Answers to questions about the Behaviour Composer and the Relative Agreement sample model

[Ken Kahn]

Dear Scott (and I've cc-ed the Modelling4All list where others may be
able to help),

First a clarification of "self", "other", and "prototype". Prototypes
are individuals whose behaviour and state are defined within the
Behaviour Composer. Typically one of those behaviours is to add
several copies of the prototype to the model.

Micro-behaviours contain NetLogo code that sometimes needs to refer to
the agent running the code and that is what "self" is for. "the-other"
is a convention that we use to model pairwise interactions. See
http://modelling4all.wikidot.com/modelling4all-programming-guide#toc2
for details.

Regarding copying models and where the "live" models are copied when
you load a URL that contains "frozen=" or "copy=". Also when you load
a model via a button. This copy is yours and you need to save the URL
(or rely upon the cookies of your browser) to load it another time.
Your model is private until you put the model URL on a public web
page.

Regarding the sample Relative Agreement model and the associated
library of micro-behaviours they were designed to follow closely the
cited paper. To extend the model to have two different opinions per
person would require making minor variants of several of the micro-
behaviours to use different variable names for opinions and certainty.
Then the person prototypes would have both the original behaviours and
these variants. Additional micro-behaviours would be needed to model
interactions between these opinions.

By the way, after this model was constructed a way was added to this
system to enable the three prototypes "Normal Person", "Positive
Extremist", and "Negative Extremist" to share a list of micro-
behaviours rather than duplicating them. I'll update the model and
library.

I hope this helps.

Best,

-ken

2009/9/20 Scott Bennett

Ken, I was ultimately able to get into the material using both Firefox
and Explorer.

One basic question. In the code for opinion dynamics in a
neighbourhood context, are "self" and "other" what you call
prototypes (generic agents I guess) in your modelling system? If not
what are the labels for the prototypes in that model? Just trying to
get the terminology straight.

I am having a little trouble seeing how I could copy an existing model
like that and then insert new commands into the code, but I am sure
this is just my lack of experience. Where does the copied model go to?
Is it in the space for my own models?

More generally, how would one approach creating a model that had
different waves or cycles to it based on different issues? For
example, let us we have an opinion spread model relating to issue 1-
which could be called environmental concern. We also have another
issue which could be called economic concerns. The joint importance of
the two could be modelled so that the importance of one was negatively
correlated with the importance of the other. Something like an issue
novelty effect.

Anyway, I realize you can't respond to every question you get, but if
you could clarify the most basic one that would be a useful starting
point for me.

thanks very much-seb-090919