Newsgroups: comp.ai.genetic
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!udel!gatech!howland.reston.ans.net!pipex!sunic!news.chalmers.se!news.gu.se!gd-news!d6222.shv.hb.se!sa209
From: sa209@utb.shv.hb.se (Claes Andersson)
Subject: Re: The evolution of communication
Message-ID: <sa209.75@utb.shv.hb.se>
Sender: usenet@gdunix.gd.chalmers.se (USENET News System)
Nntp-Posting-Host: d6222.shv.hb.se
Organization: Department of Scocial Science
References: <56572.s9101799@mail.student.utwente.nl>
Date: Thu, 17 Nov 1994 17:34:35 GMT
Lines: 72

In article <56572.s9101799@mail.student.utwente.nl> "Albert van Breemen" <s9101799@mail.student.utwente.nl> writes:

>Some weeks ago I (= student at the Department of Electrical Engineering, 
>University of Twente, The Netherlands) heard for the first time about GA 
>and I was quit fascinated. When shortly after I saw the MPEG-movie of Karl 
>Simm's I was sure about it: I had to come up with a project of my own. 
>After some time a came with the following idea:

>   'The evolution of communication'.

>The idea is to have an environment with 'creatures'. By evolution this 
>creature have to develop some kind of communication. It interrest me, if 
>the environment is changed, what kind of communication there will be and 
>for what purpose the creature will begin to communicate.


>I can use two starting point here:

>1) Nothing at all! So both the creature and communication have to evolved.
>2) I can start with 'creature' which has already been simulated by others.
>   (I think here of a simulation I saw of a cockroach)

>The best start is 2 because now i don't have to spent time on the 
>evolvement of the creature. But there are some proplems I have:


>* What is exactly communication between two or more creatures?
>  Is reacting on signs of the environment of the creature (whether these
>  signs are generated by an other creature or the environment) a kind
>  of communication or not? Or is this just reacting?
>  
>* What kind of fitness function I have to use?
>  Karl could define the lenght of the path as fitness, I can not define 
>  'a fine chat between creatures' as fitness.

>* Does anyone know why, when and how the first creature in nature began to
>  communicate?


 My theory is that communication is a simple result of how the lifeforms 
percept their environment. Percepting ones environment is of course one of 
the things that evolve fstest because there is a great selection pressure 
for it. When the environment consists mainly of other lifeforms, you have 
communication. Communication, according to R. Dawkins, is a "gate-way" for 
neural signals from one specimen to another. I suggest you read R. Dawkins "
The Extended phenotype". It's a very interesting book. 

 To have emergent communication between animates, simply do as follows:
1: Give them the ability to percept their environment or at least a 
   possibility to develop such traits.

2: The ability to affect the environment in some way. This could 
   be very simple, just let them compete for the same energy resources. 
   This can be extended in many ways.	

Claes Andersson. University of Bors. Sweden








>================================================
>A.J.N. van Breemen
>lipperkerkstr 128
>7511 DD  ENSCHEDE, The Netherlands
>tel: 053-317819
>email: a.j.n.vanbreemen@student.utwente.nl
>================================================
>       
