In contemporary evolutionary biology, an organism is said to behave altruistically when its behavior benefits other organisms, at a cost to itself. The costs and benefits are measured in terms of reproductive fitness or expected number of offspring. So by behaving altruistically, an individual reduces the number of offspring it is likely to produce by itself, but increases the number that other individuals are likely to produce. Eusociality can be considered an extreme form of altruism in animal communities. This is the highest level of social organization in animals. To be considered eusocial, an animal society should meet the following criteria: reproductive altruism (which involves reproductive division of labor and cooperative alloparental brood care), overlap of adult generations, and permanent (lifelong) philopatry. Eusociality was firstly described in social insects, and later discovered in several other organisms including eusocial rodents (several rodent species of the African family Bathyergidae) and shrimps. The next level in the hierarchy of social organizations in animals is cooperative (communal) breeding. This level corresponds to “semisociality” in a classification system of social levels originally suggested by C. D. Michener (1969) and then developed by E. O.Wilson in his Sociobiology: A New Synthesis (1975). As distinct from eusociality which is based on irreversible determination of sterile and fertile casts, cooperative breeding is less rigorous. It refers to a breeding system in which individuals other than parents (“helpers”) behave altruistically providing additional care for offspring. Many species possess more flexible types of social organization than eusociality and semisociality. However, some social systems are based on facultative division of labor and on temporal limits on breeding for some members of communities.
Altruistic behavior in animal communities is based, to a greater or lesser extent, on the division of roles between individuals depending on their behavioral, cognitive, and social specialization. Social specialization is connected with social roles and tasks performed by community members. Behavioral specialization can be expressed in differences in diets, techniques of getting food, selective reactions to certain stimuli, escaping predators, nestling, and so on. Relatively stable groups can exist in populations that differ by complexes of behavioral characteristics (Bolnick et al. 2003). Some specimens can possess complex behavioral repertoires which enable them to learn pretty fast and effectively within a specific domain. This ability can be called individual cognitive specialization. Cognitive specialization in animal communities is based on the inherited ability of some individuals to form associations between some stimuli easier than between other stimuli and thus more readily learn certain behaviors. Altruistic behavior in animals does not necessarily rely on intelligence and the ability to learn. However, the presence of “cognitive specialists” facilitates the tuning of integrative reactions of a whole animal community to unpredictable influences in its changeable environment.
Altruistic behavior of animals is still enigmatic for evolutionary biologists in many aspects. Charles Darwin famously developed a group-selection explanation for the apparent self-sacrificing behavior of neuter insects; however, he found this phenomenon difficult to explain within the frame of his theory of evolution by natural selection. Analysis of these problems became possible on the basis of ideas of gene dominance and fitness outlined by R. Fisher (1930). J. B. Haldane (1932) suggested that an individual’s genes can be multiplied in a population even if that individual never reproduces, providing its actions favor the differential survival and reproduction of collateral relatives. In the 1960s and 1970s, two theories emerged which tried to explain evolution of altruistic behavior: kin selection (or inclusive fitness) theory, due to W. Hamilton (1964), and the theory of reciprocal altruism, due primarily to R. L. Trivers (1979) and J. Maynard Smith (1974). The theory of reciprocal altruism is an attempt to explain the evolution of altruism among non-kin. The basic idea is straightforward: it may benefit an animal to behave altruistically toward another, if there is an expectation of the favor being returned in the future: “If you scratch my back, I’ll scratch yours” principle. Whereas “kin altruism” is based on animals’ ability to recognize relatives and to adjust their behavior on the basis of kinship, reciprocal altruism requires certain cognitive prerequisites, and among them the ability to recognize community members and to keep in mind aftereffects of repeated interactions. A good example here is the finding by G. S. Wilkinson (1984) of blood sharing in vampire bats, which is based on partner fidelity among non-kin individuals. In primates, there is experimental evidence that reciprocal altruism relies on sophisticated cognitive abilities that make current behavior contingent upon a history of interaction and calculation of mutual rewards and punishes (deWaal 2000).
Both kin altruism and non-kin altruism in animal societies are based on the division of roles and thus on great individual variability that includes behavioral, cognitive, and social specialization. It is worth noting that in many eusocial species social specialization is based on the division of roles between members of morphologically distinct castes. For example, termites, social aphids, social shrimps, naked mole rats, and some ants produce special casts of soldiers. In contrast, cognitive specialization is based on intricate distinction between individuals reflected in their learning abilities rather than on morphological and physiological traits.
There are several variants of division of social roles in animal communities, from division of labor in kin groups to a thin balance between altruism and “parasitism” within groups of genetically unrelated individuals. Task allocation in animal communities can impose restrictions on the display of intelligence by the members. For instance, eusocial rodents, termites, and ants condemned to digging or babysitting or suicide defending cannot forage, scout, or transfer pieces of information. Furthermore, subordinate members of cooperatively breeding communities sacrifice their energy and possibly cognitive skills to dominating individuals, serving as helpers or even as sterile workers.
Cognitive specialization in animal communities is based on the ability of some individuals to learn faster within specific domains. In eusocial animals, cognitive specialization between groups of sterile workers can serve for the maintenance of colony integrity. For example, in Myrmica ants, some members of a colony learn to catch difficult-to-handle prey much easier and earlier in the course of the ontogenetic development than others do. These individuals can serve as “etalons” for those members of communities that possess poorer skills and can learn from others by means of social learning (Reznikova and Panteleeva 2008). In several highly social ant species (such as red wood ants), a rare case of cognitive specialization between team members have been described (Ryabko and Reznikova 2009). There are stable teams within ants’ colony each containing one scout and 4–8 foragers. Only scouts are able to solve complex problems and pass information to other team members. For instance, scouts memorize and transfer the information about a sequence of turns toward a goal; they also can perform simple arithmetic operations. Such feats of intelligence cannot be performed by the foragers. Cognitive specialization within ants’ team very much increase effectiveness of solving problems while searching for food (Reznikova 2007). In ant species with low level of social organization, specialization does not predict individual efficiency. Surprisingly, little is known about cognitive specialization in other eusocial organisms, although there is some evidence of great differences in cognitive abilities between individuals in some eusocial species. For example, in honeybees, a few active foragers in a hive can solve problems demanding abstraction and classificatory abilities at a similar level with monkeys and dogs (Mazokhin-Porshnyakov 1969), and in naked mole rats, some colony members use tools while gnawing on substrates (Shuster and Sherman 1998).
In cooperatively breeding animals, members of a society sacrifice their specific behavioral and cognitive abilities to provide food and protection for the few reproductive community members and their offspring. Social groups can be made up of individuals who specialize in certain helping behaviors or those who perform a number of behaviors to differing degrees. For example, in a gregarious bird, noisy minor Manorina melanocephala, a considerable number of subordinates that were never seen to provision the young, help intensively with predator mobbing. Furthermore, bad provisioners contribute more to mobbing than good provisioners (Arnold et al. 2005). In meerkats, there is a high variation between helpers in provisioning rates as well as in their exploratory activity. Meerkats exhibit teaching of prey-handling skills and social learning of the use of new landmarks, so individual variability in learning capacities of helpers influences the prosperity of a group (Clutton-Brock 2002; Thornton and Malapert 2009). Both in cooperative breeders and in highly social species with more flexible breeding systems, behavioral and cognitive specialization is tightly connected with division of labor during joint actions. For example, cooperative hunting is based on clearly coordinated actions of individuals which are specialized on different tasks demanding different behavioral peculiarities and cognitive skills. Such division of labor includes “flush and ambush” strategies in Harris’ hawks, “driving and blocking” subtasks in chimpanzees, “center and wing” roles in lionesses, “driver and barriers” subtasks in bottlenose dolphins, and so on (see Gazda et al. 2005, for a review).
In virtually all cooperative species, there are large individual differences in altruistic behavior, the causes and consequences of which remain poorly understood despite 30 years of research on the evolution of cooperation. To date, very few studies have used experiments to test the role of cognitive specialization in integrity of animal communities. The intriguing problem is how behavioral and cognitive flexibility interacts with inherited propensities of community members. As altruistic behavior in many species is based on social specialization and division of labor within communities, it is likely that cognitive specialization manifests itself only within distinct strategies. It is still an open question whether clearly distinct behavioral strategies used by community members are based on an evolutionary stable package of features or they are based on flexible decision making. For example, “professional specialization” in cooperatively hunting mammals, such as wolves, wild dogs, and lions, seems rather flexible. However, it is still enigmatic if there is room for intelligence, or cognitive specialization in these and many other situations is based on a high level of inherited predisposition. A hypothesis which explains how community members can learn efficiently complex forms of behavior, based on behavior fragments that they already have in their repertoire, was suggested by Reznikova and Panteleeva (2008). It could be adaptive for members of different species to have dormant “sketches” of complex behavioral patterns being implemented on several carriers and then distributed by means of social learning. The authors call this “distributed social learning” because fragments of useful behavioral programs are distributed among members of a population and remain cryptic until appropriate changes in the environment occur, such as climate changes or appearance of new abundant prey, or new predators, and so on. Indeed, it could be rather costly for animal brains to be equipped with complex stereotypes for all possible vital situations. Propagation of complex stereotypes, new for certain populations, is based on relatively simple forms of social learning which underlies species’ predisposition to learn certain behaviors and does not require feats of intelligence from animals. This hypothesis has been experimentally tested on ants, and there is much to be done for investigating how it can work in vertebrates. There remains an explanatory gap between the growing body of data on displaying cognitive specialization in highly social species and our understanding of possible relations between altruistic behavior and cognitive specialization.