According to this post this project is defined as follows:
Reynard is a seedling effort to study the emerging phenomenon of social (particularly terrorist) dynamic in virtual worlds and large-scale online games and their implications for the intelligence community.
He tells us a secret: Beyers once was a girl in SL because people tend to answer questions in interviews more willingly than when a guy does that.
What do we know about the link about the virtual and the real world characteristics? There is data e.g. from Sony from Everquest who measured quite a lot, like who was close to whom and when, who has what in their inventory etc. They also performed a lot of surveys (both via phone and online).
Another type of research was simulating real life behaviour in a virtual world. E.g. if you waved your hand then the avatar would do the same.
The big question here is: How does what you do in a virtual world influence what you do in real life and vice versa.
One finding: If you got a very good looking and strong avatar they found out that these people also behave stronger etc. for about 1 hour in real life.
Another study was where they take on the avatar of a very old person looking frail and feeble. The results was that the real people were less discriminating old people. Easier excusing of forgetting things or slow driving.
(sorry, I did not get the names of all the researchers).
But there are links both ways.
We will also have more and more the problem that there are people out there online who want to get other people. Spys cannot find that much out yet in e.g. Second Life but that might change. But it also already happened if you look back at the banking and fraud situation in summer 2007.
What do the Reynard Project research projects look like?
One thing might be a bot identifying group memberships and other things. What do you think you’d learn from that? Can you figure out that person’s nationality? Can you find out that this person is a successful business women or an artist?
As for nationality what you can do is to look at the language they type in, look at their time zone and their group memberships. If voice is used then it’s far easier to tell.
Another aspect is inventory/clothing. This is harder to study because you cannot simple peek in there from the outside and even if you can you cannot really tell e.g. how „stylish“ it is or whatever. That might be more easily done by an avatar filming and later looking and accessing it.
Can you tell if a female avatar is really a female person or a male? How would you tell this?
Another question: What do people think they know about the real life persons behind the avatars around them? Is this person really who they say they are? and this will change because presenation will always become better and thus you will stronger believe that an avatar is who they say they are.
Raynard Project is starting with research programs in Q4 2009.
(personal note: To me this whole research sounds like a quite huge privacy problem)