That question is addressed to Richard, but I assume it's open to all
comers. I'll take a crack at it.
Surviving long enough to reproduce in a social setting requires that the
individual adopt a survival strategy that addresses how one deals with
others. In populations where individuals cooperate with one another, one
is faced with a choice. "Do I expend my own energy and resources to help
the other guy? If I help him and he does not reciprocate, then he's
benefited from my resources and still has all of his own upon which to
draw. He's way up, and I'm way down. If he helps me but I don't help him,
then HE'S out in the cold and I'm way ahead. If we both lend a hand, we'll
each be better off than if we had not co-operated, but if I don't help him
and he doesn't help me, neither one of us has lost much. We don't do as
well as we would have had we coordinated our efforts, but I do much better
than I would have were I to help him and he were to exploit me. Do I hold
up my end, or do I attempt to exploit my neighbor, accepting his aid but
not reciprocating?"
This is the prisoner's dilemma (PD hereafter). Keep the faith or sell out?
In a single round PD, selling out always maximizes your expected return.
Single round means you chose to keep the faith or sell out, your partner
does likewise, and then the benefits get distributed and the game is over.
Decission theory provides a number of different strategies for weighing the
potential risks and benefits; some conservative, some risky, but they all
favor the sell-out.
Life is more like a repeated PD. You make your decision to help or
exploit, your partner does likewise, you reap the rewards of round one, and
move on to round two. Survial is far more like the repeated PD than it is
like the single round variety. You can cooperate or exploit today, and you
may score big, but you're going to have to do it all over again tomorrow,
and if you find yourself paired with the same person you exploited
yesterday, they're not as likely to extend their trust to you a second
time.
The question is, when you meet someone for the first time, do you try to
exploit them (or refuse to cooperate because you expect they'll try to
exploit you) or do you extend your trust to a stranger and risk getting
screwed? There have been many a computer simulation devoted to testing the
effectiveness of various strategies for coping with a repeated PD.
A noise-free environment is one in which there is no chance that you will
mistake an attempt to cooperate for a sell out and there is no chance of
selling out by mistake. In a noise-free environment, the most effective
strategy is 'tit for tat.' This means that when you meet someone for the
first time, you extend your trust and devote your resources to a
cooperative effort. If your partner reciprocates, then when you see him in
the future, you'll know that he's to be trusted and so you expend your
resources on him. In extending a helping hand, you've forged a valuable
cooperative relationship.
If, on the initial round of the PD with a new partner, your partner accepts
your aid but does not return it, then you'll know not to cooperate with him
in the future. Start off extending aid and then give what you get 'tit for
tat.'
'Tit for tat' is not as effective when there is some chance that you will
misread your partner's actions as a sell-out when he, in fact, meant to
cooperate. If you give 'tit for tat' from the very first sell out, then
you might opt out of what might have been a mutually beneficial
relationship. When there is some chance of misinterpretation, it's best to
be a bit more forgiving. "Tit for two tats' is more effective in a noisy
environment than 'tit for tat.'
Social environments are very noisy. As anyone over the age of 15 knows,
communication can be very difficult. In noisy social enironments, your
best bet is to be pretty darn forgiving. Extending faith, hope, love, and
mercy, to your partners in the repeated prisoner's dilemma of real world
survival is an effective survival strategy.
Of course, at some point, it becomes obvious that someone is indeed
habitually exploitative. Even accounting for noise, it is clear that he's
just not going to hold up his end of the social bargin. At that point,
good survial strategies will direct one to cut off the free-loader and
cease expending one's resources on them. Loving thy nieghbor even after it
becomes clear that he won't love you back is not a good survival strategy.
I'll be interested to see what Richard has to say in responce to your
question. My background is in philosophy, so I tend to think in terms of
decision theory. Richard's background is in computers, so I expect he'll
have a different approach to answering the question.
Take care. -KMO
Resistance is Futile.
http://www.missouri.edu/~c538128/