-use considerations. We present a sensor-aware generative framework that separates latent behavioral dynamics from sensing-induced observation distortion to synthesize observed-domain trajectories that are behaviorally plausible while reproducing proxy-referenced observation distortions. The framework combines a run-level semi-Markov ethology model, occupancy calibration, and state-conditioned kinematic generation with a regime-dependent Ultra-Wideband observation channel that explicitly captures Line-of-Sight and Non-Line-of-Sight sensing conditions. Using four UWB sessions, this proof-of-concept study models three states—exploring, feeding, and burrowing—and evaluates realism through state occupancy, state-conditioned kinematic divergence, residual-domain agreement, and mean-squared displacement across time lags. We further assess whether sensor-aware conditioning improves robustness under LoS/NLoS domain shift in downstream trajectory classification. Sensor-aware conditioning yields stable mixed-domain performance with AUC = 0.995, whereas condition-agnostic baselines decline to AUC = 0.974 and AUC = 0.901. These results support the feasibility of sensor-aware in silico ethology as a proof-of-concept framework for controlled robustness studies and algorithm evaluation under proxy-referenced observation distortion. Because the present evaluation is based on four UWB sessions and uses a smoothed UWB-derived reference trajectory rather than independent ground truth, broader applications to synthetic-cohort generation, disease modeling, and statistical power-analysis workflows should be considered future directions requiring validation in larger datasets. Full article

Automated recognition of equine–human interaction behaviors from video represents a significant challenge in computational ethology, with critical applications spanning animal welfare assessment, equine-assisted services evaluation, and safety monitoring in equestrian environments. Existing approaches to animal behavior recognition typically focus on single species in isolation, rely solely on facial expression analysis while ignoring full-body posture, or employ flat classification architectures that fail under the severe class imbalances characteristic of naturalistic behavioral datasets. Furthermore, no prior framework integrates simultaneous analysis of both human and equine body language for cross-species interaction classification. This paper presents a novel hierarchical classification framework integrating multi-modal computer vision features to distinguish behavioral states during horse–human encounters. Our methodology employs three complementary feature extraction pipelines: YOLOv8 for spatial relationship modeling, MediaPipe for human postural analysis, and AP-10K for equine body language interpretation. From 28 annotated interaction videos comprising 50,270 temporal samples across five horse breeds, we extract 35 discriminative features capturing proximity dynamics, body orientation, and species-specific behavioral indicators. To address severe class imbalance (18.3:1 ratio between affiliative and avoidant categories), we implement cost-sensitive gradient boosting with automatic class weight optimization within a two-stage hierarchical architecture. The first stage classifies interactions into three parent categories (affiliative, neutral, avoidant) achieving 73.2% balanced accuracy, while stage two discriminates six fine-grained sub-behaviors achieving 88.5% balanced accuracy (under oracle parent-category routing; cascaded end-to-end performance is 62.9% balanced accuracy due to Stage 1 error propagation, identifying parent classification as the primary bottleneck). Notably, our system achieves 85.0% recall on safety-critical avoidant behaviors despite their representation of only 3.8% of the dataset. Extensive ablation studies demonstrate that equine pose features contribute most critically to classification performance, while comprehensive cross-validation analysis confirms model robustness across diverse interaction contexts. The proposed framework establishes the first systematic multimodal cross-species behavioral assessment pipeline in human–animal interaction research, with direct implications for improving equine welfare monitoring and rider safety protocols. Full article

Artificial Intelligence (AI) is being increasingly used for equid health and welfare. This study aimed to establish consensus on where and how AI should be developed to achieve maximum benefit in this field. A workshop involving 41 stakeholders generated statements about current welfare concerns, areas for AI development, and barriers and solutions to AI use. Statements were circulated through Delphi surveys (acceptance set at 75% agreement). One-hundred-and-six statements reached agreement. Ethological needs not being met and poor equid management practices were key welfare concerns. Participants identified that insufficient owner/carer knowledge and understanding were important factors contributing to welfare concerns. Priority areas for AI development included assessment of equid wellbeing, as well as individual and population-level monitoring. Barriers included limited understanding of both equine behaviour and AI, biased, unethical, or insufficient data collection, difficulties developing accurate models, challenges to validation, and uncertainty around interpretation. Proposed solutions included development of evidence-based, unbiased AI systems, following best practice guidelines, requiring approval/regulation of AI tools, collaboration, and education of AI users. This is the first study to identify stakeholders’ opinions about where AI is likely to have the greatest benefit for equids, potential barriers, and solutions. The findings should be used to prioritise funding and development. Full article

Continuous digital monitoring of sheep behaviour shows potential for early stress detection. In Part 1 of this study, a novel accelerometer-based behaviour-recognition system using a nRF52832 microcontroller with Bluetooth wireless data transfer was developed and validated. A dedicated algorithm was developed to focus on the automatic detection of rumination, which also enables the classification of resting/idling and eating. The system achieved accuracies of 0.87 (rumination), 0.90 (resting/idling), and 0.86 (eating). Specificities were 0.87, 0.95, and 0.94; sensitivities 0.89, 0.80, and 0.60; and precisions 0.79, 0.88, and 0.73, respectively. In Part 2, four sheep were continuously monitored for 24 h to establish baseline behavioural durations. Animals were then relocated in pairs to an unfamiliar enclosure for a further 24 h observation period. Relocation resulted in a significant reduction in rumination time (−45.6%, p < 0.05) and a significant increase in resting/idling (+47.9%, p < 0.05), while time spent eating decreased but did not reach statistical significance (−36.2%). These findings indicate that detecting deviations from baseline rumination and resting/idling durations may serve as suitable ethological parameters for automated, sensor-based stress alerts. With further technical refinement and validation, the developed system shows strong potential as a reliable, non-invasive tool for monitoring key sheep stress indicators. Full article

Laboratory mice are the most widely used model organisms in biomedical and behavioral research, yet growing concerns regarding reproducibility and translational validity have highlighted the substantial influence of housing and husbandry conditions on experimental outcomes. Although domestication is often assumed to have rendered laboratory mice fully adapted to artificial environments, evidence from ethology indicates that many core behavioral and physiological needs remain conserved. As a result, standard laboratory housing may generate chronic stress, alter behavior, and introduce systematic bias into experimental data. This narrative review critically examines how ethological constraints persisting after domestication interact with key environmental factors, social housing, environmental enrichment, ambient temperature, and lighting regimes to shape welfare and experimental validity in laboratory mice. Rather than providing an exhaustive overview of mouse behavior, the review adopts a problem-oriented and solution-focused approach, highlighting specific welfare-related mechanisms that can distort behavioral and physiological readouts. Particular attention is given to social isolation and aggression in male mice, the role of nesting material in mitigating thermal stress, and the effects of circadian disruption under standard and reversed light–dark cycles. By integrating ethological theory with laboratory animal welfare research, this review argues that housing conditions should be regarded as integral components of experimental design rather than secondary technical variables. Addressing welfare-related bias through evidence-based refinement strategies is essential for improving reproducibility, enhancing data interpretability, and strengthening the scientific validity of mouse-based research. Full article

Shelter environments frequently expose dogs to chronic stress and anxiety, which can compromise their welfare and reduce their chances of adoption. Recent interest in psychedelic-assisted approaches has suggested potential therapeutic applications in veterinary behavioral medicine, although empirical evidence remains scarce. This study aimed to evaluate the combined effects of low-dose 1-cyclopropionyl lysergic acid diethylamide (1cp-LSD), a legal lysergamide prodrug of LSD in several countries, and ethological intervention (EI) on the behavior and welfare of shelter dogs. Twenty dogs were randomly assigned to four groups: pharmacological intervention, ethological intervention, combined treatment, or control. The ethological sessions were conducted by veterinary behaviorists, and pharmacological treatment consisted of 10 µg of 1cp-LSD administered orally for three weeks. Blinded evaluators assessed animals using validated anxiety and welfare scales, including a treatment expectation scale, before, during and after the intervention. Results showed that the combined condition consistently outperformed single interventions, significantly enhancing sociability, calmness, and positive emotional reactivity. Importantly, these improvements persisted for three weeks following treatment cessation, indicating sustained benefits beyond the active intervention phase. These findings provide preliminary evidence for the potential of integrating low doses of psychedelics with behavioral therapy in shelter settings. Future studies with larger cohorts and refined pharmacokinetic data are required to confirm safety, elucidate mechanisms, and optimize protocols for clinical application in veterinary practice. Full article

Timely, objective pain recognition in dairy cattle is essential for welfare assurance, productivity, and ethical husbandry yet remains elusive because evolutionary pressure renders bovine distress signals brief and inconspicuous. Without verbal self-reporting, cows suppress overt cues, so automated vision is indispensable for on-farm triage. Although earlier systems tracked whole-body posture or static grimace scales, frame-level detection of facial micro-expressions has not been explored fully in livestock. We translate micro-expression analytics from automotive driver monitoring to the barn, linking modern computer vision with veterinary ethology. Our two-stage pipeline first detects faces and 30 landmarks using a custom You Only Look Once (YOLO) version 8-Pose network, achieving a 96.9% mean average precision ($mAP$) at an Intersection over the Union (IoU) threshold of 0.50 for detection and 83.8% Object Keypoint Similarity (OKS) for keypoint placement. Cropped eye, ear, and muzzle patches are encoded using a pretrained MobileNetV2, generating 3840-dimensional descriptors that capture millisecond muscle twitches. Sequences of five consecutive frames are fed into a 128-unit Long Short-Term Memory (LSTM) classifier that outputs pain probabilities. On a held-out validation set of 1700 frames, the system records 99.65% accuracy and an F1-score of 0.997, with only three false positives and three false negatives. Tested on 14 unseen barn videos, it attains 64.3% clip-level accuracy (i.e., overall accuracy for the whole video clip) and 83% precision for the pain class, using a hybrid aggregation rule that combines a 30% mean probability threshold with micro-burst counting to temper false alarms. As an early exploration from our proof-of-concept study on a subset of our custom dairy farm datasets, these results show that micro-expression mining can deliver scalable, non-invasive pain surveillance across variations in illumination, camera angle, background, and individual morphology. Future work will explore attention-based temporal pooling, curriculum learning for variable window lengths, domain-adaptive fine-tuning, and multimodal fusion with accelerometry on the complete datasets to elevate the performance toward clinical deployment. Full article

This commentary evaluates contemporary equestrian sport governance through the lens of equine welfare science. Drawing on evidence from the FEI Sport Forum 2025 debates, the IFHA Racing Integrity Handbook, media coverage of horse sport, recent scientific presentations, regulatory texts, and published research, we identify systemic shortcomings in how horse welfare is assessed, prioritised, and protected. Key issues include reliance on performance as a proxy for welfare, inadequate “fit-to-compete” protocols, neglect of horses’ mental states, coercive tack and equipment practices (e.g., double bridles, tight nosebands, ear hoods), pharmacological and surgical interventions that mask pain, euphemistic regulatory language (e.g., whip “encouragement”), and inconsistent implementation of welfare rules. Through a series of case studies, from dressage and show jumping forums to racing integrity handbooks, we illustrate euphemistic language, defensive group dynamics, dismissive rhetoric towards evidence-based criticism, and a troubling “stable blindness” that sidelines the horse’s perspective. We conclude that meaningful reform requires (1) embedding validated behavioural and physical welfare indicators into all competition and pre-competition protocols, (2) transparent, evidence-inclusive rule-making under a precautionary principle, (3) genuine engagement with independent equine welfare experts, and (4) establishment of empowered, impartial oversight bodies to ensure that stated codes of conduct translate into consistent, enforceable practice. Only by catering to the horse’s subjective experiences and applying modern ethological and bioethical standards can equestrian sport retain its social licence and ensure integrity in all areas of competition management. Full article

The relentless drive to meet global demand for poultry products has pushed for rapid intensification in chicken farming, dramatically boosting efficiency and yield. Yet, these gains have exposed a host of complex welfare challenges that have prompted scientific scrutiny and ethical reflection. In this review, I critically evaluate recent innovations aimed at mitigating such concerns by drawing on advances in behavioral science and digital monitoring and insights into biological adaptations. Specifically, I focus on four interconnected themes: First, I spotlight the complexity of avian sensory perception—encompassing vision, auditory capabilities, olfaction, and tactile faculties—to underscore how lighting design, housing configurations, and enrichment strategies can better align with birds’ unique sensory worlds. Second, I explore novel tools for gauging emotional states and cognition, ranging from cognitive bias tests to developing protocols for identifying pain or distress based on facial cues. Third, I examine the transformative potential of computer vision, bioacoustics, and sensor-based technologies for the continuous, automated tracking of behavior and physiological indicators in commercial flocks. Fourth, I assess how data-driven management platforms, underpinned by precision livestock farming, can deploy real-time insights to optimize welfare on a broad scale. Recognizing that climate change and evolving production environments intensify these challenges, I also investigate how breeds resilient to extreme conditions might open new avenues for welfare-centered genetic and management approaches. While the adoption of cutting-edge techniques has shown promise, significant hurdles persist regarding validation, standardization, and commercial acceptance. I conclude that truly sustainable progress hinges on an interdisciplinary convergence of ethology, neuroscience, engineering, data analytics, and evolutionary biology—an integrative path that not only refines welfare assessment but also reimagines poultry production in ethically and scientifically robust ways. Full article

Post-Traumatic Stress Disorder (PTSD) is a multifaceted psychiatric disorder triggered by traumatic events, leading to prolonged psychological distress and varied symptoms. Rat models have been extensively used to explore the biological, behavioral, and neurochemical underpinnings of PTSD. This review critically examines the strengths and limitations of commonly used rat models, such as single prolonged stress (SPS), stress–re-stress (S-R), and predator-based paradigms, in replicating human PTSD pathology. While these models provide valuable insights into neuroendocrine responses, genetic predispositions, and potential therapeutic targets, they face challenges in capturing the full complexity of PTSD, particularly in terms of ethological relevance and translational validity. We assess the degree to which these models mimic the neurobiological and behavioral aspects of human PTSD, highlighting areas where they succeed and where they fall short. This review also discusses future directions in refining these models to improve their utility for translational research, aiming to bridge the gap between preclinical findings and clinical applications. Full article

Individual identification is an important ability for humans and perhaps also for non-human animals to lead social lives. It is also desirable for laboratory experiments to keep records of each animal while rearing them in mass. However, the specific body parts or the acceptable visual angles that enable individual identification are mostly unknown for non-human animals. In this study, we investigated whether artificial intelligence (AI) could distinguish individual medaka, a model animal for biological, agrarian, ecological, and ethological studies, based on the dorsal view. Using Teachable Machine, we took photographs of adult fish (n = 4) and used the images for machine learning. To our surprise, the AI could perfectly identify the four individuals in a total of 11 independent experiments, and the identification was valid for up to 10 days. The AI could also distinguish eight individuals, although machine learning required more time and effort. These results clearly demonstrate that the dorsal appearances of this small spot-/stripe-less fish are polymorphic enough for individual identification. Whether these clues can be applied to laboratory experiments where individual identification would be beneficial is an intriguing theme for future research. Full article

Agonistic behaviours are often directed at other animals for self-defence or to increase distance from valued resources, such as food. Examples include aggression and counter-predator behaviours. Contemporary diets may boost the value of food as a resource and create unanticipated associations with the humans who deliver it. At the same time the domestic horse is asked to carry the weight of riders and perform manoeuvres that, ethologically, are out-of-context and may be associated with instances of pain, confusion, or fear. Agonistic responses can endanger personnel and conspecifics. They are traditionally grouped along with so-called vices as being undesirable and worthy of punishment; a response that can often make horses more dangerous. The current study used data from the validated online Equine Behavioural and Research Questionnaire (E-BARQ) to explore the agonistic behaviours (as reported by the owners) of 2734 horses. With a focus on ridden horses, the behaviours of interest in the current study ranged from biting and bite threats and kicking and kick threats to tail swishing as an accompaniment to signs of escalating irritation when horses are approached, prepared for ridden work, ridden, and hosed down (e.g., after work). Analysis of the responses according to the context in which they arise included a dendrographic analysis that identified five clusters of agonistic behaviours among certain groups of horses and a principal component analysis that revealed six components, strongly related to the five clusters. Taken together, these results highlight the prospect that the motivation to show these responses differs with context. The clusters with common characteristics were those observed in the context of: locomotion under saddle; saddling; reactions in a familiar environment, inter-specific threats, and intra-specific threats. These findings highlight the potential roles of fear and pain in such unwelcome responses and challenge the simplistic view that the problems lie with the nature of the horses themselves rather than historic or current management practices. Improved understanding of agonistic responses in horses will reduce the inclination of owners to label horses that show such context-specific responses as being generally aggressive. Full article

The Chinese giant salamander (Andrias davidianus), an endangered amphibian species endemic to China, has been previously evaluated with regards to its phyletic evolution, zooecology, and ethology, but molecular mechanisms underlying its skin pigmentation remain unknown. Herein, a skin transcriptome database of different colored salamanders was established using RNA-seq, and a total of 47,911 unigenes were functionally annotated. Among these unigenes, a total of 1252 differentially expressed genes (DEGs) were annotated in the seven public databases, and six DEGs were validated by qPCR between five different skin colors and eight tissues. The results showed that TYR, TYRP1, and ASIP were significantly differentially expressed between different body colors, while TYR, TYRP1, and DCT were highly expressed in skin tissue. The full-length complementary DNA of TYR was cloned and analyzed between normal and yellow phenotypes. Three nucleotide sequence deletion sites were identified in the coding region of TYR, leading to premature termination of transcription and translation in yellow individuals. Our study provides useful data for the further study of the molecular mechanisms of melanin formation, and a valuable reference for the breeding of specific skin colors in other salamanders. Full article

Modern herpetoculture has seen a rise in welfare-related habitat modifications, although ethologically-informed enclosure design and evidence-based husbandry are lacking. The diversity that exists within snakes complicates standardizing snake welfare assessment tools and evaluation techniques. Utilizing behavioral indicators in conjunction with physiological measures, such as fecal glucocorticoid metabolite concentrations, could aid in the validation of evidence-based metrics for evaluating snake welfare. We increased habitat cleaning, to identify behavioral or physiological indicators that might indicate heightened arousal in snakes as a response to the disturbance. While glucocorticoid metabolite concentrations increased significantly during a period of increased disturbance, this increase was not associated with a significant increase in tongue-flicking, a behavior previously associated with arousal in snakes. Locomotion behavior and the proportion of time spent exposed were also not affected by more frequent habitat cleaning. These results demonstrate the need to further investigate the behavioral and physiological responses of snakes to different aspects of animal care at a species and individual level. They also highlight the need to collect baseline behavioral and physiological data for animals, in order to make meaningful comparisons when evaluating changes in animal care. Full article