Interaction among autonomous decision-makers is usually modelled in economi
cs in game-theoretic terms or within the framework of General Equilibrium.
Game-theoretic and General Equilibrium models deal almost exclusively with
the existence of equilibria and do nor analyse the processes which might le
ad to them. Even when existence proofs can be given, two questions are stil
l open. The first concerns the possibility of multiple equilibria, which ga
me theory has shown to be the case even in very simple models and which mak
es the outcome of interaction unpredictable. The second relates to the comp
utability and complexity of the decision procedures which agents should ado
pt and questions the possibility of reaching an equilibrium by means of an
algorithmically implementable strategy. Some theorems have recently proved
that in many economically relevant problems equilibria are not computable.
A different approach to the problem of strategic interaction is a "construc
tivist" one. Such a perspective, instead of being based upon an axiomatic v
iew of human behaviour grounded on the principle of optimisation, focuses o
n algorithmically implementable "satisfycing" decision procedures. Once the
axiomatic approach has been abandoned, decision procedures cannot be deduc
ed from rationality assumptions, but must be the evolving outcome of a proc
ess of learning and adaptation to the particular environment in which the d
ecision must be made. This paper considers one of the most recently propose
d adaptive learning models: Genetic Programming and applies it to one the m
ostly studied and still controversial economic interaction environment, tha
t of oligopolistic markets. Genetic Programming evolves decision procedures
, represented by elements in the space of functions, balancing the exploita
tion of knowledge previously obtained with the search of more productive pr
ocedures. The results obtained are consistent with the evidence from the ob
servation of the behaviour of real economic agents.