The nomenclature used to describe animals working in roles supporting people can be confusing. The same term may be used to describe different roles, or two terms may mean the same thing. This confusion is evident among researchers, practitioners, and end users. Because certain animal roles are provided with legal protections and/or government-funding support in some jurisdictions, it is necessary to clearly define the existing terms to avoid confusion. The aim of this paper is to provide operationalized definitions for nine terms, which would be useful in many world regions: “assistance animal”, “companion animal”, “educational/school support animal”, “emotional support animal”, “facility animal”, “service animal”, “skilled companion animal”, “therapy animal”, and “visiting/visitation animal”. At the International Society for Anthrozoology (ISAZ) conferences in 2018 and 2020, over 100 delegates participated in workshops to define these terms, many of whom co-authored this paper. Through an iterative process, we have defined the nine terms and explained how they differ from each other. We recommend phasing out two terms (i.e., “skilled companion animal” and “service animal”) due to overlap with other terms that could potentially exacerbate confusion. The implications for several regions of the world are discussed.

Creating intelligent systems capable of recognizing emotions is a difficult task, especially when looking at emotions in animals. This paper describes the process of designing a “proof of concept” system to recognize emotions in horses. This system is formed by two elements, a detector and a model. The detector is a fast region-based convolutional neural network that detects horses in an image. The model is a convolutional neural network that predicts the emotions of those horses. These two elements were trained with multiple images of horses until they achieved high accuracy in their tasks. In total, 400 images of horses were collected and labeled to train both the detector and the model while 40 were used to test the system. Once the two components were validated, they were combined into a testable system that would detect equine emotions based on established behavioral ethograms indicating emotional affect through the head, neck, ear, muzzle, and eye position. The system showed an accuracy of 80% on the validation set and 65% on the test set, demonstrating that it is possible to predict emotions in animals using autonomous intelligent systems. Such a system has multiple applications including further studies in the growing field of animal emotions as well as in the veterinary field to determine the physical welfare of horses or other livestock.

Studies show that horses express favoritism through shared proximity and time and demonstrate unique affiliative behaviors such as allogrooming (mutual scratching) with favorite conspecifics. Allogrooming also occurs more frequently during stress and has been observed to occur more frequently in domestic herds than feral. The role of partner preference, lateralization, and duration of allogrooming as measures of social bonding has remained unclear. The present study looked at two socially stable herds of mares (n = 85, n = 115) to determine the frequency, duration, visual field of view and partner preference during allogrooming in both pasture settings (low stress) and confined settings (higher stress). One hundred and fifty-three videos for both herds were coded for allogrooming behaviors with 6.86 h recorded in confined conditions and 31.9 h in pasture settings. Six allogrooming sessions were observed in the pasture setting with an average duration of 163.11 s. In confined settings, a total of 118 allogrooming sessions were observed with an average duration of 40.98 s. Significant (p < 0.01) differences were found between settings for duration (s), number of allogrooming pairs, and frequency of allogrooming (per min) for each herd. All observed allogrooming sessions involved pairs of favored conspecifics (one partner per horse). The current study suggests that horses may have friendships that can be observed through the demonstration of specific affiliative behaviors during times of stress with more frequent, but shorter affiliative interactions with preferred partners during times of stress. This context suggests that horses adhere to the “tend and befriend” principles of friendship in animals.

Lateralised behavioural responses to environmental stressors have become more frequently used as indicators of social welfare in animals. These lateralised behavioural responses are under the control of asymmetrical brain functions as part of the primary functions of most vertebrates and assist in primary social and survival functions. Lateralised behavioural responses originating from the left hemisphere are responsible for processing familiar conditions, while the right hemisphere is responsible for responding to novel stimuli in the environment. The forced lateralisation and side preference tests have been used to determine the visual lateralised behavioural responses in livestock to environmental stressors. Limb preference during movement has also been used to determine motor lateralisation. Although behavioural investigations in livestock have recorded lateralised behavioural responses to environmental stressors, there are still limitations in the implication of lateralisation to other conditions, such as restraint and invasive procedures. Thus, it is important to have a non-invasive measure for these lateralised behavioural responses. Recently, lateralised behavioural responses have been correlated with the use of infrared temperature of external body surfaces, such as the eyes and coronary bands of limbs. This review summarised the different forms of the lateralised behavioural responses in livestock, especially cattle and horses, to environmental stressors, and the association between these responses and the relevant external body surfaces’ infrared temperature, with the purpose of improving the use of non-invasive measures in assessing welfare conditions in animals. The combination of the lateralised behavioural responses and infrared temperature of external body surfaces to environmental stressors could improve the assessment strategies of welfare conditions and the related additional husbandry interventions that could be applied to improve the welfare of farm animals.

This research applies unsupervised learning on a large original dataset of horses in the wild to identify previously unidentified horse emotions. We construct a novel, high-quality, diverse dataset of 3929 images consisting of five wild horse breeds worldwide at different geographical locations. We base our analysis on the seven Panksepp emotions of mammals “Exploring”, “Sadness”, “Playing”, “Rage”, “Fear”, “Affectionate” and “Lust”, along with one additional emotion “Pain” which has been shown to be highly relevant for horses. We apply the contrastive learning framework MoCo (Momentum Contrast for Unsupervised Visual Representation Learning) on our dataset to predict the seven Panksepp emotions and “Pain” using unsupervised learning. We significantly modify the MoCo framework, building a custom downstream classifier network that connects with a frozen CNN encoder that is pretrained using MoCo. Our method allows the encoder network to learn similarities and differences within image groups on its own without labels. The clusters thus formed are indicative of deeper nuances and complexities within a horse’s mood, which can possibly hint towards the existence of novel and complex equine emotions.

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.