[MUD-Dev2] [DESIGN] Rewards

[Sean Howard]

"Mike Sellers" <mike at onlinealchemy.com> wrote:
> You're confusing two different things.  Behaviorism as a theory (that
> internal cognitive states are irrelevant to psychological inquiry) has
> fallen out of favor for very good reasons.  But operant conditioning (the
> companion to classical or Pavlovian conditioning) is *extremely* useful
> and reliable.

Operant conditioning... that sounds familiar. I think that's called
brainwashing.

> I disagree completely with the idea that rewards should never be used as
> a social control mechanism: much of our real-world government is based on
> rewarding behavior that is seen as socially useful, and in games we know
> well that *people do what we reward them for.*

But the effects of which have never properly been studied because we just
buy into the idea wholesale without ever questioning it. I read an example
of some elementary school in Switzerland (or somewhere not in the US)
where they enacted a token system. Behave and you get a token, which are
redeemable for toys, extra recess time, and whatever. It was eventually
revoked because it instilled a mercenary attitude amongst the children.

The moment you introduce a social control mechanism like that, you
instantly allow any behavior that is not included by that mechanism. If
tokens are all that matters, then philosophical or moralistic reasoning
goes away.

Yes, people do what we reward them for, but this is obviously not the best
solution. Can anyone argue that the corporate world is a better place
after becoming a glorified Skinner Box? Yes, people will work weekends for
a bonus, but should they have to? Are their lives better because of it? Is
the product better because of it? EVERYTHING I have ever read or
experienced in the game industry says that the worker suffers, the product
suffers, the company suffers. We like to wrap up the negative impacts of
this kind of crap as "morale", but it is far deeper and more complicated
than that - which is why our solutions to offer MORE incentives always
backfire.


> As above, the question is do we reward them extrinsically for
> fundamentally un-fun behaviors ("oooh, watch me press the 1 key over and
> over to 'attack' this thing"), or do we find ways to create situations
> and player behaviors that are intrinsically rewarding?

There's that word again... "fun". Is that the only kind of rewarding you
know? Because I've felt plenty rewarded for simply hanging out with my
friends watching bad movies and eating pizza, exploring my neighborhood on
foot, or writing an email. Give me a wall and I'll paint on it. Give me a
mountain and I'll climb it. Give me a door, and I'll go through it. Give
me a friend...

The designer doesn't have to make these things "fun". In fact, any effort
to do so will fundamentally unbalance the nature of what makes these
things rewarding in the first place.

> To the degree that we can do the latter, we will also reward
> various forms of socialization, as humans are inherently motivated by
> positive social interactions.

Yes, but they don't need to be rewarded for it. The action is its own
reward. You could hogtie a group of people, blindfold them, stick them in
the dark, and they will STILL find a way to interact with each other. You
don't have to reward something that people will dying trying to do anyway.
It's redundant and it creates a situation where you only reward PART of
the socialization process. You reward communicating with other players,
but not what the communication is. You reward the act of fighting in a
group, but not the quality of that group. In other words, you are
rewarding - no, enforcing - only the most shallow understanding of these
things and then standing around slackjawed wondering why all the people in
your game are jerks. Perhaps worse, you accept it as inevitable.

-- 
Sean Howard