a model of adaptive preferences?

["C.N.Gomersall" <gomersni@xxxxxxxxxxxxxxxxx>]

I risk being blackballed and sent to Coventry, but just thought I'd take
advantage of the "silence du mort" to post an idea of mine on the list for a
second time. (The first time, I made the mistake of getting in the way of the
Davidson seminar. Or else it was just a lousy idea, but at least it's now more
likely that you'll be able to spare a moment and tell me.)

I set out below a situation that may be a model with more general application.
At the end are a few questions: requests for references, comments, &c. I'd like
to hear from you **even if you think the model is simplistic, irrelevant, or
obscure.** Here goes....

Twenty or thirty people are at a Christmas party. Each has been told to bring
a gift, valued at no more than $5, to wrap it and to place it on the table.
Lots are drawn to determine people's rank order. The first person chooses a
parcel and opens it. The second person does the same, but then he or she may
exchange that gift for the one that the first participant received. The latter
has no say in whether or not this exchange takes place.... Subsequent
participants follow the same rules, i.e., they are able to force an exchange of
gifts on *anyone* preceding them.

Suppose that someone before me has opened an envelope containing a $5 bill. When
I open my parcel, I find some trinket that I should like to exchange for the $5,
but I reason that someone following me might also prefer it and that, in such a
case, I'd be landed with that later person's gift which might be even less
desirable than the one which I now hold. Thus, I modify my desire to exchange--I
adapt my preferences--and, instead, I stick with my present gift (no pun
intended).

As a relative newcomer, I'd appreciate any and all responses to points such as
the following:

1.  Is this a useful model even if, in "real life", there is no pecking order
such as the one I've described? What matters, I think, is that people might act
as if there were such an order or, more weakly, as if there could in some sense
be "later" players whose possible behaviour/choices would need to be considered.
The model might still work even if people's subjective pecking orders were
incompatible, i.e., if I thought that you are further along it than I was, and
you took the opposite view.

2.  Any suggestions as to literature on the _mechanics_ of adaptive preferences?
(E.g., those of you in the "democracy and media" discussion and its predecessor
who felt that readers/citizens adaptively prefer to veg out?) I must admit I'm
groping towards a more rigorous expression of the game I've described--groping
being the right word here, since adaptation could take place through a process
of crypto-tatonnement? (Sorry about that.)

3.  For some reason I see this game as a potentially rich analogy--to what, I'm
not exactly sure. Am I being naive? creative? obtuse? or all of the above?

Thanks, apologies, and best wishes from someone who is both a lurker AND a
newcomer.

_______________________________________________________________
  Tel. 319 387 1133    Fax  319 387 1088    Home 319 382 4676
      C.N.Gomersall   Luther College   Decorah Iowa 52101