Deep learning detects subtle facial expressions in a multilevel society primate.

Facial expressions in nonhuman primates are complex processes involving psychological, emotional, and physiological factors, and may use subtle signals to communicate significant information. However, uncertainty surrounds the functional significance of subtle facial expressions in animals. Using artificial intelligence (AI), this study found that nonhuman primates exhibit subtle facial expressions that are undetectable by human observers. We focused on the golden snub-nosed monkeys (Rhinopithecus roxellana), a primate species with a multilevel society. We collected 3427 front-facing images of monkeys from 275 video clips captured in both wild and laboratory settings. Three deep learning models, EfficientNet, RepMLP, and Tokens-To-Token ViT, were utilized for AI recognition. To compare the accuracy of human performance, two groups were recruited: one with prior animal observation experience and one without any such experience. The results showed human observers to correctly detect facial expressions (32.1% for inexperienced humans and 45.0% for experienced humans on average with a chance level of 33%). In contrast, the AI deep learning models achieved significantly higher accuracy rates. The best-performing model achieved an accuracy of 94.5%. Our results provide evidence that golden snub-nosed monkeys exhibit subtle facial expressions. The results further our understanding of animal facial expressions and also how such modes of communication may contribute to the origin of complex primate social systems.

Flexible grouping patterns in a western and eastern chimpanzee community.

Primate social organizations, or grouping patterns, vary significantly across species. Behavioral strategies that allow for flexibility in grouping patterns offer a means to reduce the costs of group living. Chimpanzees (Pan troglodytes) have a fission-fusion social system in which temporary subgroups ("parties") change in composition because of local socio-ecological conditions. Notably, western chimpanzees (P. t. verus) are described as showing a higher degree of bisexual bonding and association than eastern chimpanzees, and eastern female chimpanzees (P. t. schweinfurthii) are thought to be more solitary than western female chimpanzees. However, reported comparisons in sociality currently depend on a small number of study groups, particularly in western chimpanzees, and variation in methods. The inclusion of additional communities and direct comparison using the same methods are essential to assess whether reported subspecies differences in sociality hold in this behaviorally heterogeneous species. We explored whether sociality differs between two communities of chimpanzees using the same motion-triggered camera technology and definitions of social measures. We compare party size and composition (party type, sex ratio) between the western Gahtoy community in the Nimba Mountains (Guinea) and the eastern Waibira community in the Budongo Forest (Uganda). Once potential competition for resources such as food and mating opportunities were controlled for, subspecies did not substantially influence the number of individuals in a party. We found a higher sex-ratio, indicating more males in a party, in Waibira; this pattern was driven by a greater likelihood in Gahtoy to be in all-female parties. This finding is the opposite of what was expected for eastern chimpanzees, where female-only parties are predicted to be more common. Our results highlight the flexibility in chimpanzee sociality, and caution against subspecies level generalizations.

News and Perspectives: Words matter in primatology.

Postings on social media on Twitter (now X), BioAnthropology News (Facebook), and other venues, as well as recent publications in prominent journals, show that primatologists, ecologists, and other researchers are questioning the terms "Old World" and "New World" due to their colonial implications and history. The terms are offensive if they result in erasing Indigenous voices and history, ignoring the fact that Indigenous peoples were in the Americas long before European colonization. Language use is not without context, but alternative terminology is not always obvious and available. In this perspective, we share opinions expressed by an international group of primatologists who considered questions about the use of these terms, whether primatologists should adjust language use, and how to move forward. The diversity of opinions provides insight into how conventional terms used in primatological research and conservation may impact our effectiveness in these domains.

DeepWild: Application of the pose estimation tool DeepLabCut for behaviour tracking in wild chimpanzees and bonobos.

Studying animal behaviour allows us to understand how different species and individuals navigate their physical and social worlds. Video coding of behaviour is considered a gold standard: allowing researchers to extract rich nuanced behavioural datasets, validate their reliability, and for research to be replicated. However, in practice, videos are only useful if data can be efficiently extracted. Manually locating relevant footage in 10,000 s of hours is extremely time-consuming, as is the manual coding of animal behaviour, which requires extensive training to achieve reliability. Machine learning approaches are used to automate the recognition of patterns within data, considerably reducing the time taken to extract data and improving reliability. However, tracking visual information to recognise nuanced behaviour is a challenging problem and, to date, the tracking and pose-estimation tools used to detect behaviour are typically applied where the visual environment is highly controlled. Animal behaviour researchers are interested in applying these tools to the study of wild animals, but it is not clear to what extent doing so is currently possible, or which tools are most suited to particular problems. To address this gap in knowledge, we describe the new tools available in this rapidly evolving landscape, suggest guidance for tool selection, provide a worked demonstration of the use of machine learning to track movement in video data of wild apes, and make our base models available for use. We use a pose-estimation tool, DeepLabCut, to demonstrate successful training of two pilot models of an extremely challenging pose estimate and tracking problem: multi-animal wild forest-living chimpanzees and bonobos across behavioural contexts from hand-held video footage. With DeepWild we show that, without requiring specific expertise in machine learning, pose estimation and movement tracking of free-living wild primates in visually complex environments is an attainable goal for behavioural researchers.

L'étude du comportement animal nous permet de comprendre comment différentes espèces et différents individus naviguent dans leur monde physique et social. Le codage vidéo du comportement est considéré comme une référence: il permet aux chercheurs d'extraire des ensembles de données comportementales riches et nuancées, de valider leur fiabilité et de reproduire les recherches. Toutefois, dans la pratique, les vidéos ne sont utiles que si les données peuvent être extraites efficacement. La localisation manuelle de séquences pertinentes parmi des dizaines de milliers d'heures prend énormément de temps, tout comme le codage manuel du comportement animal, qui nécessite une formation approfondie pour être fiable. Les approches d'apprentissage automatique sont utilisées pour automatiser la reconnaissance de modèles dans les données, ce qui réduit considérablement le temps nécessaire à l'extraction des données et améliore la fiabilité. Toutefois, le suivi des informations visuelles pour reconnaître un comportement nuancé est un problème difficile et, à ce jour, les outils de suivi et d'estimation de la pose utilisés pour détecter le comportement sont généralement appliqués lorsque l'environnement visuel est hautement contrôlé. Les chercheurs en comportement animal sont intéressés par l'application de ces outils à l'étude des animaux sauvages, mais il n'est pas clair dans quelle mesure cela est actuellement possible, ni quels outils sont les mieux adaptés à des problèmes particuliers. Pour combler ce manque de connaissances, nous décrivons les nouveaux outils disponibles dans ce paysage en évolution rapide, proposons des conseils pour la sélection des outils, fournissons une démonstration pratique de l'utilisation de l'apprentissage automatique pour suivre les mouvements dans les données vidéo des grands singes sauvages et mettons nos modèles de base à disposition pour utilisation. Nous utilisons un outil d'estimation de la pose, DeepLabCut, pour démontrer l'apprentissage réussi de deux modèles pilotes d'un problème extrêmement difficile d'estimation et de suivi de la pose: les chimpanzés et les bonobos sauvages vivant dans la forêt et représentant plusieurs animaux dans différents contextes comportementaux à partir de séquences vidéo tenues à la main. Avec DeepWild, nous montrons que, sans nécessiter d'expertise spécifique en apprentissage automatique, l'estimation de la pose et le suivi des mouvements de primates sauvages vivant en liberté dans des environnements visuellement complexes est un objectif réalisable pour les chercheurs en comportement.

Sequence Order in the Range 1 to 19 by Chimpanzees on a Touchscreen Task: Processing Two-Digit Arabic Numerals.

The sequence of Arabic numerals from 1 to 19 was taught to six chimpanzees, three pairs of mother and child. Each chimpanzee participant sat facing a touchscreen on which the numerals appeared in random positions within an imaginary 5-by-8 matrix. They had to touch the numerals in ascending order. Baseline training involved touching the adjacent numerals from 1 to X or from the numeral X to 19. Systematic tests revealed the following results: (1) The range 1 to 9 was easier than 1 to 19. (2) Adjacent numerals were easier than nonadjacent ones. (3) The "masking" (memory task) caused deterioration of performance. All these factors depended on the number of numerals simultaneously presented on the screen. A chimpanzee named Pal mastered the skill of ordering two-digit numerals with 100% accuracy. Human participants were tested in the same experiment with the same procedure. Both species showed relative difficulty in handling two-digit numerals. Global-local information processing is known to be different between humans and other primates. The assessment of chimpanzee performance and comparison with humans were discussed in terms of the possible difference in the global-local dual information processing of two-digit numerals.

Garrano Horses Perceive Letters of the Alphabet on a Touchscreen System: A Pilot Study.

This study aimed to use a computer-controlled touchscreen system to examine visual discrimination in Garrano horses (Equus caballus), an endangered breed of pony belonging to the Iberian horse family. This pilot study focused on the perceptual similarity among letters of the alphabet. We tested five horses in a one-male unit (OMU) living permanently in a semi-free enclosure near their natural habitat in Serra d'Arga, northern Portugal. Horses were trained to nose-touch black circles that appeared on the screen. Then, they were tested for discrimination of five letters of the Latin alphabet in Arial font, namely O, B, V, Z, and X, using a two-choice discrimination task. The confusion matrix of letter pairs was used to show the MDS and to identify the relative contribution of shape features. The results showed perceptual similarities among letters with curvatures pitted against those of straight lines. Shape perception in horses seems to share features with that of humans and other animals living in different niches. The touchscreen system proved to be an objective and innovative way of studying cognition in the socially organized group of horses. The automated system can promote the welfare of captive horses by maximizing their freedom of movement.

Apex and ApeTouch: Development of a Portable Touchscreen System and Software for Primates at Zoos.

We report on the development and testing of a portable touchscreen apparatus and accompanying software program for primate enrichment, cognitive research, and husbandry applications. For zoos considering using technology to bolster scientific efforts or enhance the welfare of primates in their care, touchscreen activities offer a solution that has a long and proven record of primate use in laboratory settings as well as a history of usage in the zoo world. We review the options that are available for zoos to build their own touchscreen systems and we offer as an alternative our pre-built apparatus, Apex, and primate software suite, ApeTouch, both of which are tailored for use in a zoo setting. The efficacy and utility of these tools are demonstrated in a training study with four macaque groups of different species that were previously naïve to touchscreens. All of the groups in the study learned to use the device and displayed a consistent engagement with the touchscreen tasks over 95 daily sessions of exposure. In the final stage of the training, two of the four groups displayed an above-chance level performance on a numerical sequencing task.

To drum or not to drum: Selectivity in tree buttress drumming by chimpanzees (Pan troglodytes verus) in the Nimba Mountains, Guinea.

Chimpanzees live in fission-fusion social organizations, which means that party size, composition, and spatial distribution are constantly in flux. Moreover, chimpanzees use a remarkably extensive repertoire of vocal and nonvocal forms of communication, thought to help convey information in such a socially and spatially dynamic setting. One proposed form of nonvocal communication in chimpanzees is buttress drumming, in which an individual hits a tree buttress with its hands and/or feet, thereby producing a low-frequency acoustic signal. It is often presumed that this behavior functions to communicate over long distances and is, therefore, goal-oriented. If so, we would expect chimpanzees to exhibit selectivity in the choice of trees and buttresses used in buttress drumming. Selectivity is a key attribute of many other goal-directed chimpanzee behaviors, such as nut-cracking and ant dipping. Here, we investigate whether chimpanzees at the Seringbara study site in the Nimba Mountains, Guinea, West Africa, show selectivity in their buttress drumming behavior. Our results indicate that Seringbara chimpanzees are more likely to use larger trees and select buttresses that are thinner and have a greater surface area. These findings imply that tree buttress drumming is not a random act, but rather goal-oriented and requires knowledge of suitable trees and buttresses. Our results also point to long-distance communication as a probable function of buttress drumming based on selectivity for buttress characteristics likely to impact sound propagation. This study provides a foundation for further assessing the cognitive underpinnings and functions of buttress drumming in wild chimpanzees.

Field experiments find no evidence that chimpanzee nut cracking can be independently innovated.

Cumulative culture has been claimed a hallmark of human evolution. Yet, the uniqueness of human culture is heavily debated. The zone of latent solutions hypothesis states that only humans have cultural forms that require form-copying social learning and are culture-dependent. Non-human ape cultural behaviours are considered 'latent solutions', which can be independently (re-)innovated. Others claim that chimpanzees, like humans, have cumulative culture. Here, we use field experiments at Seringbara (Nimba Mountains, Guinea) to test whether chimpanzee nut cracking can be individually (re-)innovated. We provided: (1) palm nuts and stones, (2) palm fruit bunch, (3) cracked palm nuts and (4) Coula nuts and stones. Chimpanzee parties visited (n = 35) and explored (n = 11) the experiments but no nut cracking occurred. In these experiments, chimpanzees did not individually (re-)innovate nut cracking under ecologically valid conditions. Our null results are consistent with the hypothesis that chimpanzee nut cracking is a product of social learning.

Exploring greetings and leave-takings: communication during arrivals and departures by chimpanzees of the Bossou community, Guinea.

In human fission-fusion societies, ritualized non-linguistic signal exchanges that include gestures, vocalizations, and facial expressions are regularly observed at both arrivals (greetings) and departures (leave-takings). These communicative events play an important role in the formation and maintenance of social relationships. Wild chimpanzees also form large communities that split into smaller fluid parties during daily activities, with individuals moving freely between them. However, in chimpanzees only greetings have been reported. This study explores signal exchanges in the Bossou chimpanzee community during fissions (departures) and fusions (arrivals) given an individual's social rank, kinship, position as traveller or party-member, the level of potential threat, and the party size and presence of mature males. We analysed three time periods (1993-1994; 2003-2004; 2013-2014) during which the composition and social hierarchy of the community varied. We show that the occurrence and form of communication during fission and fusion events are mediated by social factors, including rank, kinship, and party size and composition. Individuals were more likely to communicate during fusions than during fissions, communication was more likely to be produced towards a higher-ranking individual and to non-kin individuals, but the tendency to communicate in general increased with an increase in social rank. The presence of more individuals, and in particular mature males, decreased the likelihood of communication. Communication during fusions supported patterns reported in previous studies on greetings, and our results support the argument that, if present, leave-takings are not a common feature of chimpanzee social interactions. Current methodological difficulties regarding the function of declarative signals hinder our ability to discriminate potential parting rituals within communication before departures. Given similar methodological difficulties, we also provide a note of caution in the interpretation of all signals produced during fusions as 'greetings'.