I have a slightly interesting point to make here. I once went about trying to prove the ideas of psychological egoism. It was earlier in my youth, and jaded philosophy (not that I have achieved great age or overcome all jadedness but still). I wanted to try and show repeated acts of selflessness could occur simply out of complex interactions of reward and punishment. So I turned to Artificial Intelligence.
Essentially I created a ?society? of very simple nodes that interacted with growing and eating food. The first thing to define real quickly is my reward and punishment. I defined two main sources for each, internal reward and punishment and external reward and punishment. Internal is what your mind and body do to you. I treated this as a very powerful motivating force, but it could be tempered by external reward and punishment as well. The source of this external reward/punishment was other nodes. If enough nodes attempted rewarded or punished you for an action it could override internal drives. Thus external realities could create ?selfish? selfless acts.
The society was essentially based around food. The idea was simple, any given node could perform three actions; it could eat some food out of a central area, water and care for the food in the area, or do nothing. Every cycle used up energy for the nodes and if there energy went to zero they died off. Eating provided energy, water and caring increased overall food stores, nothing did nothing. The less available energy a node had the stronger internal reward or punishment a node received for its actions. So a node low on energy that did not eat during a cycle received a high internal punishment, or vice versa. Other nodes in the vicinity of a given node could observe behavior and offer either complaints or praise that acted on the overall motivation of the given node. The lower the food supplies for the whole group the more likely and stronger the condemnation for eating or praise for water. The total external and internal weighted reward/punishment lead to the nodes actions. So the nodes actions were motivated purely by efforts to maximize its overall reward and minimize punishment. My idea was that altruistic acts, watering when personal energy was low, would become common in this environment yet any given node was still motivated by selfish desires. One added piece I did was allow additional nodes to be created during high times of surplus, these next generation nodes inherited the group average tendency to perform a given action, so that essentially actions were teachable to next generation nodes.
Well all the technical jargon behind, I discovered what I thought were some interesting things. What I found was that a society based on these rules almost always fell apart with in a few generations. The next step I did was randomize all the initial conditions and allowed the program to repeatedly attempt to create a stable society, where nodes could maintain adequate food supplies and most nodes survived multiple iteration cycles. What I found after 5 days of crunching was that the only society that could ever become stable were societies where nodes had an initial propensity for performing selfless acts regardless of reward/punishment.
So I pretty much disproved my initial hypothesis. Whatever application this might have to humans, I leave open to personal review.
|