Artificial Social Life

views updated


Most demographic research either develops or uses some kind of theory or model: for instance, a theory of fertility or a model of the class system. Generally, such theories are stated in discursive English, although sometimes the theory is represented as an equation (for example, in regression analysis). In the 1990s researchers began to explore the possibilities of expressing theories as computer programs. The advantage is that social processes can then be simulated in the computer and in some circumstances it is even possible to carry out "experiments" on artificial social systems that would otherwise be quite impossible.

Although the simulation of social dynamics has a long history in the social sciences, the advent of much more powerful computers, more powerful computer languages, and the greater availability of data have led to increased interest in simulation as a method for developing and testing social theories.

The logic underlying the methodology of simulation is not very different from the logic underlying statistical modeling. In both cases, a model is constructed (for example, in the form of a computer program or a regression equation) through a process of abstraction from what are theorized to be the actually existing social processes. The model is then used to generate expected values that are compared with empirical data. The main difference between statistical modeling and simulation is that the simulation model can itself be "run" to produce output, while a statistical model requires a statistical analysis program to generate expected values.

Advantages of Simulation

Paradoxically, one of the main advantages of simulation is that it is hard to do. To create a useful simulation model, its theoretical presuppositions need to have been thought through with great clarity. Every relationship to be modeled has to be specified exactly and every parameter has to be given a value, for otherwise it will be impossible to run the simulation. This discipline means that it is impossible to be vague about what is being assumed. It also means that the model is potentially open to inspection by other researchers in all its detail. These benefits of clarity and precision also have disadvantages, however. Simulations of complex social processes involve the estimation of many parameters and adequate data for making the estimates can be difficult to come by.

Another benefit of simulation is that it can, in some circumstances, give insights into the "emergence" of macro level phenomena from micro level action. For example, a simulation of interacting individuals may reveal clear patterns of influence when examined on a societal scale. A simulation by Andrzej Nowak and Bibb Latané (1993), for example, shows how simple rules about the way in which one individual influences another's attitudes can yield results about attitude change at the level of a society, and a simulation by Robert Axelrod (1995) demonstrates how patterns of political domination can arise from a few rules followed by simulated nation-states.

Agent-Based Simulation

The field of social simulation has come to be dominated by an approach called agent-based simulation (alternatively called multi-agent simulation). Although other types of simulation such as those based on system dynamics models (using sets of difference equations) and microsimulation (based on the simulated aging of a survey sample to learn about its characteristics in the future) are still undertaken, most simulation research now uses agents.

Agents are computer programs (or parts of programs) that are designed to act relatively autonomously within a simulated environment. An agent can represent an individual or an organization, according to what is being modeled. Agents are generally programmed to be able to "perceive" and "react" to their situation, to pursue the goals they are given, and to interact with other agents, for example by sending them messages. Agents are generally created using an object-oriented programming language and are constructed using collections of condition–action rules. The agent examines its rules to identify those whose conditions hold true in its current situation and then executes ("fires") the actions determined by just those rules. The effect of firing the rules will normally be to alter the agent's situation, and thus in the next cycle a different set of rules will fire.

Agent-based models have been used to investigate the bases of leadership, the functions of norms, the implications of environmental change on organizations, the effects of land-use planning constraints on populations, the evolution of language, and many other topics. Examples of research can be found in the Journal of Artificial Societies and Social Simulation.

While most agent-based simulations have been created to model real social phenomena, it is also possible to model situations that could not exist in our world, in order to understand whether there are universal constraints on the possibility of social life (for example, can societies function if their members are entirely self-interested and rational?). These are at one end of a spectrum of simulations ranging from those of entirely imaginary societies to those that aim to reproduce specific settings in great detail.

An interesting variant on agent-based modeling is to include people in place of some or all of the computational agents. This transforms the model into a type of multiplayer computer game, which can be valuable for allowing the players to learn more about the dynamics of some social setting (for example, business students can be given a game of this type in order to learn about the effects of business strategies). Such games are known as participatory simulations.

The Potential of Simulation

Although computer simulation can be regarded as simply an another method for representing models of social processes, it encourages a theoretical perspective which emphasizes emergence, the search for simple regularities that give rise to complex phenomena, and an evolutionary view of the development of societies. This perspective has connections with complexity theory, an attempt to locate general principles applying to all systems which show autonomous behavior–including not only human societies, but also biological and physical phenomena.

See also: Simulation Models.


Axelrod, Robert. 1995. "A Model of the Emergence of New Political Actors." In Artificial Societies: The Computer Simulation of Social Life, ed. Nigel Gilbert and Rosaria Conte. London: UCL.

Carley, Kathleen, and Michael Prietula. 1994. Computational Organization Theory. Hillsdale, NJ: Lawrence Erlbaum.

Epstein, Joshua M., and Robert Axtell. 1996. Growing Artificial Societies: Social Science from the Bottom Up. Cambridge, MA: MIT Press.

Gilbert, Nigel. 1999. "Computer Simulation in the Social Sciences." Special issue of American Behavioral Scientist 42.

Gilbert, Nigel, and Klaus G. Troitzsch. 1999. Simulation for the Social Scientist. Milton Keynes, Eng.: Open University Press.

Nowak, Andrzej, and Bibb Latané. 1993. "Simulating the Emergence of Social Order from Individual Behaviour." In Simulating Societies: The Computer Simulation of Social Phenomena, ed. Nigel Gilbert and Jim Doran. London: UCL Press.

internet resource.

Journal of Artificial Societies and Social Simulation. <>.

Nigel Gilbert