Re this and other recent threads . . . . I may be speaking up where I
should not . . . . but since "agents" generally correspond to humans in the
real world, I wonder if there has been interest in examining evolutionary
psychology's basis to build agents that capture more features of humans?
Evolutionary psychology (being an offshoot of evolutionary biology) takes
Hamilton's view that we are the result of a lot of gene selection for
"inclusive fitness." Thus economic modeling picks up only part of what
motivates people. From the genes' viewpoint, what is important is
replication, by ourselves, by our children, by our close relatives. We
thus evolved psychological traits in the ancestral environment of small
tribes to seek high status (among other things).
Today this is expressed when a lawyer gives up much higher pay to become a
judge, swapping economic considerations for higher social status. The fact
that this has no effect on the number of children he has only means that
evolution hasn't had time to "realize" that we don't live in small tribes
(where high status got you an extra wife or two) any more.
While there are multiple feedback levels involved, evolution is so slow we
can treat the psychological traits based on it as constant, but we do need
to recognize 1) that they exist, and 2) they were adaptive in a world
largely gone, 3) they may *not* be adaptive for the genes of related groups
or the genes of individuals who no longer live in tribes as
hunter-gatherers. For example, the psychological traits leading to war
were seriously maladaptive for Southwest American corn farmers (circa 1250
CE) while they were not for hunter-gatherers.
Evolved psychological traits almost certainly depend on a large number of
genes and thus have a bell curve distribution, even before they are further
spread out by experience. This is fortunate, otherwise every lawyer would
be striving equally hard to become a judge. :-)
Keith Henson
PS, there is more on this subject if list readers are interested.
|