X-Message-Number: 4298
From: "Joseph J. Strout" <>
Newsgroups: sci.cryonics
Subject: Re: Improving Memory
Date: Tue, 25 Apr 1995 13:24:10 -0700 (PDT)
Message-ID: <>
References: <3nidn6$>

On 25 Apr 1995, John K Clark wrote:

> There is an interesting article in the April 7 1995 issue of
> CELL about long term memory. Jerry Yin used genetic engineering
> to produce a strain of fruit flies that overproduced a protein
> named CREB. He found that these artificial flies were much
> better at producing long term memories than natural fruit flies.
> It takes 10 repetitions for normal flies to learn something well
> enough to remember it 7 days latter, the genetically improved
> flies only needed one trial.		...

I haven't read the article (and thanks, John, for pointing it out), but 
I'd like to offer a note of caution here.  This experiment, as described, 
does not demonstrate that the flies are more intelligent or have *better* 
memories, but merely that they learn *faster*.  This could be a 
disadvantage in real-life situations, for example, this one trial (or 
several trials) may wipe out any previous knowledge the fly had about the 
stimulus.  This is exactly what you might see in an artificial neural 
network if you set the learning rate too high.  If the flies really could 
be 10 times smarter after a simple gene modification, then why didn't 
evolution do it?  These are just some of the things to consider when we 
talk about trying to "improve" our brains.

Just my $0.01999999999 worth...

,------------------------------------------------------------------.
|    Joseph J. Strout		Department of Neuroscience, UCSD   |
|    		http://sdcc3.ucsd.edu/~jstrout/    |
|------------------------------------------------------------------|
|             Check out the Mind Uploading Home Page:              |
|     http://sunsite.unc.edu/jstrout/uploading/MUHomePage.html     |
`------------------------------------------------------------------'


Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=4298