Animal Behaviour

Predation is a driving force in the ecology and evolution of prey, and primates exhibit diverse anti-predator strategies for minimizing risk. Because these behaviors can be costly, individuals must balance costs and benefits when responding to perceived threats. The cognitive capacity and behavioral plasticity of baboons make them an ideal taxon for studying the context-dependent variation in anti-predator strategies. Here, we used an autonomous, motion-activated playback experiment to study the behavioral responses of chacma baboons (Papio ursinus griseipes) to simulated predator encounters in Gorongosa National Park, Mozambique. We compared responses in 2021, when predator densities were relatively low, to responses in 2024, after predation increased due to lion (Panthera leo) population recovery and African wild dog (Lycaon pictus) reintroduction. We compared flight and vigilance responses to vocalizations of these common predators with responses to leopard (Panthera pardus), historically a key predator; spotted hyena (Crocuta crocuta), a rare predator; and cheetah (Acinonyx jubatus), absent historically and currently. We also assessed how responses varied with habitat, age-sex class, presence of offspring, and group size. Across 916 predator playbacks, baboons fled in 19% and displayed vigilance in 71% of trials. When predator density was higher, baboons displayed weakened antipredator responses, consistent with the risk allocation hypothesis. Baboons were more likely to flee in response to lion and wild dog cues. Juveniles fled more frequently than other demographic classes, while adult females with offspring were more vigilant. Overall, responses were highly heterogeneous, reflecting the substantial intraspecific variation and behavioral flexibility characteristic of baboons.

Traditionally, primates have been considered primarily visual animals. However, studies across a variety of taxa suggest that, in the context of food evaluation, the reliance on this sense might be more nuanced that previously thought, with dietary specialization and food item properties leading to differences in sensory prioritization. We performed a field-based study assessing the use of sensory cues during food evaluation as well as food-related behaviours such as muzzle contact in two mixed-sex groups of wild vervet monkeys including three age classes over a period of five months (nmonkeys = 44). Using a total of 18868 food evaluation observations collected over 44 hours of focal follows, we found that vervets mainly relied on their sense of vision when evaluating food (96.8% of all instances). Sensory usage varied according to food category and sex differences were only observed in the use of smell for a subset of these. Juveniles initiated muzzle contact and used tactile inspection more often than adults whereas females received muzzle contact more often than males. In addition, the low rejection rates suggest that most food items were familiar to the vervets regardless of age and sex. These findings are in line with optimal foraging theory according to which the food evaluation process should be adapted to the familiarity of food items and allows individuals to maximize their intake of energy and critical nutrients, while minimizing the time and effort in food evaluation.

Sentinel behaviour occurs when individuals use raised positions to scan for predators while the rest of the group forages. Here, we investigated whether a colonial cooperatively breeding species that forages in large groups, the sociable weaver, Philetairus socius, displays sentinel behaviour. This behaviour has been reported in species with similar ecology, behaviour and foraging habits, (e.g. ground foraging in open habitats where aerial predators are common) and, hence, we expected that it could occur in sociable weavers. On the other hand, sentinel behaviour appears to be less common in species that live in very large groups. We used an experimental set-up consisting of an artificial feeding station and perches to assess occurrence of sentinel related behaviours: (i) perching events > 30s on an elevated position, (ii) head-movements and (iii) alarm calling. Birds were seldom observed perching while others fed, and those that did, perched for periods that were too short to be considered as sentinel behaviour (less than 5s on average). Our results suggest that this behaviour is uncommon or even absent in sociable weavers. We discuss whether other factors such as foraging in very large groups, or interspecific foraging associations might make sentinel behaviour less important in this species.

Object play - seemingly non-functional interactions with objects - can promote the development of foraging skills, tool use, and behavioral innovation. Among Catarrhine monkeys, it was described in macaques and baboons. Wild geladas, although closely related to baboons, have been described as lacking object play (observed only in captivity) linked to their specialized grazing ecology. Here, we provide the first evidence of both social and solitary object play in a wild gelada population (NOMUs=13) at Debre Libanos (Ethiopia) and compare it with object play in sympatric olive baboons (Nindividuals=42). Notably, immature geladas engaged in object play both socially and solitarily, but the latter case was most frequent also with novel objects introduced by researchers. Solitary object play occurred at levels comparable to those of baboons, challenging previous reports of limited object interest in geladas. This finding aligns with the occurrence of object play in phylogenetically related species and with the retention in wild geladas of arboreal behavior and fruit consumption and hand morphology enhancing fine manipulation. Hence, object play in geladas under certain environmental conditions may reflect a biologically rooted capacity and underscores the importance of ecological variability, as distinct behavioral ecotypes can emerge across different populations of the same species.

Protogynous sex change, where individuals first function as females and later as males, is a key life-history strategy among polygynous reef fishes. In haremic systems, sex change is typically socially regulated, with dominants suppressing subordinates sex change through aggression. Females within a harem form a size-based hierarchy that can remain stable in most species through the threat of eviction. We studied a different situation in the cleaner wrasse Labroides dimidiatus, where larger females have incomplete control, as they spend most of their time alone at their own cleaning territory. We tracked over 400 individuals for 12 months, recording growth, behavior, social organization, and sex change. We confirmed earlier reports that both sexes direct aggression primarily at those ranked immediately below them. However, we observed 30 cases where smaller females outgrew larger ones, revealing hierarchy instability. Of 42 sex change events, 43% occurred in presence of the male, and half of these early sex changers were not the largest female, but individuals overlooked by the male. Fast growth relative to harem-mates and harem switching increased the likelihood of sex change. Local population densities also influenced growth and sex change, with individuals in high-density demes growing faster and changing sex at larger sizes. Our findings reveal flexible sex change dynamics in a system with incomplete social dominance. Such incomplete control and observations that becoming male confers both higher reproductive success and survival highlight the need to expand game-theoretical and life-history frameworks to encompass such strategic flexibility. Lay summaryDominant cleaner wrasse cannot fully control subordinates as individuals occupy distinct core areas. Tracking 400 fish for a year, we found that smaller females could outgrow initially larger ones, and early sex change despite a larger male. Fast growth and harem switching increased the chances of becoming male. Population density also shaped these strategies. Our findings reveal flexible sex change dynamics in a system where becoming male confers both higher reproductive success and survival.

Animals must continually balance foraging with the risk of predation. In complex natural environments, this means quickly distinguishing between threats and harmless situations. We investigated how site-associated coral reef fishes decide to escape in response to visual cues mimicking predator attacks, using controlled underwater presentations of looming stimuli at varying speeds. We measured escape responses across species and social contexts, comparing them to predator attack speeds observed in the same habitat. Escape responses were highly sensitive to the speed of the looming stimulus, with no responses occurring at low speeds. The speeds triggering escape matched those of predator attacks, whereas cruising swim speeds never triggered a response. Species employed distinct antipredator strategies: Brown Chromis foraged away from shelter with high responsiveness, whereas Bicolor Damselfish remained shelter-dependent with lower escape propensities. Contrary to expectations, the social factors did not affect responses in this study. These findings demonstrate that reef fish are highly sensitive to the approach speed of objects, with species-specific strategies further shaping behaviors. By combining realistic visual threats with natural predator attack data, this study offers insight into how animals make escape decisions in complex, real-world environments.

Many animals communicate using sequences of signals, but identifying recurrent, non-random signal combinations remains methodologically challenging. Collocation analyses are increasingly popular approaches for detecting which signals animals combine at rates greater than expected by chance. However, existing methods for animal collocation analysis face several limitations that reduce their statistical rigour: they lack uncertainty estimates, fail to control for non-independence in sampled data, and do not account for inflated family-wise error rates when identifying attraction among many different signal types. These limitations restrict the broader applicability of animal collocation analysis, including preventing robust comparisons of signal combination strength between cohorts (e.g. populations, sexes or age classes). We adapt a novel form of Multiple Distinctive Collocation Analysis using Pearson residuals (MDCA-Pr) that addresses these statistical limitations, and validate its use in animal communication research in three ways: first, using numerous simulated datasets of different sizes and levels of signal recombination; second, using simulated data to evaluate the performance of MDCA-Pr in intercohort comparisons, and third, by demonstrating how MDCA-Pr can be applied to compare the vocal sequences produced by male and female captive-living common marmosets (Callithrix jacchus). MDCA-Pr shows high sensitivity, including at small sample sizes, and generally low false-positive rates, which we further reduce by applying additional criteria for identifying attraction between signals. During intercohort comparisons, MDCA-Pr is conservative, with low false-positive rates, and statistical power increases with sample size. MDCA-Pr is a robust method for evaluating signal attraction in animal communication and enables accurate intercohort comparison of animal signal combinations. Significance StatementBy assessing the performance of MDCA-Pr on simulated animal-like data, we demonstrate that this method reliably detects signal combinations within and across animal cohorts, while overcoming statistical limitations of previous collocation analyses. We present an analytical pipeline for applying MDCA-Pr to animal signal data, including for intercohort comparisons, enabling identification and comparison of combinatorial strategies across entire signal repertoires. We illustrate this approach by comparing call combination strategies of male and female common marmosets when presented with food under experimental conditions, finding similar combinatorial strategies between sexes. MDCA-Pr therefore permits rigorous characterization of animal signal combinatoriality and opens avenues for investigating how demographic, social, and group-level factors influence combinatorial patterns.

Despite growing evidence that many animals can evaluate quantities, the ecological relevance of numerical cognition remains debated, particularly outside vertebrates. Would individuals still rely on numerousness if less computationally demanding cues, visual features extracted at the early stage of visual processing, were available to assess quantity? In primates, individuals show a numerical bias as they tend to rely on the number of items rather than non-numerical cues, such as total area, to categorize quantities. In this study, we trained free-flying honeybees to discriminate between two and four items in conditions where numerosity covaried with the total area and perimeter (Experiment Size) or the convex hull (Experiment Space) cues, mimicking ecological contexts. Transfer tests assessed which numerical or non-numerical cues were learned and preferentially used by the bees. Bees primarily relied on numerousness over these non-numerical cues. Individual analyses revealed two consistent strategies: a "numerical bias" strategy, in which bees encoded numerical information while ignoring non-numerical cues, and a "generalist" strategy, where bees flexibly switched between cues and favored non-numerical information when cues conflicted. We further reported improved discrimination when smaller quantities appeared on the left and larger ones on the right, consistent with an oriented mental number line. Together, these findings demonstrate a spontaneous numerical bias in honeybees and reveal that individuals within the same species can adopt distinct strategies when evaluating quantity. Our findings also suggest that distantly related taxa like bees and primates may have independently evolved comparable mechanisms for quantity evaluation.

In mammals, loud, high-pitched, and harsh-sounding calls typically accompany heightened emotional arousal, particularly during distress such as separation. However, whether subtle arousal reductions can be detected through acoustic analysis within a single negative context remains unclear. We investigated whether source-related acoustic parameters of puppy whines reflect arousal modulations induced by calming interventions during maternal separation. Thirty-five eight-week-old Beagle puppies were recorded under four conditions combining synthetic appeasing pheromone and a pressure harness. Vocal behavior, activity, whine duration, and intensity, did not significantly differ across treatments, suggesting interventions did not suppress separation-related vocal responses. Nevertheless, calming products selectively altered acoustic parameters known to index arousal in dog vocalizations. Puppies receiving combined treatments produced whines with lower fundamental frequency (fo) and reduced fo variability, while pheromone exposure increased call tonality, reflected by reduced jitter and shimmer and elevated harmonics-to-noise ratios. Spectral entropy remained unchanged, possibly because the proportion of whines containing nonlinear phenomena did not vary across conditions. Reductions in fo, fo variability, and acoustic roughness are consistent with established correlates of lower arousal in mammals, suggesting source-related vocal parameters sensitively capture subtle arousal shifts even when overt vocal behavior remains stable, supporting their use as bioacoustic indicators for evaluating welfare interventions.

Sleeping and feeding are crucial for survival of any animal. In case of large predators, knowing where these activities occur can help us understand their behavioural adaptations for coexisting with people and could help mitigating human-carnivore conflicts. Leopard (Panthera pardus) is an elusive and highly adaptable large felid that mostly lives outside protected areas and can survive also in close proximity to humans. However, most leopard research in Africa has been conducted in protected areas and we poorly understand leopards habitat selection while resting and hunting. To shed light on their coexistence with humans, we investigated habitat features influencing leopard selection of resting and kill sites on farmlands in central Namibia, using generalized linear mixed models (GLMM) under a use-availability study design and blinded field-sampling. Leopards primarily selected resting sites that were located in mountainous, steep, rugged terrain and sites with good concealment while kill sites were selected in mountainous habitats. Human infrastructure did not affect leopard resting and kill site selection. Thus, the capacity of leopards to perform essential life-supporting behaviours while coexisting with people appears to be primarily driven by their ability to remain concealed, rather than spatially avoiding humans.

We assessed whether pigs provide consolation, referring to targeted affiliation that attenuates a partners stress, under experimental conditions that manipulated exposure to stressed partners. Using a within-subject design, 74 pigs were tested in three contexts: a helping task in which group members could observe and help a trapped focal pig to return to the group, a direct-reunion, in which group members were naive to the experience of a separated focal pig until reunion, and an undisturbed control. We measured affiliative and non-affiliative interactions, anxiety behaviours and changes in salivary cortisol. Only the helping context satisfied most consolation criteria: there were selective increases in unidirectional affiliative contacts from the observer to the focal pig, non-affiliative interactions remained at baseline, and focal pigs showed fewer anxiety behaviours. In contrast, direct-reunions triggered increases in affiliative and non-affiliative interactions and higher anxiety. Cortisol increased during both direct-reunions and helping, but its level was not linked to affiliation. Results add to growing evidence for consolation behaviour in pigs and suggest best practices for reintegrating pigs into groups. Graded reintroductions that allow observers to assess the emotional state of targets may promote social buffering, whereas abrupt regrouping may trigger more generalized arousal or personal distress.

Honeypot ants represent an example of convergent evolution, where a group of workers specialized in storing liquid food in their crops (i.e., stomach) has independently evolved multiple times across different ant genera. While seasonal resource scarcity and arid conditions are thought to drive the evolution of repletism, the role of environmental variables in this process has not been tested. With this is mind, species ensemble models were computed to assess suitability and richness areas, and the importance of predictors. Predictor importance was compared between genera and groups occupying a similar geographical area. Niche overlap and similarity between honeypot ant species were also evaluated to determine whether they occupy similar environmental spaces. Similarity was mainly found within genera, and Leptomyrmex and Myrmecocystus showed striking niche differences. Overall, Leptomyrmex distribution was mainly influenced by atmospheric bioclimatic variables like precipitation and temperature, while Myrmecocystus had soil bioclimatic variables as the most important predictors for their current distribution. Our results indicate that honeypot ants species currently do not occupy the same environmental space, and are not experiencing the same contemporary environmental stressors. While our results suggest that contemporary environmental factors cannot explain the convergence of honeypot ants, future research will examine past climatic conditions along with investigations into the ant genomes to understand more about the causes and consequences of the convergence.

Socially-controlled sex changing fishes provide powerful model systems for investigating sexual development and phenotypic plasticity in both behavior and physiology. The remarkable sexual transformation these fishes undertake is strongly influenced by their position in dominance hierarchies. However, the behavioral mechanisms underlying hierarchical formation remain understudied, particularly among female groups. Here, we investigated the role of winner-loser effects among females in establishing social dominance in a female-to-male sex changing fish. Individuals with prior losing experiences were more likely to lose subsequent size-matched fights, demonstrating clear loser effects, while there was no evidence for winner effects. Initial mirror aggression and some prior fighting behaviors, particularly submission, significantly and positively correlated with aggression in size-matched fights and subsequent mirror aggression; however, contest outcomes were not altered by these factors. Additionally, mirror aggression increased significantly only in subjects that drew size-matched fights. These findings demonstrate complex fighting dynamics in female-female competition and confirm the presence of loser effects in a sequential hermaphroditic species. These effects may represent evolutionarily advantageous mechanisms underlying sex change, thereby offering further context for examining how social rank advantages drive sexual transition.

Although predation is a major driver of group living across taxa and the antipredator benefits of grouping are well established, the energetic costs experienced by groups under predation remain largely unexplored. In the current study, we use wild, white mullet (Mugil curema, Valenciennes 1836), to provide the first real-time quantification of the energetic cost of escape in schooling fish using intermittent, closed-loop respirometry. We found that small groups exposed to predators showed a 53.8% increase in their organismal metabolic rate (MO2) as compared to groups without predator exposure. When we evaluated antipredator behaviors such as escape response, group cohesion, and displacement of the group centroid, we found a positive correlation to energetic costs. We then investigated whether escape responses are socially modulated by comparing the energetic costs of escape across solitary individuals, solitary individuals with visual access to a group, and groups. We found that escape frequency and energetic costs to predation were comparable across social contexts, indicating that escape may be an intrinsic survival response independent of cues from group members. Furthermore, we found that fish exposed to predators showed markedly reduced feeding, suggesting that predation constrains energy acquisition in addition to imposing direct energetic costs. Our results provide the first direct quantification of the energetic costs of escape in a schooling fish, offering new insights into the physiological trade-offs underlying collective antipredator defenses.

BackgroundIn hibernating mammals, the timing of den entry and exit reflects complex interactions among environment, physiology, and energetic constraints, with consequences for fitness. Consequently, shifts in denning phenology can affect population dynamics, particularly under climate change. Reliable estimation of denning timing is therefore critical, yet current methods often rely on GPS-derived movement data, limited by coarse sampling intervals, detection issues, and the inability to distinguish true inactivity from active presence at the den site. In this study, we developed and apply a method to estimate denning phenology in a brown bear population in south-central Sweden using accelerometer-derived activity data. Our approach employs adaptive, individual-specific thresholds to account for variation in baseline activity across bears, focusing on day-to-day changes to identify the start and end of inactivity periods. This method allows flexible and reproducible detection of den entry and exit dates, overcoming limitations associated with fixed thresholds and small sample sizes. ResultsWe compared activity-based estimates with GPS-derived den occupancy and examined variation in denning behavior across demographic groups. Analyzing 388 bear-winters, the method successfully identified inactivity periods in 360 cases. The method failed to identify clear start and end dates of hibernation for 28 (7%) bear-winters, which were characterized by unusually high or low daily activity levels at the boundaries of the inactivity period. Den site occupancy ranged from September 5 to June 2, with durations of 112-260 days, whereas inactivity periods detected from activity data extended from September 6 to May 13, lasting 83-217 days. Our comparison of activity-based and GPS-based methods indicates that bears may arrive at the den site several weeks before the onset of inactivity, with timing varying among demographic groups. ConclusionWe show that activity-based analysis provides a robust framework for estimating denning phenology, distinguishing actual inactivity from site presence, and improving understanding of the timing and variability of bear denning behavior. Applying an individual-level activity-based method improves accuracy in assessing ecological mechanisms underlying hibernation in bears and other hibernators, while also enhancing interpretation of environmental drivers and providing a reliable tool to monitor phenological shifts in response to climate change.

Drumming--rhythmic, percussive sound production using body parts or external objects--is rare among non-human animals, with confirmed tool-assisted cases previously limited to primates and Palm Cockatoos. Here, we report the first documented instance of spontaneous, tool-assisted drumming in a Galah (Eolophus roseicapilla). A captive, male Galah produced rhythmic tapping by striking a coconut shell against a metal bowl. Across 14 recorded sessions, the bird displayed consistent temporal structure characterised by two stable tapping rates (approximately 0.8 s and 0.2 s inter-onset intervals) arranged into recurring phrases. This pattern indicates a simple hierarchical rhythmic organisation with a 4:1 ratio between metrical levels. The birds behaviour emerged without training, apparent reinforcement, or known exposure to conspecific or human drumming models, suggesting an intrinsic capacity for rhythmic tool use. Although the function of the behaviour remains unclear--play, nutrient extraction, or communicative signalling--these observations extend known rhythmic and tool-using abilities within cockatoos and raise new evolutionary questions. Our findings highlight the potential for rhythmically structured, instrumental behaviour to arise in a broader range of avian taxa than previously recognised, motivating further comparative and experimental work on the cognitive and biomechanical foundations of drumming in parrots.

Vocal accommodation is the process by which individuals adjust their vocalizations to resemble those of social partners. This phenomenon is widespread in social animals and can reinforce affiliation, signal group identity, and facilitate coordination. Most studies of vocal accommodation have focused on convergence in the acoustic structure of individual calls. Whether social partners also converge in how calls are arranged into sequences remains largely unknown. We examined vocal convergence during pair formation in common marmosets (Callithrix jacchus) by recording phee sequences from nine dyads composed of three males and three females before pairing and again four months after, in two audience contexts: when individuals interacted vocally with their partner or with an opposite sex stranger. We quantified similarity between individuals in call sequence-structure using transition probabilities, bigram frequencies, repeat-length distributions, and local alignment, and quantified similarity in acoustic structure using spectral parameters, MFCCs, and dynamic time warping. We found vocal convergence on a sequence level. After pair formation, partners became more similar in sequence structure when calling to strangers, whereas no change was detected in partner directed sequences. In contrast, call acoustic structure did not change in either context. Because vocal repertoires are constrained by anatomy and physiology, reorganizing existing call types into different combinations may provide a flexible route for modifying signals without altering the acoustic structure of individual calls. Our results provide evidence that social bonds can drive sequence level vocal convergence in a non-human primate, suggesting that vocal flexibility may arise not only through changes in acoustic structures but also through changes in how calls are organized over time.

Humans order numbers in space from left to right, with smaller quantities represented preferentially in the left hemispace and larger ones in the right hemispace. The direction of this mental number line (MNL), or more generally of number-space associations (NSA), is influenced by cultural habits such as reading and writing direction. However, a growing body of evidence from pre-verbal infants and non-human animals suggests that number-space mappings may also have biological foundations. In non-human primates, evidence for a directional MNL remains mixed, partly due to small sample sizes and methodological heterogeneity. Here, we tested samples of rhesus (Macaca mulatta) and crab-eating macaques (Macaca fascicularis) across two experiments using spontaneous food-related tasks. In Experiment 1, monkeys chose between identical food quantities (1x1 to 24x24) presented on the left and right. No systematic spatial choice bias emerged as a function of numerical magnitude, and hand use did not differ across exact numerical pairs, although exploratory analyses revealed magnitude-related modulations of manual responses. In Experiment 2, monkeys were habituated to small (4x4) or large (16x16) quantities and subsequently tested with the alternative quantity. Result showed significantly more leftward choices following numerical decreases (16[->]4) and more rightward choices following numerical increases (4[->]16), indicating that relative numerical context, rather than absolute magnitude, elicited directional spatial biases. These findings suggest that in macaques, number-space associations emerge most robustly in comparative contexts involving expectancy violations of magnitude.

Pretend play is a hallmark behavior in childhood where children create nonliteral meanings. Empirical data supporting the role of social cognition and the decoupling from literality are still scarce during early development. We explored here how the comprehension of pretense affects the visual exploratory behavior of toddlers (n = 44) and adults (n = 65) when they were exposed to short video clips in which an actress performed either real actions (e.g., eating jelly) or pretend actions (e.g., pretending to eat with imaginary food), while varying the complexity of those actions. We analyzed participants exploration of the face in the videos as exploitation of social information. We showed that all observers paid more attention to the face in pretend scenarios than in real ones, measured as longer total looking time in adults and more fixations and revisits to the face in both age groups. We also found more gaze shifts (a measure of information sampling) between the face and the moving hand in the pretend videos in both age groups, mainly at the initial stages of the actions. Additionally, analyses of the scanpaths structure using gaze entropy showed less order in the exploration of pretend videos in both age groups, suggesting that pretense involved greater uncertainty and increased information seeking. The less structured trajectories were observed again mainly in complex pretend scenarios. Taken together, our gaze results indicate that from its developmental origins, the comprehension of pretense relies on social processes linked with information seeking and exploration. Significance StatementDevelopmental theories have long debated whether pretend games are born in conjunction with social capacities in the second year or become integrated later in life. Our study shows that, much like adults, toddlers visually explore pretend scenes gathering more social information and in a less structured manner compared to real-world scenarios, suggesting that the emerging capacity to play with the meaning of things is linked with that of thinking of other minds early in life.

Collective behavior in animal societies can buffer individual costs and confer resilience to environmental challenges. However, the mechanisms by which groups sustain function when members are compromised remain poorly understood. In the presented study, we investigate how social context shapes collective fanning, a thermoregulatory behavior critical for colony function, in Western honeybees (Apis mellifera). Using oxytetracycline (OTC), a known physiologically disruptive antibiotic to honeybees, to selectively impair certain group members, we tested our hypothesis that the presence of untreated bees would rescue the fanning response in mixed-composition groups. We show that groups containing untreated individuals fan at levels comparable to fully untreated groups, despite the presence of OTC-impaired bees. This preservation of collective thermoregulatory function was correlated with both treated and untreated individuals in mixed groups shifting their interaction dynamics and social network positions. These findings reveal a decentralized mechanism of collective resilience, whereby behavioral compensation by individuals sustains group-level thermoregulation under partial disruption. Our results provide a framework for understanding how social insect colonies maintain function in the face of individual-level perturbations, with broader implications for predicting the limits of collective resilience in animal societies experiencing increasing environmental pressures.