Division of labor, characterized by within-individual task consistency and between-individual variability, is a fundamental aspect of social organization across various species, from eusocial insects to humans. Understanding the emergence of this division of labor is a central challenge in biology. Existing models often rely on pre-existing interindividual differences, such as response thresholds determining task engagement. Individuals with low thresholds readily undertake tasks, reducing the stimulus for others with higher thresholds to engage, thus leading to specialization. However, this approach doesn't explain the emergence of division of labor in homogeneous groups lacking pre-existing differences. This study addresses this gap by proposing a model based on the realistic assumption that individual nutrition levels fluctuate, triggering foraging when low. The model incorporates the common phenomenon of resource sharing within animal groups to investigate its role in the spontaneous emergence of division of labor in initially identical individuals. The research explores different resource-sharing scenarios to assess its effect on task specialization and the overall degree of division of labor.

The existing literature on division of labor largely focuses on response threshold models, which posit that individuals differ in their thresholds for initiating a task based on perceived stimuli. These models successfully explain task specialization when pre-existing differences in response thresholds exist. However, they fall short in explaining the emergence of division of labor from homogenous groups of initially identical individuals. Studies on social insects have highlighted the link between nutritional differences and division of labor, with foragers often exhibiting different nutritional states compared to nurses. This suggests a potential connection between resource allocation and task specialization. Other research explores the influence of group size on division of labor, with some suggesting an increase in specialization with larger groups. However, empirical evidence supporting this relationship is not entirely consistent. The role of resource sharing in promoting task specialization has been investigated in microbial systems, but these models often differ from animal group dynamics in terms of resource availability and sharing mechanisms. The present study aims to fill these gaps by focusing on the role of resource sharing in the self-organized emergence of division of labor in initially identical individuals.

The researchers developed an individual-based simulation model in continuous time. Each simulation comprised a group of N individuals (typically 100), each cycling between foraging and nursing (performing other tasks). Individuals possessed a nutrition level (n) ranging from 0 to Nmax (100). Simulations initiated with all individuals having an identical initial nutrition level (Ninit = 50). The nutrition level decreased over time at a metabolic rate (m) that differed depending on the task (mfor for foraging, mnur for nursing). Task choice was determined by an individual's perceived nutrition level relative to a critical threshold (μ = 50). To account for imperfect assessment, the perceived nutrition level was drawn from a normal distribution around the true level with a standard deviation (σ = 1). Foraging trips had a fixed duration (tfor = 5), after which individuals chose whether to forage again or nurse. Four resource-sharing scenarios were implemented: (1) No sharing; (2) Equal sharing between foragers and one nursing individual; (3) Dominance-based sharing, with resources allocated proportionally to pre-assigned dominance values; (4) Nutrition-based sharing, where dominance was determined by the individual's nutrition level. The degree of division of labor was quantified using the metric D from Duarte et al. (2012), ranging from -1 (strict alternation) to +1 (full specialization). The model was implemented in C++, analyzed and visualized in R using various packages.

The simulations revealed that resource sharing is sufficient for the emergence of division of labor, even in initially identical individuals. In the absence of resource sharing, individuals exhibited strict alternation between foraging and nursing (D = -1). With equal sharing, a significant degree of division of labor emerged (D > 0.5), indicating individuals specializing in either foraging or nursing for extended periods. Differences in metabolic rates between foraging and nursing reinforced division of labor. Lower metabolic rates during nursing resulted in stronger specialization, leading to bimodal distributions of nutrition levels. Unequal resource sharing, driven by dominance, also strongly promoted division of labor, leading to maximal levels of specialization. Interestingly, even when dominance was dynamically determined by nutritional status, strong division of labor emerged. The group size did not significantly affect the emergence of division of labor, although the degree of specialization could be influenced by small group size effects. Even scenarios with altruistic resource sharing to individuals close to foraging thresholds resulted in some degree of division of labor, albeit weaker.

The findings demonstrate that resource sharing, a common phenomenon in biological systems, is a sufficient mechanism for the self-organized emergence of division of labor, even without pre-existing individual differences. This contrasts with traditional response threshold models that assume stable interindividual differences throughout the lifetime. The model provides a mechanistic explanation for how the interaction between nutrition-driven foraging and resource sharing leads to task specialization. The results suggest that division of labor might arise from pre-existing mechanisms not specifically selected for regulating labor but for other purposes. This is a crucial point as it highlights that pre-existing interindividual differences are not strictly required for the emergence of division of labor; the feedback mechanism generated through resource sharing plays a significant role. The study challenges assumptions of traditional models and offers a framework for future experimental testing by focusing on evaluating whether division of labor arises from stable individual differences or feedback mechanisms influenced by parameters like nutrition levels.

This research demonstrates that resource sharing alone can generate a robust division of labor in groups of initially identical individuals. This mechanism is likely to be relevant across various biological systems, from animal groups to the evolution of multicellularity. Future research could explore the interplay between resource sharing, dominance hierarchies, and the evolution of cooperation in promoting division of labor, and examine the effects of varying resource availability and distribution patterns on the dynamics of task specialization.

The model employs several simplifying assumptions that could be explored in future research. The fixed durations of foraging and nursing trips and the simplified assessment of nutrition levels may not fully capture the complexity of real-world behaviors. The model also assumes a constant resource availability. Varying environmental conditions and resource scarcity might significantly affect the emergence and stability of division of labor. Further, the model doesn't incorporate explicit evolutionary dynamics, focusing solely on the self-organization aspect. Extending the model to incorporate evolutionary processes would provide further insights into the long-term consequences of resource sharing on division of labor.