Social groups
Degrees and qualities of sociality
We’ve previously described populations as groups of conspecific individuals living in the same geographical area (which we could define by a minimal gene flow measure). We also evidenced various types of interactions between these species.
When members of a same population have repeated interactions that exhibit a significant mutualist ("win-win") facet, they are said to engage in cooperative behavior. The set of individual is then called a social group and the network of cooperative behaviors is the social structure that governs it.
Social groups exist with extremely wide variations across a range of key facet:
Group size: from pairs to millions.
Cohesiveness: the degree to which group members remain in spatial proximity.
Connectedness: the density of social interactions (who interacts with whom, and how often).
Permeability: the ease with which one can enter or leave the group.
Compartmentalization: the degree to which a group is divided into subgroups, castes, or roles.
Differentiation: the extent to which individuals within the group differ in behavior, morphology, or reproductive status.
Communication: the richness, bandwidth, and modality of information exchange.
Let’s describe the gradation in sociality that can be observed in the animal world.
Solitary
animals have no recurring interaction with others with the exception of courtship, mating, and other reproduction-linked behaviour, which are generally considered to be outside the scope of sociality, since they are shared by all animals. Solitary species includes reptiles like turtles or snakes, as well as many fish, amphibians and insects.
Gregarious
species tend to form loose, temporary aggregations, like a school of fish, a flock of starlings or a swarm of locusts. That behavior is cooperative since every individual benefits from the pattern, and arises from simple self-organization, based on local interactions, with any overview of the whole group. The typical benefits are information sharing, safety in numbers, warmth through huddling. Yet, gregariousness is generally not considered as a mark of proper sociality, since it lacks the hallmark of parental care and communal living.
Presocial
structures include a wide range from the most basic forms of sociality to the most complex, falling just short of the eusociality of ants or bees.
Presociality’s evolutionary history seems to have followed two routes:
parental care has evolved independently dozens of time. With enormous variations in the intensity and duration of care, it is characterized by the investment of time and energy in the survival and maturation of offspring.
Species that engage in parental care but not in any other form of social interactions (like the jaguar) are called parasocial.communal species cooperate in the building and sharing of their nests, thus reducing predation risk and making thermoregulation cheaper.
Species that engagein communal cooperation but nothing else are on the subsocial route.
From thoses bases, three hallmarks of sociality can be the majorx defining features of more advances states:
Cooperative brood care, the participation by adults to the rearing of younglings who are not their own.
Overlap between generations, the fact that parents live long enough, not only to see their offspring to adulthood, but to contribute, as distinct generation of adults, to the group’s labor.
Reproductive division of labor, the division of a group between sexual castes, some of which do not attempt to reproduce.
When one of these three hallmarks is present, the group’s structure is called quasisocial. When two are present it is called semisocial.
Eusocial
structure is achieved when all three aforementioned hallmarks are present.
This structure is exemplified in familiar insect colonies (ants, termites, most bees and wasps) and, more anecdotally, a few mammals (mole-rat species).
A colony of ants blurs the line between a collection of organisms and a multicellular one: individuals are born into specialized castes (foragers, gardeners, soldiers,...), and only a few reproduce. It bears mentioning that those castes are not genetically determined but rather developmentally enforced. Individuals with near-identical genomes turn completely distinct morphologically and behaviorally from their diet (quality and quantity) and other environmental factors. They can also follow a "career", holding different roles as they age, following a pre-set partner. This system is much more flexible in the short term than pure genome-adjusting sexual selection: the ratio of castes can be adjusted quickly based on colony need, as expressed via pheromonal feedback more soldiers under predator pressure, more minor workers when food is abundant, etc.
The evolutionary success of eusociality is unquestionably immense. A single termite colony can count dozens of million individuals and cover 50,000 square meters. The aggregate biomass from eusocial insects represents 30 to 40% of the terrestrial animal biomass.
Determinants of sociality
Ecological pressure
As for other forms of behavior, the most immediate drivers for social behavior are the selective forces imposed by the ecosystem.
Predation is perhaps the most potent single driver of sociality. Group living reduces predation risk through several mechanisms:
The dilution of risk through group size.
The vigilance from multiple indivuals.
Collective defense through mobbing, alarm, deterrence,...
The confusion of a predator’s targeting.
Food distribution acts in a different way. When food is abundant and homogeneously distributed, solitary foraging may be optimal. When it is clumped but unpredictably located, the cost of sharing with a group is offset by the benefit of large searching parties. When food requires a cooperative effort to obtain (like large prey), group hunting evolves.
Habitat structure also influences sociality. Open environments like the savanna force species to put more premium on vigilance, since individuals are very exposed visually. Complex habitats like dense jungles can also decrease social cohesion and make small groups more optimal.
Population density acts as a form of auto-constraint. It is of course downstream of some of the aforementioned environmental factors (food sources, predators...) as well as intrinsic characteristics (mobility, sexual behavior...). In turn, it determines social behavior as a major input of the cost-benefit analysis between competition and cooperation.
The typical pattern is that of a dial: low-density populations are more solitary, as self-reliance and seclusion are encouraged; beyond a certain density threshold, the cost of defending one’s territory from peers becomes overwhelming and social strategies emerge as viable options; but when density reaches extreme levels, stress, conflict from myriad interactions, and hierarchical imbalances lead to social breakdown, ranging from simple group fission to general dispersion.
Phylogenetic inertia
A given ecological context could be matched with an optimal social organization from an adaptive perspective. But all species carry ancestral traits as a result of their evolutionary history. The extent to which they resist further evolutionary change is called phylogenetic inertia, and constitutes another prime determinant of sociality.
Inherited traits that translate into social blueprints define strong boundaries for ecological influence: regardless of environmental pressures, great apes never cease to live together and jaguars do not band into herds...
Often the success or failure of a social mechanism hinges on a genetic preadaptation, a previously existing trait that is functional in a different, perhaps insignificant, context, but lays dormant as a potential stepping stone to a new social adaptation.
Relentless selection will then ensure that where some specific social organization is best adapted to a certain environment, some species will likely fill that niche. But most likely, this will not be due to genotype evolution (which deploys over longer timeframes) nor to a "fair" behavioral competition between neighboring populations. Successful social mechanisms are often born at the nexus of specific environmental pressures and pre-existing behavioral patterns or abilities.
Supporting this view, it has been observed that sociality is reversible. When conditions change, social breakdown can occur. The enabling traits stay, but serve different purposes.
Communication
One key enabling ability for all forms of sociality is communication. The flip side of animals having sensory perceptions is the ability to produce the type of signals others can perceive. Obviously, most animals can see and hear, but their mere existence alsomeans they can also be seen and be heard by others. Often, the stricture of a species’ environment have led to strategies that minimize this sensory footprint (camouflage, stealth...). But another strand of evolutionary pressure has been conducive to voluntary signals, and the phenomenon of communication.
Communication can be a solitary exercise, with autocommunication. Bats emit sound that bounce off surfaces and give them information about their surroundings. This monologue is a form of communication because the signal comes back encoded with environmental data.
Whenever it involves multiple individuals, the ability to communicate becomes the first, and most fundamental, expression of sociality in animals.
The modes of communication simply mirror the senses we have described:
Chemical communication: the emission of pheromones, which have the advantage of lingering in the environment and of an extremely wide lexicon, but the drawback of slow and relatively localized transmission.
Visual communication, whose asset is a variety of modes like color, pattern, shape, movement as well as speed. The drawback is the directionality and sensitivity to light conditions (a free line of sight and some amount of ambient light is required).
Acoustic communication, which cumulates versatility and speed (sound is much slower than light, but at the scales at which animals operate, it is still conveyed quasi instantaneously) with resilience (sounds propagates around obstacles, is indifferent to day and night, etc.).
Mechanic communication takes touching as its main form. Compared to the others, it is of course characterized by its local aspect, but makes up for it with the vast expanse of sensory reception areas. An animal sees with its eyes and hears with is hears, but can be touched anywhere on its body, which expands the expressive universe of tactile communication in a different way from previous examples.
Natural selection and altruism
The problem of altruism has been a central theoretical challenge, since Darwin noted that the existence of sterile worker insects seemed to pose a fatal objection to his theory. Besides that extreme example, we are familiar with examples of altruistic behavior in the animal world: "soldiers" sacrificing themselves to protect the group, "betas" accepting to eschew reproduction, etc.
The development of behavioral studies saw ethologists routinely resort to the idea that animals acted for the good of the species, or of the group. In the wrong hands, this is anthropomorphic fallacy; but even more rigorous thinkers seemed to fall back on ths idea, ignoring its frontal contradiction with Darwin’s ideas.
The implication of such an assumption is that the unit of selection in the evolution process is not the organism, as thought by Darwin, but the species, or perhaps that both co-exist to varying degrees over time and taxa. This simplistic idea did not withstand scrutiny: the selection mechanism which groups are subject to simply do not cut deep enough to circumvent the brutality of individual selection: maladapted organisms die early, and their genes vanish from the pool.
The breakthrough came with the theory of kin selection. It predicts that altruism should be preferentially directed toward close relatives, and that the willingness to help should scale with the degree of relatedness. A behavior’s cost to oneself can then be outweighed by a benefit to others, pro rata to their genetic proximity.
These predictions are broadly confirmed by numerous cases of kin-preference in social mammals. More spectacularly, kin selection got strong support from the supreme social set, the eusocial insects, due to one of their peculiarities: haplodiploidy. This means that males carry only one set of chromosomes, while females carry pairs. The effect is that full sisters share 75% while a mother shares only 50% with their offspring. Kin selection theory would then entail that sisters derive more benefit from helping their mother make more sisters than to produce their own younglings. This looks startingly like a stepping stone to reproducive castes in eusocial organizations!
Originally, kin selection was strongly associated with a framework where genes are the unit of selection. Conceptual debates are far form settled yet. There is no question that kin selection is empirically verified but has also been shown to be compatible, or even equivalent, with models of multi-level selection, which introduce more complex systems where units of selection co-exist at different levels, typically: gene, organism, kin group, wider group, species, etc.
Social structures
Group size
is not arbitrary; it reflects an optimum between the benefits and costs of adding members, depending on combinations of environmental factors and intrinsic ones (intragroup competition, disease transmission, information...). It is of course adjustable, via fusion-fission mechanisms; the degree of adjustability itself is an adaptive feature in relation to environment seasonality.
For the most cognitively sophisticated social species, chimpanzees being the canonical example, fission and fusion are the default modes. A community of on hundred of these apes can share a territory, but at any given moment it is fragmented into small, fluid parties of two to ten, with ever-shifting membership. Spider monkeys, spotted hyenas, bottlenose dolphins, and elephants follow similar patterns. Such a regime places extraordinary demands on social cognition: each animal must track the current location and recent behavior of dozens of associates who are often out of sight, remember who was with whom, who fought and who reconciled, who shared food and who stole, and update those records continuously.
Roles and castes
denote the behavioral specialization of individuals. In eusocial insects, this is predicated on morphological caste differentiation during development (queens, workers, soldiers look very different). In vertebrate societies, roles are more flexible and subject to potential rotation: a sentinel can watch for predators while others forage, an alpha can mediate disputes, etc.
Dominance hierarchies
represent important instances of role structures: they are rankings that determine priority of access to resources (food, mates, resting sites). They can be linear (pecking orders) or more complex (with coalitions, tree structures, context-dependence...). Dominance is typically established through aggressive encounters but maintained, once established, through ritualized signals (threats, submissive postures). The stability of a dominance hierarchy represents an optimum between the benefits of reducing conflict when every individual accept their ranks, and the chance at better outcomes if the structure is upended. Among mammals, troops, bands, packs, pods, herds and other groups types abound, representing. awide array of dominace structures. Gorilla groups center on a single silverback presiding over unrelated females and their offspring; wolves and wild dogs organize around a single breeding pair flanked by related helpers; elephant families are headed by a matriarch whose accumulated knowledge is a critical resource. Some species push this further into multi-level groups, with social organization at several nested tiers. The primage gelada navigates a four-tiered system: a one-male unit (a male, his several females, and their offspring), several units associating into a team, teams clustering into bands sharing a home range, and bands congregating into herds of up to one thousand individuals at sleeping cliffs. These structures are not confined to primates: elephants and dolphins form alliances at multiple levels, too. Individuals must steer through overlapping social contexts and adjust behavior accordingly.
Aggression
in social animals is not pathological; it is a regulated, functional behavior that mediates competition for resources. We can distinguish between contest competition (direct, aggressive, in which the winner monopolizes the resource) and scramble competition (indirect, in which all have access to the resource but see their share decrease as more competitors enter the fray). Resources are often tied to geography, and territoriality, the active defense of a defined area, is a major driver of agression. The spatial configuration of territories reflects the distribution of resources and the intensity of competition. Territories can be assigned to solitary animals as well as social groups.
The flip side of aggression is conflict management: any group in which individuals interact repeatedly must contain aggression lest it destroys the cooperative fabric that makes group living valuable. The first line of defense is ritualized aggression. When aggression does occur, many species engage in reconciliation: post-conflict affiliative contact between former opponents. A related phenomenon is consolation: a bystander approaching and affiliating with the victim rather than the aggressor. Policing is a more proactive form of conflict management: for instance, the alpha male can intervene impartially in disputes between subordinates, breaking up fights and restoring peace.
Such patterns suggest a rudimentary sense of social order, an expectation about how group members should behave toward one another.
Mating systems
are thought to be determined by differential parental investment. The sex that invests more in each offspring (usually the female, through gestation, lactation, or egg production) becomes the limiting resource for the other sex. The less-investing sex (usually the male, then) competes for access to the more-investing sex, leading to sexual selection for traits that increase mating success. This produces the familiar patterns of sexual dimorphism: males that are larger, more ornamented, more aggressive, or more conspicuous than females.
The ratio of investing to competing individuals determines the mating system:
When both sexes invest approximately equally, monogamy is favored.
When one sex invests much more than the other, the less-investing sex competes intensely and polygamy results.
In any case, the mating pair constitutes the simplest social unit of all. In many species, the bond is fleeting — a brief encounter, then separation — but in others, it persists through a breeding season, or for life. The natural extension of the pair bond is the family group, parents and their dependent offspring, which manifest when some degree of care is displayed.
Parental care
is another major axis of variation in social systems. We have already noted that the investment from the two parents can be markedly different, and that in some social structures, brood care could become communal, involving many more individuals.
Since time and energy are finite, there is also a trade-off is between the number of offspring and the intensity of parental care. Some species produce large numbers of eggs and rely on the random survival of some individuals until adulthood. Others choose heavy parental care, with the aim of securing a high probability of survival for a few younglings. Such species are associated with larger body sizes, slower developments, and extended periods of juvenile dependence, all of which is thought to enable deeper learning and the honing of complex mental or behavioral patterns.