 |
Center for Nonlinear Studies |
This work focuses on the combination of laboratory experiments (lab) with human subjects and agent-based models (ABM). This combination has mutual advantages. Lab experiments are a powerful means to obtain well grounded information on the micro-foundations of complex social outcomes that facilitate cumulative scientific progress more than other standard qualitative or quantitative methods (Fehr and Gintis 2007). This is because, in lab experiments: (a) individuals are subjected to incentives, (b) they interact according to rules of the game that are clearly defined and easily understandable, (c) what is being observed is the concrete behavior of individuals and not a self-representation by the subject him/herself (as during an interview), (d) the presence of the observer less affects the behavior of individuals than in field observations, and (e) it is possible to manipulate certain theoretically crucial factors (e.g., incentives, rules of the game, interactions) and verify their consequences at a micro and macro level (e.g., Selten 1998). For these reasons, the lab can offer sound, clean and informative data on agents' behavior which is pivotal for evidence-based models. On the other hand, agent-based modeling complements the laboratory since it allows to explore complex interaction structures, external validity, and long time evolution which is impossible to investigate in the lab. Agent-based models can both allow for macro-level implication analyses of experimental evidence and provide new interpretations for the lab. More in details, considering that several techniques (e.g., symbolic regression – Genetic Programming, numerical regression, cluster analysis are at disposal for describing and classifying the behavior of experimental subjects in order to include such a knowledge in ABM, I will present the application of such different techniques to the data collected in a well well-known experiment (Andreoni 1995) of voluntary provision of public goods. Different techniques lead not only to different results but also to different kinds of behavioral descriptions, and consequently to different levels of analytical capabilities of ABM. Finally two ongoing research works combining lab behavioral knowledge and ABM will be briefly sketched in order to further point out the analytical power of the methodology on focus.
Host: Donatella Pasqualini