3Evolutionary Perspective on Social

Cognitions

Introduction

The term “social cognition” describes all abilities necessary to act I adequately in a social system.The understanding of a social environment) presupposes representations of the other agents as intentional and goal-l directed. Social cognition is a common skill among various species, howeverl we can observe distinct levels of complexity. As proposed by Tomasellol (2004), we can distinguish between the following degrees (sorted byj increasing sophistication):

• Dyadic engagement: Sharing behavior and emotions; Direct! interaction of two animate agents.

• Tryadic engagement: Sharing goals and perception; Some goalCaricaure of ChJles directed agents acting together towards some shared goal. Darwin himself

• Collaborative engagement: Joint intentions and attention; Two or more agents acting (in complementary roles), according to a coordinated action plan with mutual knowledge and the possibility of helping the other in his role.

The latter is considered to be the crucial difference between humans and all other species: Human beings possess a uniqe motivation to share psychological states with other persons.

When we describe psychological features of animals or humans, and we accept evolution as the process which designs everything in biosphere, the question arises whether psychological characteristics and their development are explainable in terms of genetic variation and natural selection. There are three main-theories of evolutionary development of the social structure of modern man:

1. group selection,

2. kin selection,

3. reciprocal altruism

From selection to sociality

While we behold development of life on earth, we soon take a evolutionary stance to explain the change of life over time. That means we try to reconstruct the way life developed on earth and also the modality of the processes that control these changes. In reference to social behaviour of creatures there are many evolutionary theories to explain this. As proposed by Gaulin and McBurney (see references below), we attend to three major theories: Group selection, an intuitive approach to describe social behaviour. Kin selection, a more advanced theory considering laws of genetics to cause social behaviour. Reciprocal alturism is a sophisticated approach to treat the individual in social barter-deals. These theories all have in common that the goal of the described agents is to pass on their genetic material into the next generations. And this common goal creates the social interaction we can observe.

Group Selection

Vero Wynne-Edwards (1906-1997) proclaimed this theory first in the 1960's. On an evolutionary perspective a group is a number of individuals who affect the fitness of each other. Notice that biological relatedness and periods of time are not taken into account with this definition. Group selection now means that if any of the individuals of a group is doing benefit to its group, the group is more likely to survive and pass its predisposition to the next generation. This again improves the chance of the individual to spread its genetic material. So in this theory an alturist is more likely to spread his alleles than a non-alturistic organism. The distinction to the „classical“ theory of evolution is that not only the fittest individuals are likely to survive, but also the fittest groups, so to speak the ones with the most cooperation.

Let's consider an example: Take some bacteria in the human mouth. These bacteria are very slow moving. The first group of bacteria, which perform cell division as fast as possible, soon waste all their nutriments and have no resources left. Instead, they have to face their bordering bacteria colonies. So the first group is facing death very fast.

The second group of bacteria, which perform a more moderate cell division, leave more resources to their bordering colonies. Whereas first group offspring always have to compete for the resources with its neighbours, the second groups offspring have savings of resources and so survives more likely a longer time period. The altruism of the second group makes it fitter as a whole. In the case of bacteria swimming in a growing medium, in which the bacteria can move freely, the fast group always overwhelms the second group. This indicates the problem of the group selection theory: it needs certain circumstances to describe things properly. Additionally, every theory about groups should include the phenomenon of migration. So in this simple form, the theory is not capable in handling selfish behaviour of some agents in altruistic groups: Alturistic groups which include selfish members would turn into pure selfish ones over time, because alturistic agents would work for selfish agents, thereby increasing the cheaters' fitness while decreasing their own. So generally, group selection is a poor explanation for altruism and sociality.

Kin Selection

A more sophisticated approach to explain cooperative behaviour is kin selection. Kin selection theory is embedded in the natural selection theory and the inclusive fitness theory (the later one arose from kin selection), which we do not inspect further here. William D. Hamilton and John M. Smith puplished the kin selection theory in 1964. The improvment compared to group selection is, that not the individuals are seen as actor, but the genes are identified as the actual „players“. The theory declares: An agent acts cooperatively, if the adressee is genetically related to him, because he wants the genetic material to spread which he has in common with the beneficiary. The more genetic material two agents have in common, the more they will cooperate. To help its own kin, an agent improves the chance its own genetic material to spread, at least the part he has in common with its kin.

As a rule of thumb we propose: "The relative benefit to the adressee of an altruistic action, has to be higher than the costs to the giver."

Hamiltons rule also considers the relatedness between giver and adressee of a favor, which he expressed in his famous formula:

r * B > C

where

r is the relatedness (value between zero and one) - it shows how akin the actor and the

beneficiary are;

B is the benefit to the beneficiary;

C are the costs to the giver

If the left value is greater than the costs of the right side, the behaviour increases the fitness of the giver and should therefore be done.

Reciprocal Alturism

In general reciprocal alturism is to do benefit to any organism in expectation of a reciprocation, so to speak a benefit pay-back. This behaviour establishes a system by which social interaction can be viewed as barter-deals. These barter-deals are concrete described by the prisoner's dilemma, which Albert W. Tucker formalized as follows:

Two suspects, A and B, are arrested by the police. The police have insufficient evidence for a conviction, and, having separated both prisoners, visit each of them to offer the same deal: if one testifies for the prosecution against the other and the other remains silent, the betrayer goes free and the silent accomplice receives the full 10-year sentence. If both stay silent, the police can sentence both prisoners to only six months in jail for a minor charge. If each betrays the other, each will receive a two-year sentence. Each prisoner must make the choice of whether to betray the other or to remain silent. However, neither prisoner knows for sure what choice the other prisoner will make. So the question this dilemma poses is: What will happen? How will the prisoners act?

Enhanced prisoner's dilemma looks like this:

Agents A and B making a barter-deal.

If A cheats on B, whereas B cooperates with A, B gets no points, A gets more than average

points.

If A cooperates with B and vise versa, both get average points.

If both cheat on each other, none of them gets points.

Iterated prisoner's dilemma reminds strongly on social interactions and many strategies for it were designed. A famous iterated prioner's dilemma strategie is tit for tat, which can be viewed as specification of reciprocal alturism. Among the „simple“ stategies, it is the most effective one. Tit for tat means if an organisms cooperation is reciprocated, it keeps the cooperation running, otherwise it stops cooperating with the organism, who did not cooperate. Because it is often hard to determine who is cooperating and who only pretends to be reliable, but actually acts selfish, organisms developed mechanisms to detect cheaters and selfish agents in altruistic groups in order to withhold favors from them. The development of cognitive abilities to detect cheaters is important for altruistic groups to survive - as we have seen, migration of selfish agents into altruistic groups infects it which in the end leads in dying out of the altruistic allele. Therefore, to keep altruism efficient and still benefit from its advantages, complex organism have developed various social cognitive abilities.

Possible selection pressures favoring human sociality

From an evolutionary perspective, sociality is another adaptation of some species to their environment - just like eyes, pelt, legs and the like. We know that adaptations are constantly tested against their environment, and that those traits (genes) who increase the fitness of the bearer are more likely to get passed on to the next generation. So what are the benefits and costs of living-within-a-group in general? One of the first principal costs is of course increased competition in almost any area of life. Animals that live solitary don't have to fight for rare resources such as foods or possible mates with companions. In addition, groups tend to attract more predators, as you can imagine. However, these two issues also have a positive side: A group tends to be more successful in finding food and is more powerful in defense against mentioned predators: More eyes/ears simply see/hear more than fewer do. This leads us to the first conclusion (which in principal can be generalized for all adaptations): Selection favors social habits if the benefits of living in a group outweights the costs.

Knowing that, what were the conditions of our ancestors in the Environment of Evolutionary Adaptedness (EEA)? Of course nobody can exactly say how they lived, but by considering life of contemporary great apes and hunter-gatherer societies, we can make a reasonable guess on which we can draw some conclusions: Presumably, early humans inhabited more open, less forested country than the majority of primates, which certainly increased their exposure to predators. Also, open habitats typically contain herds of grazing animals. Thus, if our ancestors' diet depended on meat, cooperative hunters maybe were more successful than solitary ones. Other food than meat also may support living in a group: If resources are generally rich, but scattered over the landscape, the costs of sharing the found food with others is low - while the benefits of getting help in finding them in the first place are very high. Another possible pressure was revealed by observing social great apes like gorillas: Neither their food comes in patches (they eat leaf material and do never hunt), nor have they natural enemies after reaching adulthood. It seems that the standard reasons for sociality have not much force on them -but one aspect of their behaviour gives a hint anyway: Killing among members of their own species. Especially female apes are at risk of having their infants killed by males that did not father those infants, so female gorillas cluster around strong, powerful males that are capable of protecting their children. Life of chimpanzees also includes fatal violence, concentrated onto struggles between communities: Males invade other groups to kill infants of unfamiliar females or even other males, if the numerical advantage is sufficient. These kinds of risk clearly favor sociality over living in solitude.

So we get a small overview of the probable pressures on our ancestors that lead to sociality:

• inhibiting open country,

• hunting,

• gathering rich but scattered patches of food,

• hostilities/infanticide among neighbours.

Social Cognition

Having talked about various processes and theories about evolution, the further discussion in this chapter will require detailed elaboration on the term 'social cognition'. While 'cognition' for itself is already a rather complex term, 'social cognition' is even more specific. In order to make it accessible and to understand the general idea behind it, several main aspects and the respective vocabulary to talk about it need to be introduced. The suggestions of Tomasello et al 2004 will serve as a basis for this.

First of all it has to be said that, when talking about 'social cognition', we talk about human cognitive skills, for, as we will see, humans are the only species abilities in the respective sense evolved. Trying to find the reasons for this fact will go hand in hand with inquiring into the nature of social cognition as a species unique human faculty.

Drawing a rough sketch of the topic, what we are interested in is the ability to participate with others in collaborative actions with not only common but also shared goals. The term introduced to describe this aspect is called shared intentionality. This presupposes the ability to read the intentions of other agents, as well as a broad background of cultural learning during the development of a human being. Especially the latter is a strictly human phenomenon, since it arises from a unique motivation to share psychological states, which in turn needs unique forms of cognitive representation for doing so. Hence it is proposed that the creation of linguistic symbols is closely related, as well as the rise of social norms and the establishment of social institutions.

The human faculty of social cognition

Around a child's first birthday the capability emerges to understand the intentions of other people. Humans are by far the most skillful species at reading the minds of specimen, i.e. successfully guessing or reasoning about what other fellow humans perceive, intend, believe, know, desire, etc. - which is crucial in order to decide what someone else is doing in the first place. While the action might be quite the same, it is the respective intention according to which the action has to be judged. For example, seeing somebody breaking a window has to be judged differently if somebody has just lost his keys, if he tries to break in somewhere or if someone simply exerts wanton destruction. While reading the intentions of another agent is not a strictly human capability, humans also collaborate and interact culturally, that means we have complex collaborative activities, shared symbolic artifacts and social institutions that allow for communication and structure, which lead to powerful abstract levels and organizational concepts like societies, states, etc., and give the possibility to convey knowledge on these levels from one generation to another, hereby creating a vast complexity and variety over historical time.

It is proposed that reading intentions and cultural learning give rise to species unique processes of cultural cognition and evolution.

In order to explain the level of complexity found in human cultural and collaborative activities, we need the term of shared intentionality. Apparently it is a strictly human faculty to participate in collaborative actions that involve shared goals and socially coordinated action plans, which is also called joint intentions. This requires an understanding of the goals and perceptions of other involved agents, as well as sharing and communicating these, which again seems to be a strictly human behavior. This then may have brought forth elaborate cognitive representations for dialogs, like the human faculty for language, mathematics and the creation of social practices and institutions. For further discussion, the questions of how humans come to understand intentional action and how they participate in actions born of shared intentionality will have to be dealt with.

Understanding intentional action

For our purposes, we will consider intentional action as an organism's intelligent behavioral interaction with its environment and the factors that play a role in that.

According to cybernetic theory this presupposes an organizational structure consisting of three components:

1. a goal,

2. perceiving the environment,

3. acting towards the goal by changing the environment.

This is a circular organization that causes action, which in turn causes a change in perception, which again determines the action. In this way, the system is self-regulating. This model is used by Tomasello et al to describe human intentional action.

Let's look at an example: A closed box which a person wants open. The latter obviously is the goal state, however, apparently it is necessary to distinguish between an external goal, which is the actual state of an open box in the environment, and an internal goal, which is something like a mental representation of the goal state, that the person needs to operate towards that goal.

Furthermore, a sharp distinction is drawn between a goal and an intention. Whereas a goal describes merely the state desired by the person, an intention includes this goal and the means to achieve it, that is a sort of action plan the person decided for to achieve the goal. In the given example, this might be cutting the box open with a sharp knife or scissors. As mentioned earlier, since an intention includes the goal, the same action can have different intentional interpretations. As we saw with breaking a car window for example, it can either be a means to get to the keys, or merely an expression of wanton destruction.

The results of the action are a change in the state of the environment, which, according to a person's internal goal, can be a failed attempt, a success, where the state of the environment matches the internal goal, or an accident, which is in a way a failed attempt with unpredicted consequences in the environmental state. Since we are talking about a human person, the results of an action are usually accompanied by an emotional reaction, like happiness, sadness, anger or surprise.

On the part of perception, the term 'selective attention' is introduced to point out that the monitoring of the environment done by the organism is focused on the goals at hand, which means that only those facts that are relevant for opening the box in the example before are taken into active consideration. The color of the box, for example, plays an unimportant role in the action plan chosen to open the box.

Depending of the complexity of the goal, the means may include the creation of subgoals, or even a whole subplan. Also, the organism probably wants the box open for a reason, which would be in the context a higher level goal. Concerning the intention involving this potential higher level goal, the act of opening the box may itself be just a subgoal. Choosing appropriate subgoals is referred to as 'decision making' by Tomasello et al.

Finally, the action itself might be the actual goal, consider for example the act of jogging, of dancing or of singing, where the respective action is not meant to bring about some immediate goal but represents by itself already the desired state.

These are all considerations that have to be taken into account in order to understand the intentional action of other organisms, which, as was pointed earlier, is a crucial point for social cognition. Referring to children's understanding of intentional action, the latter can be divided into three groups representing more and more complex level of grasp.

The first one to be mentioned is the perception of animate action. This means that after a couple of months, babies can differentiate between motion that was caused by some external influence to some passive object, and actions that an object or organism has performed by itself, as animate being. At this stage, however, the child has not yet any understanding of potential goals the observed actor might have.

The next stage of comprehension includes the understanding that the organism acts with persistence towards achieving a goal, including trial and error, and is developed by children after about 9 months. This also means that the child knows that the person it observes has a certain perception. At this stage, a certain amount of predicting the behavior of the actor is possible. After one year of age, a child understands that an actor pursuing a goal may have a variety of action plans to achieve the goal, and is choosing between them. Furthermore a certain sense for the selective attention of the actor will have developed, and the child realizes that action and attention are directed towards a goal. This allows a broad variety of predictions of behavior of organisms in a certain environment.

By 14 months of age, children fully comprehend intentional action, including the basics of rational decision making. According to Tomasello et al, this leads to powerful forms of cultural learning, which is especially important since a child not only learns to predict behavior in an environment, but it also learns, foremost by imitation, how things are conventionally done in their culture.

Shared intentionality

According to Tomasello et al, shared intentionality might emerge whenever socially interacting agents in an environment understand each other as acting intentionally. What this means is that the agents work together towards a shared goal in collaborative interaction. Furthermore, they do that in coordinated action roles and mutual knowledge about them. The nature of the activity or its complexity is not important, as long as the action is carried out in the described fashion. It is important to mention that the notion of 'shared goals' means that the internal goals of each agent include the intentions of the others. If you take a group of apes on a hunt, for example, the apes appear to be acting in a collaborative fashion, however, it is reasonable to assume that neither do they have coordinated action roles, nor do they have a shared goal, but rather act as seen fit towards the same individual goal state. Summing up, the important characteristics of the behavior in question are that the agents are mutually responsive, have the goal of achieving something together, and coordinate their actions with distributed roles and action plans.

Tomasello et al argue that in complex social groups the repeated sharing of intentions in a particular interactive context leads to the creation of habitual social practices and beliefs, that may form normative or structural aspects of a society, like government, money, marriage, etc., which of course form the notion of society itself. Society might hence be seen as a product and an indicator of social cognition.

The social interaction that builds the ground for activities involving shared intentionality is proposed to be divided into three groups:

The first one is called 'dyadic engagement'. What is meant here is the simple sharing of emotions and behavior, by means of a direct mutual responsiveness, for example by expressing emotions. The
motivation to share emotions, repeatedly, is already a key factor for social cognition and a main difference between humans and other species, as for example primates.

The next level is called 'triadic engagement', where two agents act together towards a shared goal, while monitoring the perception and goal-direction of the other agent. Sharing a goal is one step further than simply a direct responsiveness as in dyadic engagement.

The last supposed level is called 'collaborative engagement', which comprises, as introduced earlier, joint intentions and attention. At this point the agents share a goal, act in different, even complementary roles with a complex action plan and mutual knowledge about selective attention and intentions of one another. The latter aspect allows the agents to assist each other and reverse or take over roles.

These different levels of social engagement require the understanding of the different aspects of intentional action, as introduced above, and presuppose the uniquely human motivation to share psychological states with each other.

According to Tomasello et al, human infants develop very early in ontogeny the strong motivation to share emotions, goals and perception and participate in collaborative pretense activities in fictional environments.

The special motivation to share psychological states of course needs means to do so. These means have to be certain complex cognitive representations, especially for the joint intentions that require at least two sets of action plans, since in the spirit of shared goals those of the other one have to be represented as well for true shared intentionality. Since these representations have as content mostly social engagement, Tomasello et al make use of the term 'dialogic cognitive representations' at this point. Closely related with this is the communication and use of linguistic symbols. Dialogic cognitive representations allow in some sense a form of 'collective intentionality', which is important to construct social norms, conceptualize beliefs and, most importantly, share them. This gives rise to something like social rationality: by internalizing collective norms children learn to regulate their own behavior.

In this sense, social cognition is what enables us to create culture and lays the foundation for society. With this knowledge we can now return to the discussion of how and why this particular kind of human behavior may have evolved during evolution and in what way it is useful.

References

• M. S. Gazzaniga, R. B. Ivry, G. R. Mangun; Conitive Neuroscience, Norton & Company, 1998

• M. Tomasello; Understanding and sharing intentions: The origins of cultural cognition, Cambridge University Press, 2004

• Steven J. C. Gaulin, Donald H. McBurney; Psychology: An evolutionary approach, Prentice-Hall New Jersey, 2001

live version • discussion • edit lesson • comment • report an error