The evolutionist approach to Cognitive Architectures

It has been said that a problem that is well formulated (in a certain language) also contains its solution (expressed in that same language). In this respect, probably the most important obstacle that AI research has been confronted with was the lack of a clearly-defined presentation of the problem that it tries to solve: specifically, although there have been many projects dealing with the field of AI in the past decades, most, if not all, of these projects have never been able to clearly formulate the problem they are trying to address, and thus it only came as natural that they haven't been able to devise a strategy to solve said problem.

Enter evolution.

An evolutionist approach to AI not only has the capability to clearly define an objective for the AI research problem, but it also naturally suggests the expansion of the AI research effort towards a more general goal, namely the development of a generic cognitive architecture. In an evolutionist approach, the problems that arise when trying to investigate the huge variety of cognitive architectures that can be found in the natural world can all be framed inside a coherent analytical strategy which studies the incremental adjustments that a cognitive architecture incorporates as it faces the myriad of adaptation problems it is confronted with.

Based on this kind of "evolutionist framing" approach, a crucial breakthrough can be achieved in the investigation methodology of the cognitive mechanisms, by following a clear research guideline which essentially consists of analyzing the evolutionist pressures that have sequentially shaped the various cognitive modules' architectures that can be found in the natural world, from simple insects - and below, all the way up to the modern human - and above. By systematically analyzing one such evolutionist pressure at a time, an evolving model of cognition might be developed by incremental additions of new modules that each answers one specific evolutionist pressure, thus effectively retracing the footsteps of evolution.