[starlogo-users] Do some individual turtles inherit from specificparents?

[Emre Sevinç]

Eric Klopfer wrote:

>Turtles by default inherit all of the characteristics of their (single)
>parent.  They are really cloned from one parent using the "hatch []"
>command.  However, it is possible to grab the characteristics of the other
>parent and use some combination of those.  See for example the attached
>"sexual.slogo" project which is like the rabbits model that comes with
>starlogo, only using sexual reproduction instead of asexual.
>  
>

As another cognitive science researcher I'd like to have a little bit 
more detail, you say the attached
file uses sexual reproduction instead of sexual. I've took a look at it 
and since I'm not a StarLOGO
master hacker please enlighten me:

I see that reproduction scheme is as follows:

===================================================================
to reproduce
  if energy > (hatch-threshold)
      [grab one-of-turtles-here
          [if ((species-of partner) = rabbit) and ((energy-of partner) > 
(hatch-threshold))
              [setdad partner
               reproduce2]
          ]
      ]
end

to reproduce2
  setenergy-of dad ((energy-of dad) * .667)
  setenergy energy * .667
  hatch [setenergy (((energy-of dad) + energy) / 2)
         seth random 360 jump 1]
  ifelse energy > (hatch-threshold) and (energy-of dad) > (hatch-threshold)
      [reproduce2]
      [setdad -1]
end
=======================================================================

Are really two parents used? I was a little bit confused about that 
hatch[] function. At exactly which
line(s) does a new turtle inherit the characteristic of its parents?

As for identifying each turtle, I believe each of them while being 
created (hatched?) can be given
a unique ID. Am I right? Is that possible to program in StarLOGO?

Thanks in advance,
Emre Sevinc

>-----Original Message-----
>From: starlogo-users-bounces at media.mit.edu
>[mailto:starlogo-users-bounces at media.mit.edu] On Behalf Of Rob Bowers
>Sent: Sunday, March 13, 2005 3:39 PM
>To: starlogo-users at media.mit.edu
>Subject: [starlogo-users] Do some individual turtles inherit from
>specificparents?
>
>"It is very difficult in both AGENTSHEETS and StarLOGO
>to develop agents possessing even simple cognitive
>models, and it would be impossible to create models
>using any kind of evolutionary approach" [Gilbert & Bankes, 2002].
>
>This quote is what initially directed me away from StarLOGO. I need my
>agents to
>be able to inherit traits from specific parents. For this, they would
>at least need to be individually identified. And subsequent
>generations would need to be introduced as a consequence of the
>parents "mating".
>
>However, a trusted friend strongly advised me to look deeper into
>StarLOGO. So I ask you all: Would I find StarLOGO appropriate for my
>needs, as sketched in the previous paragraph? If not, can you
>recommend a more suitable environment?
>
>I will greatly appreciate any advice. Gilbert & Bankes [2002] was the
>only review I found that addressed this issue. It is otherwise hard to
>tell from the outside, without investing rather heavily into an
>environment, which is suitable, let alone which is mOST suitable.
>
>Thankyou,
>Rob Bowers
>_______________________________________________
>starlogo-users mailing list
>starlogo-users at media.mit.edu
>http://education.mit.edu/mailman/listinfo/starlogo-users
>  
>
>------------------------------------------------------------------------
>
>_______________________________________________
>starlogo-users mailing list
>starlogo-users at media.mit.edu
>http://education.mit.edu/mailman/listinfo/starlogo-users
>  
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://education.mit.edu/pipermail/starlogo-users/attachments/20050315/2e5d2fc3/attachment.htm