From newshub.ccs.yorku.ca!torn!utcsri!rpi!uwm.edu!cs.utexas.edu!uunet!trwacs!erwin Thu Oct  8 10:10:37 EDT 1992
Article 7060 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!torn!utcsri!rpi!uwm.edu!cs.utexas.edu!uunet!trwacs!erwin
>From: erwin@trwacs.fp.trw.com (Harry Erwin)
Newsgroups: comp.ai.philosophy
Subject: Re: unpublished chapters of "The Turing Option"
Message-ID: <737@trwacs.fp.trw.com>
Date: 29 Sep 92 14:51:47 GMT
References: <1992Sep25.203827.17312@news.media.mit.edu> <1992Sep28.143838.5772@hawk.cs.ukans.edu> <1992Sep28.174024.15814@news.media.mit.edu>
Organization: TRW Systems Division, Fairfax VA
Lines: 52

On running multiple personalities:

I'm getting some feedback that this is an unusual skill, but I sometimes
deliberately run multiple personalities simultaneously. I do this when I'm
trying to understand someone else. I take sensory input from the
polysensory areas, postprocess it to shift the point of view, and then
feed it back to the polysensory area. Then I do a context switch to the
simulated personality and run it off the shifted data. If I've calibrated
the simulated personality based on previous behavior of the person, I can
get insight into the motivations and likely future behavior. It works
fairly well, although maintaining three points of view (introspective
self, self that's processing the sensory data, and simulated person) is
exhausting, and I have to ensure adequate shares of time for each, or
things collapse. I'm not schizoid, so I don't know what it would be like
to do this involuntarily, but I suspect "aspects" would run into similar
phenomena. Restricting aspects to interfacing through the polysensory
areas would probably work better than allowing them full access to the
neocortex.





























Cheers,
-- 
Harry Erwin
Internet: erwin@trwacs.fp.trw.com



