Evolution of self-organized division of labor in a response threshold model

Shift in distribution of division of labour in chronically stressed honeybee colonies after perturbation

Division of labour (DOL) in eusocial insects plays an important role in colony fitness. Honeybees face a variety of stressors that compromise the homeostasis of the colony and reduce survival and reproduction. Considering the significance of DOL in colony homeostasis, it is important to understand whether and how DOL may be altered as a result of chronic stress. Therefore, we tested whether honeybee colonies shift DOL in response to high infestation with the parasitic mite Varroa destructor. For this, we monitored chronically stressed and presumably low-stress colonies from April till December 2022. During the experiment, we applied a cold shock to test whether a perturbation resulted in a larger alteration in DOL in chronically stressed colonies. We found that after cold shock, there was a lower proportion of nurses in the chronically stressed colonies. For foragers, we found higher activity post-cold shock in chronically stressed colonies, but no difference between treatments in nectar inflow, suggesting less efficient foragers. Furthermore, we found that there was an accelerated task switch in chronically stressed colonies after the cold shock. The large changes after the perturbation may indicate inefficient task allocation due to chronic stress. Our study contributes to the understanding of social resilience and chronic stress responses in eusocial animals.

Origin of division of labor is decoupled from polymorphism in colonial animals

Division of labor, the specialization of sometimes phenotypically divergent cell types or group members, is often associated with ecological success in eukaryotic colonial organisms. Despite its many independent evolutionary origins, how division of labor emerges remains unclear. Conventional hypotheses tend toward an "economic" model, so that biological division of labor may reflect a partitioning of preexisting tasks and morphologies into specialized colony members. Here, we present an alternative model of the origin of division of labor, which can explain the evolution of new functions within a colony. We show that in colonies of the Cretaceous aged (103-96 Ma) fossil bryozoan of the genus Wilbertopora, the first cheilostome bryozoan to evolve polymorphism, preexisting morphologies were not simply partitioned among new members, but instead expanded into novel morphospace as they lost functions, specifically feeding. This expansion occurred primarily during two pulses of heightened morphological disparity, suggesting that the evolution of polymorphism corresponded to relaxed constraints on morphology and perhaps to the exploration of novel functions. Using a simple model of physiological connections, we show that regardless of the functionality of these new colony members, all nonfeeding members could have been supported by neighboring feeding members. This suggests that geometric constraints and physiological connectedness could be prerequisites for evolving both polymorphism and division of labor in modular organisms, and that a classic partitioning model of specialization cannot be broadly applied to biological systems.

A soluble model for synchronized rhythmic activity in ant colonies

Synchronization is one of the most striking instances of collective behavior, occurring in many natural phenomena. For example, in some ant species, ants are inactive within the nest most of the time, but their bursts of activity are highly synchronized and involve the entire nest population. Here we revisit a simulation model that generates this synchronized rhythmic activity through autocatalytic behavior, i.e., active ants can activate inactive ants, followed by a period of rest. We derive a set of delay differential equations that provide an accurate description of the simulations for large ant colonies. Analysis of the fixed-point solutions, complemented by numerical integration of the equations, indicates the existence of stable limit-cycle solutions when the rest period is greater than a threshold and the event of spontaneous activation of inactive ants is very unlikely, so that most of the arousal of ants is done by active ants. Furthermore, we argue that the persistent oscillations observed in the simulations for colonies of finite size are due to resonant amplification of demographic noise.

Resource adaptation drives the size-complexity rule in termites

The size-complexity rule posits that the evolution of larger cooperative groups should favour more division of labour. Examples include more cell types in larger multicellular organisms, and more polymorphic castes in larger eusocial colonies. However, a correlation between division of labour and group size may reflect a shared response of both traits to resource availability and/or profitability. Here, this possibility was addressed by investigating the evolution of sterile caste number (worker and soldier morphotypes) in termites, a major clade of eusocial insects in which the drivers of caste polymorphism are poorly understood. A novel dataset on 90 termite species was compiled from the published literature. The analysis showed that sterile caste number did increase markedly with colony size. However, after controlling for resource adaptations and phylogeny, there was no evidence for this relationship. Rather, sterile caste number increased with increasing nest-food separation and decreased with soil-feeding, through changes in worker (but not soldier) morphotype number. Further, colony size increased with nest-food separation, thus driving the false correlation between sterile caste number and colony size. These findings support adaptation to higher energy acquisition as key to the rise of complex insect societies, with larger size being a by-product.

Social Learning versus Individual Learning in the Division of Labour

Division of labour, or the differentiation of the individuals in a collective across tasks, is a fundamental aspect of social organisations, such as social insect colonies. It allows for efficient resource use and improves the chances of survival for the entire collective. The emergence of large inactive groups of individuals in insect colonies sometimes referred to as laziness, has been a puzzling and hotly debated division-of-labour phenomenon in recent years that is counter to the intuitive notion of effectiveness. It has previously been shown that inactivity can be explained as a by-product of social learning without the need to invoke an adaptive function. While highlighting an interesting and important possibility, this explanation is limited because it is not yet clear whether the relevant aspects of colony life are governed by social learning. In this paper, we explore the two fundamental types of behavioural adaptation that can lead to a division of labour, individual learning and social learning. We find that inactivity can just as well emerge from individual learning alone. We compare the behavioural dynamics in various environmental settings under the social and individual learning assumptions, respectively. We present individual-based simulations backed up by analytic theory, focusing on adaptive dynamics for the social paradigm and cross-learning for the individual paradigm. We find that individual learning can induce the same behavioural patterns previously observed for social learning. This is important for the study of the collective behaviour of social insects because individual learning is a firmly established paradigm of behaviour learning in their colonies. Beyond the study of inactivity, in particular, the insight that both modes of learning can lead to the same patterns of behaviour opens new pathways to approach the study of emergent patterns of collective behaviour from a more generalised perspective.

Evolution of self-organised division of labour driven by stigmergy in leaf-cutter ants

Social insects owe their widespread success to their ability to efficiently coordinate behaviour to carry out complex tasks. Several leaf-cutter ant species employ an advanced type of division of labour known as task partitioning, where the task of retrieving leaves is distributed between workers that cut and drop and those that collect the fallen leaves. It is not entirely clear how such highly coordinated behaviour can evolve, as it would seem to require the simultaneous mutations of multiple traits during the same generation. Here, we use an agent-based simulation model to show how task partitioning in leaf-cutter ants can gradually evolve by exploiting stigmergy (indirect coordination through the environment) through gravity (leaves falling from the treetop on the ground forming a cache). Our simple model allows independent variation in two core behavioural dimensions: the tendency to drop leaves and the tendency to pick up dropped leaves. Task partitioning readily evolves even under these minimal assumptions through adaptation to an arboreal environment where traveling up and down the tree is costly. Additionally, we analyse ant movement dynamics to demonstrate how the ants achieve efficient task allocation through task switching and negative feedback control.

Resource sharing is sufficient for the emergence of division of labour

Division of labour occurs in a broad range of organisms. Yet, how division of labour can emerge in the absence of pre-existing interindividual differences is poorly understood. Using a simple but realistic model, we show that in a group of initially identical individuals, division of labour emerges spontaneously if returning foragers share part of their resources with other group members. In the absence of resource sharing, individuals follow an activity schedule of alternating between foraging and other tasks. If non-foraging individuals are fed by other individuals, their alternating activity schedule becomes interrupted, leading to task specialisation and the emergence of division of labour. Furthermore, nutritional differences between individuals reinforce division of labour. Such differences can be caused by increased metabolic rates during foraging or by dominance interactions during resource sharing. Our model proposes a plausible mechanism for the self-organised emergence of division of labour in animal groups of initially identical individuals. This mechanism could also play a role for the emergence of division of labour during the major evolutionary transitions to eusociality and multicellularity.

Social complexity and brain evolution: insights from ant neuroarchitecture and genomics

Brain evolution is hypothesized to be driven by requirements to adaptively respond to environmental cues and social signals. Diverse models describe how sociality may have influenced eusocial insect-brain evolution, but specific impacts of social organization and other selective forces on brain architecture have been difficult to distinguish. Here, we evaluate predictions derived from and/or inferences made by models of social organization concerning the effects of individual and collective behavior on brain size, structure, and function using results of neuroanatomical and genomic studies. In contrast to the predictions of some models, we find that worker brains in socially complex species have great behavioral and cognitive capacity. We also find that colony size, the evolution of worker physical castes, and task specialization affect brain size and mosaicism, supporting the idea that sensory, processing and motor requirements for behavioral performance select for adaptive allometries of functionally specialized brain centers. We review available transcriptomic and comparative genomic studies seeking to elucidate the molecular pathways functionally associated with social life and the genetic changes that occurred during the evolution of social complexity. We discuss ways forward, using comparative neuroanatomy, transcriptomics, and comparative genomics, to distinguish among multiple alternative explanations for the relationship between the evolution of neural systems and social complexity.

Deterministic Response Threshold Models of Reproductive Division of Labor Are More Robust Than Probabilistic Models in Artificial Ants

We implement an agent-based simulation of the response threshold model of reproductive division of labor. Ants in our simulation must perform two tasks in their environment: forage and reproduce. The colony is capable of allocating ant resources to these roles using different division of labor strategies via genetic architectures and plasticity mechanisms. We find that the deterministic allocation strategy of the response threshold model is more robust than the probabilistic allocation strategy. The deterministic allocation strategy is also capable of evolving complex solutions to colony problems like niche construction and recovery from the loss of the breeding caste. In addition, plasticity mechanisms had both positive and negative influence on the emergence of reproductive division of labor. The combination of plasticity mechanisms has an additive and sometimes emergent impact.

Individual experience influences reconstruction of division of labour under colony disturbance in a queenless ant species

BACKGROUND

Division of labour (DOL) is ubiquitous across biological hierarchies. In eusocial insects, DOL is often characterized by age-related task allocation, but workers can flexibly change their tasks, allowing for DOL reconstruction in fluctuating environments. Behavioural change driven by individual experience is regarded as a key to understanding this task flexibility. However, experimental evidence for the influence of individual experience is remains sparse. Here we tested the effect of individual experience on task choice in the queenless ponerine ant, Diacamma cf. indicum from Japan.

RESULTS

We confirmed that both nurses and foragers shifted to vacant tasks when the colony composition was biased to one or the other. We also found that nurses which are induced to forage readily revert to nursing when reintroduced into balanced colonies. In contrast, foragers which are induced to revert to nursing very rarely return to a foraging role, even 19 days post reintroduction to their original colony.

CONCLUSIONS

Taken together, our results suggest that individual experience decreases the response threshold of original foragers, as they continue to be specialist nurses in a disturbed colony. However, original nurses do not appear strongly affected by having forager experience and revert to being nurses. Therefore, while individual experience does have an effect, other factors, such as reproductive ability, are clearly required to understand DOL maintenance in fluctuating environments.

The evolution of mechanisms to produce phenotypic heterogeneity in microorganisms

In bacteria and other microorganisms, the cells within a population often show extreme phenotypic variation. Different species use different mechanisms to determine how distinct phenotypes are allocated between individuals, including coordinated, random, and genetic determination. However, it is not clear if this diversity in mechanisms is adaptive-arising because different mechanisms are favoured in different environments-or is merely the result of non-adaptive artifacts of evolution. We use theoretical models to analyse the relative advantages of the two dominant mechanisms to divide labour between reproductives and helpers in microorganisms. We show that coordinated specialisation is more likely to evolve over random specialisation in well-mixed groups when: (i) social groups are small; (ii) helping is more "essential"; and (iii) there is a low metabolic cost to coordination. We find analogous results when we allow for spatial structure with a more detailed model of cellular filaments. More generally, this work shows how diversity in the mechanisms to produce phenotypic heterogeneity could have arisen as adaptations to different environments.

The evolution of division of labour in structured and unstructured groups

Recent theory has overturned the assumption that accelerating returns from individual specialisation are required to favour the evolution of division of labour. Yanni et al., 2020, showed that topologically constrained groups, where cells cooperate with only direct neighbours such as for filaments or branching growths, can evolve a reproductive division of labour even with diminishing returns from individual specialisation. We develop a conceptual framework and specific models to investigate the factors that can favour the initial evolution of reproductive division of labour. We find that selection for division of labour in topologically constrained groups: (1) is not a single mechanism to favour division of labour-depending upon details of the group structure, division of labour can be favoured for different reasons; (2) always involves an efficiency benefit at the level of group fitness; and (3) requires a mechanism of coordination to determine which individuals perform which tasks. Given that such coordination must evolve prior to or concurrently with division of labour, this could limit the extent to which topological constraints favoured the initial evolution of division of labour. We conclude by suggesting experimental designs that could determine why division of labour is favoured in the natural world.

The foraging gene as a modulator of division of labour in social insects

The social ants, bees, wasps, and termites include some of the most ecologically-successful groups of animal species. Their dominance in most terrestrial environments is attributed to their social lifestyle, which enable their colonies to exploit environmental resources with remarkable efficiency. One key attribute of social insect colonies is the division of labour that emerges among the sterile workers, which represent the majority of colony members. Studies of the mechanisms that drive division of labour systems across diverse social species have provided fundamental insights into the developmental, physiological, molecular, and genomic processes that regulate sociality, and the possible genetic routes that may have led to its evolution from a solitary ancestor. Here we specifically discuss the conserved role of the foraging gene, which encodes a cGMP-dependent protein kinase (PKG). Originally identified as a behaviourally polymorphic gene that drives alternative foraging strategies in the fruit fly Drosophila melanogaster, changes in foraging expression and kinase activity were later shown to play a key role in the division of labour in diverse social insect species as well. In particular, foraging appears to regulate worker transitions between behavioural tasks and specific behavioural traits associated with morphological castes. Although the specific neuroethological role of foraging in the insect brain remains mostly unknown, studies in genetically tractable insect species indicate that PKG signalling plays a conserved role in the neuronal plasticity of sensory, cognitive and motor functions, which underlie behaviours relevant to division of labour, including appetitive learning, aggression, stress response, phototaxis, and the response to pheromones.

Self-organization of organoids from endoderm-derived cells

Organoids constitute biological systems which are used to model organ development, homeostasis, regeneration, and disease in vitro and hold promise for use in therapy. Reflecting in vivo development, organoids form from tissue cells or pluripotent stem cells. Cues provided from the media and individual cells promote self-organization of these uniform starting cells into a structure, with emergent differentiated cells, morphology, and often functionality that resemble the tissue of origin. Therefore, organoids provide a complement to two-dimensional in vitro culture and in vivo animal models of development, providing the experimental control and flexibility of in vitro methods with the three-dimensional context of in vivo models, with fewer ethical restraints than human or animal work. However, using organoids, we are only just beginning to understand on the cellular level how the external conditions and signaling between individual cells promote the emergence of cells and structures. In this review, we focus specifically on organoids derived from endodermal tissues: the starting conditions of the cells, signaling mechanisms, and external media that allow the emergence of higher order self-organization.

Evolution of division of labour in mutualistic symbiosis

Mutualistic symbiosis can be regarded as interspecific division of labour, which can improve the productivity of metabolites and services but deteriorate the ability to live without partners. Interestingly, even in environmentally acquired symbiosis, involved species often rely exclusively on the partners despite the lethal risk of missing partners. To examine this paradoxical evolution, we explored the coevolutionary dynamics in symbiotic species for the amount of investment in producing their essential metabolites, which symbiotic species can share. Our study has shown that, even if obtaining partners is difficult, 'perfect division of labour' (PDL) can be maintained evolutionarily, where each species perfectly specializes in producing one of the essential metabolites so that every member entirely depends on the others for survival, i.e. in exchange for losing the ability of living alone. Moreover, the coevolutionary dynamics shows multistability with other states including a state without any specialization. It can cause evolutionary hysteresis: once PDL has been achieved evolutionarily when obtaining partners was relatively easy, it is not reverted even if obtaining partners becomes difficult later. Our study suggests that obligate mutualism with a high degree of mutual specialization can evolve and be maintained easier than previously thought.

Modern termites inherited the potential of collective construction from their common ancestor

Animal collective behaviors give rise to various spatial patterns, such as the nests of social insects. These structures are built by individuals following a simple set of rules, slightly varying within and among species, to produce a large diversity of shapes. However, little is known about the origin and evolution of the behavioral mechanisms regulating nest structures. In this study, we discuss the perspective of inferring the evolution of collective behaviors behind pattern formations using a phylogenetic framework. We review the collective behaviors that can be described by a single set of behavioral rules, and for which variations of the environmental and behavioral parameter values produce diverse patterns. We propose that this mechanism could be at the origin of the pattern diversity observed among related species, and that, when they are placed in the proper conditions, species have the behavioral potential to form patterns observed in related species. The comparative analysis of shelter tube construction by lower termites is consistent with this hypothesis. Although the use of shelter tubes in natural conditions is variable among species, most modern species have the potential to build them, suggesting that the behavioral rules for shelter tube construction evolved once in the common ancestor of modern termites. Our study emphasizes that comparative studies of behavioral rules have the potential to shed light on the evolution of collective behaviors.

Task allocation in a cooperative society: specialized castes or age-dependent switching among ant workers

Some ant species have multiple worker castes that differ in body size; workers in one caste remain in the colony and those in the other forage outside the colony (caste polyethism). In other species, all workers engage in both tasks, but the younger workers remain in the colony and the older workers forage (age polyethism). Here, we ask which of these two is the most efficient for colony level performance when foragers suffer a higher daily mortality than workers in the colony and when the optimal worker size differs between two tasks. We studied two models: in the stationary colony model, the colony size and composition remain constant, and the amount of excess resources that can be used for producing reproductive individuals is maximized; in the growing colony model, all of the resources obtained are used for producing new workers, and the rate of the colony growth is maximized. In both models, we observed similar results: caste polyethism is more advantageous than age polyethism if the difference in mortality between the two tasks is small and the difference in the optimal size is large. In the opposite situation, the age polyethism is more advantageous.

A computational model of task allocation in social insects: ecology and interactions alone can drive specialisation

Social insects allocate their workforce in a decentralised fashion, addressing multiple tasks and responding effectively to environmental changes. This process is fundamental to their ecological success, but the mechanisms behind it are not well understood. While most models focus on internal and individual factors, empirical evidence highlights the importance of ecology and social interactions. To address this gap, we propose a game theoretical model of task allocation. Our main findings are twofold: Firstly, the specialisation emerging from self-organised task allocation can be largely determined by the ecology. Weakly specialised colonies in which all individuals perform more than one task emerge when foraging is cheap; in contrast, harsher environments with high foraging costs lead to strong specialisation in which each individual fully engages in a single task. Secondly, social interactions lead to important differences in dynamic environments. Colonies whose individuals rely on their own experience are predicted to be more flexible when dealing with change than colonies relying on social information. We also find that, counter to intuition, strongly specialised colonies may perform suboptimally, whereas the group performance of weakly specialised colonies approaches optimality. Our simulation results fully agree with the predictions of the mathematical model for the regions where the latter is analytically tractable. Our results are useful in framing relevant and important empirical questions, where ecology and interactions are key elements of hypotheses and predictions.

Combining Multi-Agent Systems and Subjective Logic to Develop Decision Support Systems

Nowadays, the rise of the interconnected computer networks and the increase of processed data have led to producing distributed systems. These systems usually separate multiple tasks into other simpler with the goal of maintaining efficiency. This paradigm has been observed for a long time in different animal organisations as insect colonies and fish shoals. For this reason, distributed systems that emulate the biological rules that govern their collective behaviour have been developed. Multi-Agent Systems (MAS) have shown their ability to address this issue. This paper proposes Ant Colony based Architecture with Subjective Logic (ACA-SL). It is a bio-inspired model based on ant colony structures. It makes use of MAS to distribute tasks and Subjective Logic (SL) to produce Decision Support Systems (DSS) according to the combination of individual opinions. A system implementation based on the proposed architecture has been generated to illustrate the viability of the proposal. The proposed architecture is intended to be the starting point for developing systems that solve a variety of problems.

Costs of task allocation with local feedback: Effects of colony size and extra workers in social insects and other multi-agent systems

Adaptive collective systems are common in biology and beyond. Typically, such systems require a task allocation algorithm: a mechanism or rule-set by which individuals select particular roles. Here we study the performance of such task allocation mechanisms measured in terms of the time for individuals to allocate to tasks. We ask: (1) Is task allocation fundamentally difficult, and thus costly? (2) Does the performance of task allocation mechanisms depend on the number of individuals? And (3) what other parameters may affect their efficiency? We use techniques from distributed computing theory to develop a model of a social insect colony, where workers have to be allocated to a set of tasks; however, our model is generalizable to other systems. We show, first, that the ability of workers to quickly assess demand for work in tasks they are not currently engaged in crucially affects whether task allocation is quickly achieved or not. This indicates that in social insect tasks such as thermoregulation, where temperature may provide a global and near instantaneous stimulus to measure the need for cooling, for example, it should be easy to match the number of workers to the need for work. In other tasks, such as nest repair, it may be impossible for workers not directly at the work site to know that this task needs more workers. We argue that this affects whether task allocation mechanisms are under strong selection. Second, we show that colony size does not affect task allocation performance under our assumptions. This implies that when effects of colony size are found, they are not inherent in the process of task allocation itself, but due to processes not modeled here, such as higher variation in task demand for smaller colonies, benefits of specialized workers, or constant overhead costs. Third, we show that the ratio of the number of available workers to the workload crucially affects performance. Thus, workers in excess of those needed to complete all tasks improve task allocation performance. This provides a potential explanation for the phenomenon that social insect colonies commonly contain inactive workers: these may be a 'surplus' set of workers that improves colony function by speeding up optimal allocation of workers to tasks. Overall our study shows how limitations at the individual level can affect group level outcomes, and suggests new hypotheses that can be explored empirically.

A theoretical approach to the size-complexity rule

The so-called size-complexity rule claims the existence of a positive correlation between organism size and number of cell types. In this spirit, here we address the relationship between organism size and number of potential tasks that can be performed. The modeling relies on the assumption that the states of the cells within the aggregates are such that the maximum fitness is realized, but also relies on the existence of tradeoffs among the distinct functions. For group sizes larger than the number of potential tasks, fitness maximization is attained when all cells in group specialize in a given task. Under this scenario, the number of potential tasks equals the number of cell types. We have found that the morphology and the topology of aggregates, as well as the developmental mode, strongly influence the dynamics of body formation. Particularly, it has been observed that more compact structures, such as sphere-like structures, are more likely to follow the claim of the size-complexity rule, whereas more fragile structures such as linear chains, which are more vulnerable to drastic changes due to division mechanisms, can, in a broad scenario, violate the size-complexity rule.

Task switching is associated with temporal delays in Temnothorax rugatulus ants

The major evolutionary transitions often result in reorganization of biological systems, and a component of such reorganization is that individuals within the system specialize on performing certain tasks, resulting in a division of labor. Although the traditional benefit of division of labor is thought to be a gain in work efficiency, one alternative benefit of specialization is avoiding temporal delays associated with switching tasks. While models have demonstrated that costs of task switching can drive the evolution of division of labor, little empirical support exists for this hypothesis. We tested whether there were task-switching costs in Temnothorax rugatulus. We recorded the behavior of every individual in 44 colonies and used this dataset to identify each instance where an individual performed a task, spent time in the interval (i.e., inactive, wandering inside, and self-grooming), and then performed a task again. We compared the interval time where an individual switched task type between that first and second bout of work to instances where an individual performed the same type of work in both bouts. In certain cases, we find that the interval time was significantly shorter if individuals repeated the same task. We find this time cost for switching to a new behavior in all active worker groups, that is, independently of worker specialization. These results suggest that task-switching costs may select for behavioral specialization.

A Design Pattern for Decentralised Decision Making

The engineering of large-scale decentralised systems requires sound methodologies to guarantee the attainment of the desired macroscopic system-level behaviour given the microscopic individual-level implementation. While a general-purpose methodology is currently out of reach, specific solutions can be given to broad classes of problems by means of well-conceived design patterns. We propose a design pattern for collective decision making grounded on experimental/theoretical studies of the nest-site selection behaviour observed in honeybee swarms (Apis mellifera). The way in which honeybee swarms arrive at consensus is fairly well-understood at the macroscopic level. We provide formal guidelines for the microscopic implementation of collective decisions to quantitatively match the macroscopic predictions. We discuss implementation strategies based on both homogeneous and heterogeneous multiagent systems, and we provide means to deal with spatial and topological factors that have a bearing on the micro-macro link. Finally, we exploit the design pattern in two case studies that showcase the viability of the approach. Besides engineering, such a design pattern can prove useful for a deeper understanding of decision making in natural systems thanks to the inclusion of individual heterogeneities and spatial factors, which are often disregarded in theoretical modelling.

Evolution of Self-Organized Task Specialization in Robot Swarms

Division of labor is ubiquitous in biological systems, as evidenced by various forms of complex task specialization observed in both animal societies and multicellular organisms. Although clearly adaptive, the way in which division of labor first evolved remains enigmatic, as it requires the simultaneous co-occurrence of several complex traits to achieve the required degree of coordination. Recently, evolutionary swarm robotics has emerged as an excellent test bed to study the evolution of coordinated group-level behavior. Here we use this framework for the first time to study the evolutionary origin of behavioral task specialization among groups of identical robots. The scenario we study involves an advanced form of division of labor, common in insect societies and known as "task partitioning", whereby two sets of tasks have to be carried out in sequence by different individuals. Our results show that task partitioning is favored whenever the environment has features that, when exploited, reduce switching costs and increase the net efficiency of the group, and that an optimal mix of task specialists is achieved most readily when the behavioral repertoires aimed at carrying out the different subtasks are available as pre-adapted building blocks. Nevertheless, we also show for the first time that self-organized task specialization could be evolved entirely from scratch, starting only from basic, low-level behavioral primitives, using a nature-inspired evolutionary method known as Grammatical Evolution. Remarkably, division of labor was achieved merely by selecting on overall group performance, and without providing any prior information on how the global object retrieval task was best divided into smaller subtasks. We discuss the potential of our method for engineering adaptively behaving robot swarms and interpret our results in relation to the likely path that nature took to evolve complex sociality and task specialization.