Skip to main content
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Aug 29;21(4):733–740. doi: 10.1111/1749-4877.13040

To Share or Not to Share: Food Sharing in Wild Azure‐Winged Magpie (Cyanopica cyanus)

Yijin Du 1, Pu Yang 1,2, Jiale Peng 2, Chenya Wang 2, Sen Song 2, Jiani Chen 1,✉
PMCID: PMC13362278  PMID: 40878617

ABSTRACT

Food sharing is very common in the animal kingdom. Despite extensive research, the mechanisms underlying food sharing remain debated. Moreover, the majority of studies on food sharing in non‐human animals have been conducted under controlled environments. The natural behavioral characteristics and ecological factors influencing the natural selection of food sharing are still not well understood. In this study, we introduce a method to examine food‐sharing behaviors in wild birds within their natural habitat. Using two types of feeders—one permitting food sharing with conspecifics and the other providing exclusive access to food—along with infrared cameras to record feeder‐triggering events, we found that azure‐winged magpies did not exhibit proactive food‐sharing behaviors. However, they did exhibit passive food‐sharing behaviors under conditions of low food availability. Consistent with the harassment‐avoidance hypothesis, the birds prioritized the sharing feeders to decrease harassment from food snatching when food was limited. These findings suggest that food sharing is likely shaped into a passive behavioral pattern under limited resources conditions, which induce conflicts within social groups. This context‐dependent strategy may effectively reduce harassment costs, optimize individual access to resources to maximize individual benefits, and potentially enhance the survival of other group members.

Keywords: azure‐winged magpie, cooperative mechanisms, food sharing, harassment, prosociality


Darwin's theory of natural selection emphasizes competition for survival. It is intriguing how altruistic behaviors like food sharing evolved. Our study shows that sharing is not always caring. Instead of proactively sharing food, wild azure‐winged magpies reactively shared by prioritizing sharing feeders over monopoly alternatives to reduce food snatching when food was limited. This self‐interested sharing could lower the cost of competition and be preferred by natural selection.

graphic file with name INZ2-21-733-g002.jpg

1. Introduction

Prosocial behaviors, such as resource sharing, comforting, cooperating, helping, and altruistic punishment, are voluntary actions that benefit others (Eisenberg and Miller 1987). Specifically, as an important form of resource sharing, food sharing is considered to enhance social cohesion, reduce intragroup competition, lower foraging costs, and promote cooperation within animal societies (Gilby 2006; Silk et al. 2013). It is defined as a situation where two or more individuals jointly consume a food resource that could otherwise be monopolized (Stevens and Gilby 2004). Food sharing can manifest as either active or passive behavior. Active food sharing involves individuals intentionally transferring food to others, often observed among relatives. On the other hand, passive food sharing occurs when the food possessor allows others to access food without defending it, such as by letting others take food from their hands or mouths, co‐feed, or pick up food within their reach (Marshall‐Pescini et al. 2016). Notably, although food sharing also occurs among unrelated individuals (Stevens and Gilby 2004; Gilby 2006), only a few species have been documented to exhibit active food sharing (Stevens 2004; Ostojić et al. 2014; Duque and Stevens 2016). In comparison, passive food sharing is far more common and may function as a strategy to reduce potential conflicts or harassment (Gilby 2006; Silk et al. 2013; Bowden‐Parry et al. 2020).

The prevalence and mechanisms of food sharing have been studied in a wide range of species, including insects (Vahed 1998), cetaceans (Hoelzel 1991), small mammals (Judd and Sherman 1996; Brotherton et al. 2001), birds (Heinrich 1988; Evans and Marler 1994; Schwab et al. 2012; Duque and Stevens 2016), bats (Wilkinson 1984; Carter and Wilkinson 2013), and primates (de Waal 1989, 1997; Nettelbeck 1998; Gilby 2006; Gomes and Boesch 2009; Eppley et al. 2013; Crick et al. 2013). Various hypotheses have been proposed to explain food sharing, such as kin selection (Hamilton 1964; Silk et al. 2013), reciprocity hypothesis (Wilkinson 1984; de Waal 1997; Gomes and Boesch 2009; Jaeggi and Van Schaik 2011; Jaeggi et al. 2013; Jaeggi and Gurven 2013; Duque and Stevens 2016), harassment‐avoidance hypothesis (Gilby 2006; Silk et al. 2013; Bowden‐Parry et al. 2020), cooperative breeding hypothesis (Burkart et al. 2007), and so on.

However, these hypotheses have limitations that restrict their applicability across species and contexts. The extent and nature of food sharing vary significantly across species (Jaeggi et al. 2010a, 2010b; Schwab et al. 2012; Massen et al. 2015), ages (de Kort et al. 2006), habitats, social hierarchies (Jaeggi et al. 2010b; Duque and Stevens 2016), food types, and food availability (de Kort et al. 2006; Crick et al. 2013; Bowden‐Parry et al. 2020; Massen et al. 2020). Identifying the immediate biological and ecological drivers that determine which hypothesis is applied will help disentangle the complexities of the evolution of sharing behavior. However, despite extensive research on the ultimate evolutionary mechanisms of food sharing, studies on its proximate causes remain insufficient (Marshall‐Pescini et al. 2016).

Food sharing is influenced by a variety of immediate ecological and social factors. Food availability modulates sharing behavior, particularly large or indivisible items (e.g., carcasses in carnivores), are more likely to be shared due to reduced ability of any one individual to monopolize them (de Kort et al. 2006; Jaeggi and Van Schaik 2011; Crick et al. 2013; Massen et al. 2020). In contrast, scarcity intensifies competition and often suppresses sharing behavior (Stevens and Stephens 2002). Food type further influences this dynamic: easily guarded foods (e.g., small fruits) show lower sharing rates compared to bulky or dispersed foods (Jaeggi and Van Schaik 2011; Crick et al. 2013; Bowden‐Parry et al. 2020). Competitive pressure, such as food‐snatching behavior, can drive food sharing as a strategy to mitigate conflict and avoid harassment (Stevens and Stephens 2002; Stevens 2004; Gilby 2006; Silk et al. 2013; Bowden‐Parry et al. 2020).

Numerous studies on food sharing have been conducted in controlled environments. However, there has been little discussion on the integrated effects of these ecological and social factors, despite their combined influence in nature. To determine the ecological and social factors that contribute to the emergence of food sharing, we conducted a field experiment with azure‐winged magpies (Cyanopica cyanus), a species of the Corvidae family, in their natural habitat, manipulating three key variables involving food amount, food type, and competitive pressure. In addition, food sharing among non‐human primates has been extensively studied due to their complex social structures and cognitive abilities (de Waal 1997; Stevens and Gilby 2004; Jaeggi et al. 2010b; Jaeggi and Van Schaik 2011; Melis et al. 2011; Kopp and Liebal 2016). However, birds, particularly those such as corvids—often referred to as “feathered apes” due to their social nature and intelligence (Clayton and Emery 2015)—have been studied less extensively in the context of food sharing compared to non‐human primates (Marshall‐Pescini et al. 2016). azure‐winged magpies are social birds and cooperative breeders that have been observed assisting conspecifics in obtaining food (Horn et al. 2016; Massen et al. 2020). They also show proficiency in various cognitive tasks (Wang et al. 2019), making them suitable subjects for exploring group food sharing in the natural environment.

We designed two types of feeders: one allowing food sharing with conspecifics (the prosocial option) and another offering exclusive access to food (the selfish option). We recorded the feeder‐triggering events, the order in which feeders were triggered, and food‐snatching behaviors to evaluate the birds’ sharing behavior and explore how different factors—food amount, food type, and competitive pressures—influence these behaviors. This design enabled us to test whether magpies would share food regardless of context (as might be expected if they possessed strong intrinsic prosocial tendencies), or whether they would share only under specific conditions (as expected if sharing is driven by ecological context and social competition).

2. Materials and Methods

2.1. Subjects and Apparatus

The study was conducted in a wild population of azure‐winged magpies (C. cyanus), a common wild resident on the Yuzhong Campus of Lanzhou University, Lanzhou, Gansu, China. The experimental apparatus included two types of feeders: the monopoly and the sharing feeders (Figure 1A). Each feeder consisted of a plastic bottle with a food container inside, sealed with cotton. The two types of feeders were similar, except that the monopoly feeders had closed bottoms, whereas the sharing feeders had open bottoms. When a bird removed the cotton from the monopoly feeder, the food dropped onto the bottom, permitting exclusive access to a single individual. In contrast, when a bird operated the sharing feeder, the food dispersed onto the ground through the open bottom, enabling simultaneous feeding by multiple individuals.

FIGURE 1.

FIGURE 1

Experimental setup for food‐sharing experiment in azure‐winged magpies. (A) Diagram of the feeders; (B) Feeder arrangement; (C) Field setup of feeders.

2.2. Experimental Procedure

Our experimental procedure consisted of a 60‐day preliminary learning phase, followed by a formal food‐sharing experiment conducted from February to May.

2.2.1. Learning Phase

This phase included a 15‐day free‐feeding period, followed by a 45‐day period during which the azure‐winged magpies learned to remove the cotton from the apparatus to obtain food. During the initial 15 days, free food was provided to attract azure‐winged magpies to three sites on campus. The experimental feeders were presented at the same times to allow the birds to acclimate to them. Over the following 45 days, free food was removed, and one sharing feeder and one monopoly feeder spaced 20 cm apart, both baited with mealworms, were placed at each site. During this period, the birds visiting the sites learned to operate the feeders through either trial and error or social observation.

2.2.2. Food‐Sharing Experiment

Following the learning phase, we conducted the formal experiment to evaluate sharing behavior under controlled conditions of food availability and type. Experimental variables included food amount (small: 5 mL, large: 20 mL) and food type (bird food pellets or mealworms). At each site, 20 feeders were positioned in two parallel rows spaced 40 cm apart: one row contained sharing feeders, the other monopoly feeders. Within each row, feeders were spaced 20 cm apart and alternated between large and small amounts (Figure 1B,C). While feeder locations remained constant, food types (pellet and mealworm) were alternated daily. All feeders were replenished at dusk (19:00 ± 10 min) to reset experimental conditions each day.

2.2.3. Data Collection

A total of 95 h of video footage were collected throughout the formal food‐sharing experiment, using passive infrared‐triggered cameras to monitor the behaviors of azure‐winged magpies at each site. From the video, we recorded days, sites, feeder‐triggering events, triggering order (1–20 per replenishment cycle), instances of food‐snatching behaviors targeting focal birds during feeder activation, and the number of azure‐winged magpies present at each triggering event (hereafter referred to as “group size,” with a range of 0–59). Aggressive behaviors, including pecking, body shoving, and displacing focal individuals while feeding, were classified as food‐snatching behaviors. Due to the lack of individual markings, analyses focused on feeder operations within foraging groups without information on specific individuals.

2.3. Statistical Analyses

2.3.1. Feeder‐Triggering Events

A binomial GLMM was used to model whether a feeder was triggered or not. Fixed effects included feeder type, food type, food amount, pairwise interactions between these three predictors, and days (number of days throughout the experiment). The site was included as a random effect. Group size was excluded in this model because quantifying group size was unfeasible for non‐triggered events due to the prolonged duration of the non‐triggered state, during which group size fluctuates as a dynamic process.

2.3.2. Feeder Type Preference

A binomial GLMM was used to model feeder type choice (monopoly = 1, sharing = 0). Fixed effects included food type, food amount, group size, and pairwise interactions between these factors and days. The site was included as a random effect. This model assessed the influence of group size on feeder choice, based on the assumption that the presence of more birds might influence the preference for a particular feeder type.

2.3.3. Order of Feeder Triggering

A cumulative link mixed model (CLMM; ordinal package) was used to analyze the order in which feeders were triggered, treated as an ordinal response (Christensen 2019). Fixed effects included feeder type, food type, food amount, and their pairwise interactions. Site was included as a random effect. This model applied a proportional odds framework suitable for ordinal outcomes. Post hoc pairwise comparisons were conducted using estimated marginal means (emmeans) and predicted probabilities were computed by inverse logit transformation of the linear predictors.

2.3.4. Food‐Snatching Behavior

The number of food‐snatching events was modeled using a negative binomial GLMM (to control for the overdispersion of the count data, family = nbinom2, glmmTMB package) (Brooks et al. 2017) with feeder type, food type, food amount, and their pairwise interactions as fixed effects, and site as a random effect. Data without snatching events were excluded to ensure the response variable reflected actual behavioral interactions.

In addition, a non‐parametric Wilcoxon rank‐sum test was used to compare the intensity of food snatching (snatches per mL of food) between the 5 and 20 mL conditions. Normality and homogeneity of variances were violated (Shapiro–Wilk p < 0.001; Levene's test p < 0.001), justifying the use of the Wilcoxon test. The analysis was conducted separately for the two food types (mealworm and pellet). All statistical analyses were performed in R v4.1.0. Residual diagnostics were conducted using the DHARMa package (Hartig 2016) to evaluate model fit, dispersion, and residual patterns.

3. Result

3.1. Feeder‐Triggering Events

Throughout the experiment, a total of 4140 feeders were deployed across three sites, with azure‐winged magpies triggering 2894 of them. Of the triggered feeders, 1433 (49.5%) were sharing feeders. We analyzed the proportion of feeder triggering across different food types, food amount, feeder types, and days. Feeders containing mealworms were triggered significantly more often compared to those with pellets (Estimate = −0.760, SE = 0.127, z = −5.971, p < 0.001; Figure 2). Feeders containing a large food amount (20 mL) were triggered more frequently than those containing only 5 mL of food (Estimate = −0.375, SE = 0.130, z = −2.895, p = 0.004; Figure 2). The frequency of feeder triggering increased significantly with the progression of the experimental days (Estimate = 0.018, SE = 0.002, z = 10.161, p < 0.001), indicating a progressive rise in triggering events over time. However, no significant difference was observed in the likelihood of feeders being triggered between monopoly and sharing feeders (Estimate = −0.118, SE = 0.131, z = −0.900, p = 0.368; Figure 2). None of the interactions between the fixed effects showed a significant impact on feeder‐triggering events (all p > 0.05; Table S1).

FIGURE 2.

FIGURE 2

Triggering proportions by food type and amount. Mean (± SE) triggering proportions of the feeders, including the proportions of the total, monopoly, and sharing feeders under different food types (mealworm and pellet) and food amounts (20 and 5 mL). Error bars represent SE. Different lowercase letters (a, b, c, and d) above the bars indicate statistically significant pairwise differences (p < 0.05).

3.2. Feeder Type Preference

Further analysis indicated that the type of feeder triggered was neither influenced by the food amount (Estimate = −0.027, SE = 0.175, z = −0.152, p = 0.879) nor the food type (Estimate = 0.014, SE = 0.180, z = 0.077, p = 0.939). In addition, the feeder type triggered was not influenced by group size (Estimate = 0.003, SE = 0.007, z = 0.442, p = 0.659). The days also showed no significant impact on the choice between the two feeder types (Estimate = −0.002, SE = 0.002, z = −0.763, p = 0.445), indicating that feeder type preference did not change as the experiment progressed. Other pairwise interactions between the fixed effects showed no significant impact on feeder type preference (all p > 0.05; Table S2). This absence of significant preference for any feeder type suggests that active food sharing was not exhibited under these conditions.

3.3. Order of Feeder Triggering

We investigated the triggering order to determine whether there was a preference between the two types of feeders. The CLMM revealed a significant interaction between feeder type and food amount (Estimate = −0.258, SE = 0.129, z = −1.992, p = 0.046), indicating that the effect of feeder type on triggering order depends on the available food amount (Table S3). When food amount was small (5 mL), birds were significantly more likely to trigger the sharing feeder earlier than the monopoly feeder. (Estimate = 0.212, SE = 0.094, z = 2.257, p = 0.024; Figure 3). In contrast, under the 20 mL condition, no significant difference was found between feeder types in triggering order (Estimate = −0.046, SE = 0.089, z = 0.510, p = 0.610; Figure 3). [Correction added on 23 September 2025, after first online publication: In the preceding paragraph, “monopoly feeder earlier than the sharing feeder” was incorrectly published and has been corrected to “sharing feeder earlier than the monopoly feeder.”]

FIGURE 3.

FIGURE 3

Triggering order of monopoly and sharing feeders under different food amounts. Predicted probability of higher triggering order (i.e., being triggered later) for feeders under two food quantity conditions (20 and 5 mL). Probabilities are transformed from the logit scale. Error bars indicate 95% confidence intervals (CI). n.s., no significant difference. *p < 0.05.

Food type showed a significant main effect (Estimate = −0.388, SE = 0.111, z = −3.503, p < 0.001), the probability of higher triggering order was significantly higher in mealworm than in pellet (Estimate = 0.303, SE = 0.065, z = 4.648, p < 0.001), but no significant interactions were found between food type and either feeder type (Estimate = 0.180, SE = 0.130, z = 1.387, p = 0.166) or food amount (Estimate = −0.010, SE = 0.130, z = −0.080, p = 0.936). There was no significant main effect of feeder type (Estimate = −0.044, SE = 0.107, z = −0.417, p = 0.677) or food amount (Estimate = 0.128, SE = 0.109, z = 1.174, p = 0.240).

3.4. Food‐Snatching Behaviors

We analyzed the intensity of food‐snatching events under different food conditions. A Wilcoxon rank‐sum test revealed that the snatching intensity to get the same amount of food (snatches per mL of food) was significantly higher in the 5 mL condition than in the 20 mL condition for both food types (mealworm: W = 184430, p < 0.001; pellet: W = 128888, p < 0.001), indicating stronger competition when food was scarce (Figure 4A).

FIGURE 4.

FIGURE 4

Effects of food amount and feeder type on snatching behaviors. (A) Snatching strength (events per 1 mL of food), shown separately for 5 and 20 mL conditions within each food type (mealworm and pellet); (B) Total number of snatching events for monopoly versus sharing feeders, shown separately for mealworm and pellet conditions. Boxplots indicate median, interquartile range, and 1.5× IQR whiskers. ***p < 0.001.

The negative binomial GLMM revealed significant main effects of feeder type (Estimate = −0.882, SE = 0.052, z = −16.84, p < 0.001), food type (Estimate = −0.216, SE = 0.053, z = −4.06, p < 0.001), and food amount (Estimate = −0.934, SE = 0.053, z = −17.52, p < 0.001), as well as significant interactions between feeder type and food type (Estimate = 0.497, SE = 0.066, z = 7.58, p < 0.001), and feeder type and food amount (Estimate = 0.164, SE = 0.066, z = 2.48, p = 0.013). No significant interaction was found between food type and food amount (Estimate = −0.047, SE = 0.066, z = −0.70, p = 0.482). The number of snatching events was significantly lower in sharing feeders compared to the monopoly feeders across all food conditions (Figure 4B, Table S4): Mealworm‐20 mL (z = 16.84), Mealworm‐5 mL (z = 12.38), Pellet‐20 mL (z = 6.88), Pellet‐5 mL (z = 3.56); all P < 0.001.

4. Discussion

Our findings suggest that azure‐winged magpies do not exhibit proactive food‐sharing behaviors under natural foraging conditions. Contrary to our expectations, azure‐winged magpies did not trigger the sharing feeders more frequently than the monopoly feeders. Previous research has shown that azure‐winged magpies actively share food with familiar conspecifics (i.e., group members) (Horn et al. 2016; Massen et al. 2020), and these sharing behaviors have been explained by the cooperative breeding hypothesis (Horn et al. 2016). The prosocial tendencies evolved through cooperative breeding may not extend equally to all conspecifics, which could explain the absence of proactive food sharing in our study. In our setup, the foraging group may have contained unfamiliar conspecifics from different social groups, and the cooperative breeding hypothesis may not accurately predict prosocial behaviors between unfamiliar social members. Further experiments are needed to investigate this possibility. While azure‐winged magpies are cooperative breeders, we observed that group foraging behavior was maintained throughout both the pre‐breeding and early breeding periods. Moreover, although the frequency of feeder‐triggering events increased significantly over time, the bird’ preference for feeder type did not change, suggesting that breeding‐related social dynamics had limited influence on the observed food‐sharing patterns.

Alternatively, could the absence of proactive food sharing be attributed to the birds' inability to differentiate between the feeder types or their reluctance to share via creating a public resource? Or could the lack of proactive food sharing be due to the abundant and low‐cost opportunities to trigger the feeders? As the feeders were presented in an array of 20 at a time, this arrangement imposed little cost for birds searching for food. A bird could always turn to another feeder if one was occupied, which makes little difference between the two feeder types. The fact is that the birds have food‐snatching behaviors toward the foraging social mates. Although the cost of searching for food was low, triggering a feeder not only provided food but also risked inciting aggression from nearby birds, which imposes significant social costs, such as risks of aggression, displacement, and reduced feeding efficiency.

These alternative explanations seem unlikely, as we observed a preference for the sharing feeder when food amount was small (5 mL). If the magpies were unable to differentiate between the feeder types or were indifferent, no significant difference would have been expected in the triggering order. When the food amount was increased to 20 mL, no distinct preference was detected. On the other hand, could the observed priority triggering of sharing feeders reflect an inherent preference? This explanation also seems unlikely. If such a preference were inherent, whether driven by subjective bias or objective benefit, we would expect to observe consistent preference patterns across both food quantity conditions (5 and 20 mL). The response of azure‐winged magpies to varying food amounts is consistent with the context‐dependent foraging decisions seen in rufous hummingbirds (Bateson et al. 2003), indicating that they can adapt their strategies based on environmental conditions. Our observations suggest that the birds' food sharing may be a passive strategy aimed primarily at reducing harassment rather than a proactive cooperative tactic.

This interpretation is supported by the high intensity of snatching behavior under conditions of small food amount (Figure 4A), where choosing the sharing feeder effectively reduced the strength of food snatching (Figure 4B). In this context, food snatching can be understood as a form of harassment that disrupts feeding and reduces individual fitness. These results are in line with the harassment‐avoidance hypothesis, suggesting that azure‐winged magpies use the sharing feeder to minimize harassment and competition. The harassment‐avoidance hypothesis posits that sharing food provides immediate selfish benefits by reducing costly harassment and does not require complex cognitive skills, nor does it involve tracking future rewards or debts. Therefore, it is broadly applicable in explaining food sharing in large groups in natural settings (Stevens and Stephens 2002; Stevens 2004; de Kort et al. 2006; Gilby 2006). Studies in captive settings have also shown that harassment is a major motivator for food sharing. For example, chimpanzees and squirrel monkeys (Saimiri boliviensis) were found to share food more frequently when harassed, and increased harassment increased the frequency of sharing (Stevens 2004; Gilby 2006). A study on captive Asian small‐clawed otters showed that increased begging (considered to be a form of harassment) led to more food sharing (Bowden‐Parry et al. 2020). Together with our findings, these results highlight harassment as a primary motivator for food sharing among non‐human animals.

However, studies show varying impacts of food amount. In Asian small‐clawed otters, sharing behaviors were not influenced by food type or abundance (Bowden‐Parry et al. 2020). Captive jackdaws shared preferred food types more frequently than non‐preferred types (de Kort et al. 2006), and a large amount of food was found to induce proactive food sharing in captive chimpanzees (Hauser et al. 1993). Our research on wild azure‐winged magpies shows that the birds prioritized sharing only when food was scarce. Despite differences between species, particularly in response to experimental food manipulation, these variations may be attributed to the higher harassment costs faced by wild azure‐winged magpies compared to animals in captivity, where food competition might be less significant. Our results suggest that food sharing in azure‐winged magpies in the wild is likely a social strategy that is primarily passive in nature.

We found a significant effect of food type on triggering order. During the experiment, food type was fixed across all feeders within each day. Since only one type of food (either pellets or mealworms) was presented per day, this effect indicates that the higher probability of a higher triggering order in the mealworm condition is due to triggering more feeders. This interpretation is supported by the result that the triggering proportions of the total feeders were significantly higher in the mealworm condition than in the pellet condition.

Unlike traditional indoor experiments involving complex tasks like token exchange and bar‐pulling tests that require training (Tennie et al. 2016), our method enables animals to naturally choose between two types of feeders: one that promotes food sharing and another that facilitates solo feeding. This design mitigates potential biases, such as those influenced by reward anticipation or the exclusion of untrained individuals. It provides a more ecologically valid representation of natural animal behaviors. Furthermore, this method is applicable to outdoor studies across different species.

In summary, although azure‐winged magpies did not exhibit proactive food‐sharing behaviors, they demonstrated a preference for sharing feeders when available food was scarce. This indicates that azure‐winged magpies can adjust their behaviors in response to environmental changes by employing a passive food‐sharing strategy to avoid potential harassment. Our study reveals that three factors—the behavioral nature of foraging, group composition, and resource availability—are likely important for the application of the harassment‐avoidance hypothesis. From the biological perspective, cooperation is necessary for a species that is likely to arouse social conflict during resource acquisition. However, a passive approach to cooperation may be optimal for ensuring individual benefit when foraging in a group with mixed members (familiar, unfamiliar, and even other species). From an ecological perspective, limited available resources that are difficult to defend are likely to induce passive sharing as a means to avoid harassment and conflict. In addition, our approach bridges the gap between captive and wild behavior studies, offering a possible template for etho‐ecological experimentation.

Conflicts of Interest

The authors declare no conflicts of interest.

Ethics Statement

Animal experiments followed the guidelines of Lanzhou University and conformed to all applicable governmental regulations concerning the ethical use of animals and biodiversity rights.

Supporting information

Table S1 GLMM on Feeder Triggering Events

Table S2 GLMM on Feeder Type Preference

Table S3 CLMM on Order of Feeder Triggering

Table S4 Negative Binomial GLMM on Food‐Snatching Behavior.

INZ2-21-733-s001.doc (70.5KB, doc)

Acknowledgments

We sincerely thank Chuan Yan for his assistance with the statistical analysis and data visualization.

Du, Y. , Yang P., Peng J., Wang C., Song S., and Chen J.. 2026. “To Share or Not to Share: Food Sharing in Wild Azure‐Winged Magpie (Cyanopica cyanus).” Integrative Zoology 21, no. 4: 733–740. 10.1111/1749-4877.13040

Funding: This study was supported by the National Natural Science Foundation of China (Grant Number: 32070445).

Data Availability Statement

The dataset and analysis script supporting the findings of this study are publicly available at [figshare] via [https://doi.org/10.6084/m9.figshare.29644037.v1]. Analyses can be reproduced using the R script food_sharing_analysis_script.R and dataset food_sharing.csv.

References

  1. Bateson, M. , Healy S. D., and Hurly T. A.. 2003. “Context–Dependent Foraging Decisions in Rufous Hummingbirds.” Proceedings of the Royal Society of London Series B: Biological Sciences 270, no. 1521: 1271–1276. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bowden‐Parry, M. , Postma E., and Boogert N. J.. 2020. “Effects of Food Type and Abundance on Begging and Sharing in Asian Small‐Clawed Otters (Aonyx cinereus).” PeerJ 8: e10369. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Brooks, M. E. , Kristensen K., Benthem K. J., et al. 2017. “glmmTMB Balances Speed and Flexibility Among Packages for Zero‐Inflated Generalized Linear Mixed Modeling.” The R Journal 9, no. 2: 378. [Google Scholar]
  4. Brotherton, P. N. M. , Clutton‐Brock T. H., O'Riain M. J., et al. 2001. “Offspring Food Allocation by Parents and Helpers in a Cooperative Mammal.” Behavioral Ecology 12, no. 5: 590–599. [Google Scholar]
  5. Burkart, J. M. , Fehr E., Efferson C., and Van Schaik C. P.. 2007. “Other‐Regarding Preferences in a Non‐human Primate: Common Marmosets Provision Food Altruistically.” Proceedings of the National Academy of Sciences 104, no. 50: 19762–19766. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Carter, G. , and Wilkinson G.. 2013. “Does Food Sharing in Vampire Bats Demonstrate Reciprocity?” Communicative & Integrative Biology 6, no. 6: e25783. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Christensen, R. H. B. 2019. “Cumulative Link Models for Ordinal Regression with the R Package Ordinal.” R package vignette version 2023.12-4.1. Retrieved from. https://cran.r‐project.org/package=ordinal.
  8. Clayton, N. S. , and Emery N. J.. 2015. “Avian Models for Human Cognitive Neuroscience: A Proposal.” Neuron 86, no. 6: 1330–1342. [DOI] [PubMed] [Google Scholar]
  9. Crick, J. , Suchak M., Eppley T. M., Campbell M. W., and de Waal F. B. M.. 2013. “The Roles of Food Quality and Sex in Chimpanzee Sharing Behavior (Pan troglodytes).” Behaviour 150, no. 11: 1203–1224. [Google Scholar]
  10. de Kort, S. R. , Emery N. J., and Clayton N. S.. 2006. “Food Sharing in Jackdaws, Corvus monedula: What, Why and with Whom?” Animal Behaviour 72, no. 2: 297–304. [Google Scholar]
  11. de Waal, F. B. M. 1989. “Food Sharing and Reciprocal Obligations Among Chimpanzees.” Journal of Human Evolution 18, no. 5: 433–459. [Google Scholar]
  12. de Waal, F. B. M. 1997. “The Chimpanzee's Service Economy: Food for Grooming.” Evolution and Human Behavior 18, no. 6: 375–386. [Google Scholar]
  13. Duque, J. F. , and Stevens J. R.. 2016. “Voluntary Food Sharing in Pinyon Jays: The Role of Reciprocity and Dominance.” Animal Behaviour 122: 135–144. [Google Scholar]
  14. Eisenberg, N. , and Miller P. A.. 1987. “The Relation of Empathy to Prosocial and Related Behaviors.” Psychological Bulletin 101, no. 1: 91–119. [PubMed] [Google Scholar]
  15. Eppley, T. M. , Suchak M., Tinsman J., and de Waal F. B. M.. 2013. “Perseverance and Food Sharing Among Closely Affiliated Female Chimpanzees.” Primates; Journal of Primatology 54, no. 4: 319–324. [DOI] [PubMed] [Google Scholar]
  16. Evans, C. S. , and Marler P.. 1994. “Food Calling and Audience Effects in Male Chickens, Gallus gallus: Their Relationships to Food Availability, Courtship and Social Facilitation.” Animal Behaviour 47, no. 5: 1159–1170. [Google Scholar]
  17. Gilby, I. C. 2006. “Meat Sharing Among the Gombe Chimpanzees: Harassment and Reciprocal Exchange.” Animal Behaviour 71, no. 4: 953–963. [Google Scholar]
  18. Gomes, C. M. , and Boesch C.. 2009. “Wild Chimpanzees Exchange Meat for Sex on a Long‐Term Basis.” PLoS ONE 4, no. 4: e5116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hamilton, W. D. 1964. “The Genetical Evolution of Social Behaviour. I.” Journal of Theoretical Biology 7, no. 1: 1–16. [DOI] [PubMed] [Google Scholar]
  20. Hartig, F. 2016. “DHARMa: Residual Diagnostics for Hierarchical (Multi‐Level /Mixed) Regression Models.” R package version 0.4.7. Retrieved from: 10.32614/CRAN.package.DHARMa. [DOI]
  21. Hauser, M. D. , Teixidor P., Fields L., and Flaherty R.. 1993. “Food‐Elicited Calls in Chimpanzees: Effects of Food Quantity and Divisibility.” Animal Behaviour 45, no. 4: 817–819. [Google Scholar]
  22. Heinrich, B. 1988. “Winter Foraging at Carcasses by Three Sympatric Corvids, with Emphasis on Recruitment by the Raven, Corvus corax .” Behavioral Ecology and Sociobiology 23, no. 3: 141–156. [Google Scholar]
  23. Hoelzel, A. R. 1991. “Killer Whale Predation on Marine Mammals at Punta Norte, Argentina; Food Sharing, Provisioning and Foraging Strategy.” Behavioral Ecology and Sociobiology 29, no. 3: 197–204. [Google Scholar]
  24. Horn, L. , Scheer C., Bugnyar T., and Massen J. J. M.. 2016. “Proactive Prosociality in a Cooperatively Breeding Corvid, the Azure‐Winged Magpie (Cyanopica cyana).” Biology Letters 12, no. 10: 20160649. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Jaeggi, A. V. , Burkart J. M., and Van Schaik C. P.. 2010a. “On the Psychology of Cooperation in Humans and Other Primates: Combining the Natural History and Experimental Evidence of Prosociality.” Philosophical Transactions of the Royal Society B: Biological Sciences 365, no. 1553: 2723–2735. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Jaeggi, A. V. , De Groot E., Stevens J. M. G., and Van Schaik C. P.. 2013. “Mechanisms of Reciprocity in Primates: Testing for Short‐Term Contingency of Grooming and Food Sharing in Bonobos and Chimpanzees.” Evolution and Human Behavior 34, no. 2: 69–77. [Google Scholar]
  27. Jaeggi, A. V. , and Gurven M.. 2013. “Reciprocity Explains Food Sharing in Humans and Other Primates Independent of Kin Selection and Tolerated Scrounging: A Phylogenetic Meta‐Analysis.” Proceedings of the Royal Society B: Biological Sciences 280, no. 1768: 20131615. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Jaeggi, A. V. , Stevens J. M. G., and Van Schaik C. P.. 2010b. “Tolerant Food Sharing and Reciprocity Is Precluded by Despotism Among Bonobos but Not Chimpanzees.” American Journal of Physical Anthropology 143, no. 1: 41–51. [DOI] [PubMed] [Google Scholar]
  29. Jaeggi, A. V. , and Van Schaik C. P.. 2011. “The Evolution of Food Sharing in Primates.” Behavioral Ecology and Sociobiology 65, no. 11: 2125–2140. [Google Scholar]
  30. Judd, T. M. , and Sherman P. W.. 1996. “Naked Mole‐Rats Recruit Colony Mates to Food Sources.” Animal Behaviour 52, no. 5: 957–969. [Google Scholar]
  31. Kopp, K. S. , and Liebal K.. 2016. “Here You Are!—Selective and Active Food Sharing Within and Between Groups in Captive Sumatran Orangutans (Pongo abelii).” Behavioral Ecology and Sociobiology 70, no. 8: 1219–1233. [Google Scholar]
  32. Marshall‐Pescini, S. , Dale R., Quervel‐Chaumette M., and Range F.. 2016. “Critical Issues in Experimental Studies of Prosociality in Non‐Human Species.” Animal Cognition 19, no. 4: 679–705. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Massen, J. J. M. , Haley S. M., and Bugnyar T.. 2020. “Azure‐Winged Magpies' Decisions to Share Food Are Contingent on the Presence or Absence of Food for the Recipient.” Scientific Reports 10, no. 1: 16147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Massen, J. J. M. , Lambert M., Schiestl M., and Bugnyar T.. 2015. “Subadult Ravens Generally Don't Transfer Valuable Tokens to Conspecifics When There Is Nothing to Gain for Themselves.” Frontiers in Psychology 6: 885. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Melis, A. P. , Schneider A.‐C., and Tomasello M.. 2011. “Chimpanzees, Pan troglodytes, Share Food in the Same Way After Collaborative and Individual Food Acquisition.” Animal Behaviour 82, no. 3: 485–493. [Google Scholar]
  36. Nettelbeck, A. R. 1998. “Observations on Food Sharing in Wild Lar Gibbons (Hylobates lar).” Folia Primatologica 69, no. 6: 386–391. [DOI] [PubMed] [Google Scholar]
  37. Ostojić, L. , Legg E. W., Shaw R. C., Cheke L. G., Mendl M., and Clayton N. S.. 2014. “Can Male Eurasian Jays Disengage from Their Own Current Desire to Feed the Female What She Wants?” Biology Letters 10, no. 3: 20140042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Schwab, C. , Swoboda R., Kotrschal K., and Bugnyar T.. 2012. “Recipients Affect Prosocial and Altruistic Choices in Jackdaws, Corvus monedula .” PLoS ONE 7, no. 4: e34922. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Silk, J. B. , Brosnan S. F., Henrich J., Lambeth S. P., and Shapiro S.. 2013. “Chimpanzees Share Food for Many Reasons: The Role of Kinship, Reciprocity, Social Bonds and Harassment on Food Transfers.” Animal Behaviour 85, no. 5: 941–947. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Stevens, J. R. 2004. “The Selfish Nature of Generosity: Harassment and Food Sharing in Primates.” Proceedings of the Royal Society of London Series B: Biological Sciences 271, no. 1538: 451–456. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Stevens, J. R. , and Gilby I. C.. 2004. “A Conceptual Framework for Nonkin Food Sharing: Timing and Currency of Benefits.” Animal Behaviour 67: 603–614. [Google Scholar]
  42. Stevens, J. R. , and Stephens D. W.. 2002. “Food Sharing: A Model of Manipulation by Harassment.” Behavioral Ecology 13, no. 3: 393–400. [Google Scholar]
  43. Tennie, C. , Jensen K., and Call J.. 2016. “The Nature of Prosociality in Chimpanzees.” Nature Communications 7, no. 1: 13915. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Vahed, K. 1998. “The Function of Nuptial Feeding in Insects: A Review of Empirical Studies.” Biological Reviews 73, no. 1: 43–78. [Google Scholar]
  45. Wang, L. , Luo Y., Wang X., Maierdiyali A., Chang H., and Li Z.. 2019. “Azure‐Winged Magpies Solve String‐Pulling Tasks by Partial Understanding of the Physical Cognition.” Current Zoology 65, no. 4: 385–392. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Wilkinson, G. S. 1984. “Reciprocal Food Sharing in the Vampire Bat.” Nature 308, no. 5955: 181–184. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1 GLMM on Feeder Triggering Events

Table S2 GLMM on Feeder Type Preference

Table S3 CLMM on Order of Feeder Triggering

Table S4 Negative Binomial GLMM on Food‐Snatching Behavior.

INZ2-21-733-s001.doc (70.5KB, doc)

Data Availability Statement

The dataset and analysis script supporting the findings of this study are publicly available at [figshare] via [https://doi.org/10.6084/m9.figshare.29644037.v1]. Analyses can be reproduced using the R script food_sharing_analysis_script.R and dataset food_sharing.csv.


Articles from Integrative Zoology are provided here courtesy of Wiley

RESOURCES