ABSTRACT
As we have seen in earlier cross-culture economic experiments (Section 17.7), culture can matter for human decision-making and behavior. Broadly interpreted, culture is a set of routines, rules, norms, routines, and beliefs that will govern the behavioral mode of human agents. Geert Hofstede, well known for his proposed five dimensions of culture, defines culture as “the collective programming of the mind that distinguishes the members of one group of people from another” (Hofstede, 2001, p. 9; emphasis added). With this understanding, it is not surprising to expect research efforts attempting to integrate culture into models of software agents at the micro level, i.e., to build culturally sensitive agents. The aim is to map each identified culture to a set of behavioral rules, for example, decision tress (Section 12.10). Hence, different cultures are associated with different behavioral rules. Culture here is taken as exogenously given. A further attempt is, therefore, to endogenize it so as to examine how culture may arise, develop, and evolve through time. To differentiate these two different attempts, we shall call the former models exogenous agent-based cultural models and the latter endogenous.
