On Jan 29, 12:20 am, "THE BORG" outthere.com> wrote:

Game theorists have devised a prisoner's dilemma game for the specific
purpose of analyzing cooperation versus selfishness in social
interactions. The game derives its name from a scenario in which two
suspects of a crime are being interrogated in separate rooms by the
police. Neither knows what the other is going to do and has a choice
either of turning in his accomplice or of confessing to the crime
himself. In the language of the prisoner's dilemma game, ratting out
the accomplice is referred to as "cheating." Refusing to implicate the
accomplice is "cooperating."

Imagine a hypothetical scenario in which one suspect cooperates and
the other cheats. The cheat gets off scot-free in return for his
testimony that is used to convict his accomplice. Given the compelling
case that can be made against him using the cheat's testimony, the
cooperator receives a heavy prison sentence, say twenty years. If both
suspects cheat, both will be found guilty, but they will receive a
lighter sentence in return for testifying against the other, say five
years. If both suspects cooperate, the police have very little usable
evidence and can convict them only on a minor offense that gets both a
year in jail.

In this particular example of the prisoner's dilemma, cooperation
produces the best result in terms of the total number of years served
in prison. When both suspects cheat on their accomplice, they go to
jail for five years, but when both cooperate, they only get one year
each. From the point of view of the individual, however, cheating is
the better strategy, regardless of whether the other suspect cheats or
cooperates. If the other cooperates and I cheat, then I am home and
dry, no sentence. If the other cheats, then I should cheat also
because this will get me a five-year sentence rather than twenty years
behind bars. The game thus captures the central problem of altruism,
which is that the most desirable solution for the individual is not
the same as the most desirable outcome for a group (in this case a
group of only two individuals). Hence the dilemma...

Iterated Version Of The Game;

...As a once-off proposition, the prisoner's dilemma game is rather
unpredictable. Your choice would be very much determined by a guess as
to what the other person would do. If you had both been arrested
together on a previous occasion, then the fact of having cooperated in
the past would surely be critical information in swaying your current
decision.

Axelrod and Hamilton decided that a repeated (or iterated) game would
be better able to capture the dynamic flavor of altruistic behavior in
the real world. They launched a computer tournament by asking sixty-
two academics in various fields to submit a computer program that
coded for a behavioral strategy, such as always cooperating, cheating
if the other cooperated, cheating every third move, and so on. These
programs were run against each other in pairs. The programs were then
entered into a "second-generation" tournament. Strategies that did the
best received the most copies in the second generation to simulate
evolution. This process was repeated for many more generations.

One of the most successful strategies was also one of the simplest.
Known as tit-for-tat, this required an individual to cooperate when he
encounters a new partner. From then on, he does whatever the other
individual does. If a new person moves to your neighborhood, you begin
by being friendly and welcoming. If he responds warmly, you strike up
a friendship. If he acts as though he wants to be left alone, you
remain virtual strangers. That is the flavor of tit-for-tat.

Tit-for-tat has an advantage in allowing individuals to avoid initial
mistrust and begin cooperating quickly. It is vulnerable to cheats but
can be exploited only once. Tit-for-tat is also forgiving, responding
only to the most recent actions of other individuals and not holding
grudges.

Biologists believe they have detected tit-for-tat in all kinds of
animal interactions, from grooming in monkeys and impala to the
exchange of costly eggs for inexpensive sperm among fish and worms
that can produce eggs and sperm simultaneously. Chimpanzees and
baboons join forces with individuals who have helped them in social
conflicts in the past. Vampire bats refuse to share blood with roost
mates who refused them previously.

Tit-for-tat-style reciprocity has been observed among humans in a
variety of situations, some of them quite surprising. One of the
oddest was described by Robert Axelrod in his analysis of interactions
between the British and their German enemies during the deadly trench
warfare of World War II. Remember that soldiers in the trenches were
involved in a new and exceptionally dangerous warfare that was a death
sentence for the majority who were unfortunate enough to be placed in
that situation by officers who were out of their depth, to put it
charitably.

Entrenched troops spent their time shelling the enemy, thereby
participating in a horrendous and strategically pointless slaughter.
Against all orders, the troops on each side developed an etiquette of
firing only to the side of enemy positions to minimize the loss of
life. Axelrod reports one astonishing memoir of a British officer
whose position was unexpectedly shelled by the Germans, although
without casualties. Following the attack, a German officer surfaced
from his trench and shouted, "We are very sorry about that; we hope no
one was hurt. It was not our fault, it was the damned Prussian
artillery."4 Both sides had realized that with the deadly weapons
available to them it was just too dangerous to try and kill each other
all the time. It was in their mutual interest to cooperate.
Reciprocity had broken out in one of the most unexpected places. There
was no verbal agreement, which would clearly have been treason, but
the repetitive nature of trench warfare provided an opportunity for
cooperation to occur between the opposing armies.

Kindness In A Cruel World:
The Evolution Of Altruism
by Nigel Barber
http://www.amazon.com/gp/product/1591022282/