From: Anders Sandberg (asa@nada.kth.se)
Date: Wed Dec 19 2001 - 12:28:07 MST
On Wed, Dec 19, 2001 at 01:50:13PM -0500, Smigrodzki, Rafal wrote:
>
> The second point is that if we manage to survive for a very long time,
> our situation may be much like Leonard's. If we reject the notion of
> stasis, for what is the point of living forever if you are in a rut,
> then we have to embrace change. And with enough change, inevitably
> the person you were before will be lost. While the memories may not
> be totally erased, still you will be effectively a different person and
> the memories might as well belong to someone else. Of course the time
> horizon will be much larger; unlike Leonard's 15 minutes, we may have
> a thousand subjective years of continuity. But beyond that threshold,
> we will not be the same person. Like Leonard, our life will be a series
> of time-bounded moments, with continuity and causality linking them,
> but no sense of common identity.
Exactly. Nature seems to have optimized our memories for learning fast
when we are young and more slowly later (in order to retain early
important information while preventing overwriting); at least that is
the thesis of a paper I'm writing (I have a very nice little model
predicting this, that seems to fit real psychological data on
autobiographical memory). If we allow nature to have its way, we would
turn into struldbrugs unable to learn eventually, so any life extension
scheme has to involve keeping the learning rate reasonably high. Which
means that the old memories will fade away, unless we actively try to
save some of them by re-training or revieweing them at regular
intervals.
I'm thinking of making a transhumanist version of the paper when the
first is finished, looking more into what my model predicts we should
do.
-- ----------------------------------------------------------------------- Anders Sandberg Towards Ascension! asa@nada.kth.se http://www.nada.kth.se/~asa/ GCS/M/S/O d++ -p+ c++++ !l u+ e++ m++ s+/+ n--- h+/* f+ g+ w++ t+ r+ !y
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:12:40 MST