Evolving Language Experiment Experiment [RESULTS!]

[brandrinn]

I just sent out Round 4. I sent one to *Ceresz, but if he doesn't reply we can just work around it.

I think the next round will be the last one. It would be interesting to see what you all can do given more time, but that will have to wait for the next time. So far I can see some patterns emerging, and that will give us an idea about how many people and how many rounds we need to do this sort of thing in the future.

Also, Terra, I recommend you find some way of automating this process, as sending all the images will be a lot of work if you're doing it manually.

[Herr Dunkel]

I WANT IN NAO

[clawgrip]

I am interested in seeing the results. I just worry that the results may be contaminated because the participants all know what is going on.

[Herr Dunkel]

Uh duh this is a linguobabble forum.

[brandrinn]

The fact that our participants know what's going on means that the patterns will emerge much quicker than would be realistic in a random sample of people.

But the main purpose here is a) to gather methological data for future experiments, and b) to see if there is anything interesting about the patterns ZBBers make when each person is given a fraction of control over a process.

[Mr. Z]

Umm... yeah. I really don't know what to do, so I just make up patterns on purpose. Am I supposed to think of random words?

[brandrinn]

Yeah, you're supposed to think of the best word that fits the picture in your opinion. You can consciously make patterns if you want to, but our hope is that the patterns will emerge even without trying, over time.

EDIT: just waiting on two participants to finalize the results.

[brandrinn]

OK, since two people still haven't responded, I'll just post the results here. First, here are images of all the creatures in case you haven't seen them yet:

[brandrinn]

OK, I'm an idiot. I can't figure out how to import a table into a BB post. Any technical help?

[Wattmann]

OK, I'm an idiot. I can't figure out how to import a table into a BB post. Any technical help?

I don't think you can...
If you could, you might try HTML or somesuch, but elsehow IDK

[Lyhoko Leaci]

Do a screen capture of the table and then post the image on here?

[Terra]

Perhaps upload an excel file, and then give a link to it? I'm eager to see the results!

* * *

I like the pictures. Did you make these yourself? Did you just use MS Paint? How long did it take you? Do you want to make some pictures for my experiment so I don't have to use shapes?

[brandrinn]

OK, I'm sorry I was swamped yesterday, but I'll put the results up tonight, I promise.

[Terra]

And by "tonight", you mean tomorrow night?

[brandrinn]

"Tonight" means ten o'clock Korean time. Here's a table, finally!



Turtlehead sent me some extra data that's not in the chart: Mita for number 2, and Ita for number 8.

[Terra]

To make sure that I understand, each round you sent 2 images to 4 different people. Why do some cells have multiple names then?

* * *

To be honest, I don't see that many patterns. Maybe "sn-" for 2 eyes. I think that this is because each person only ever named 2 creatures, instead of the whole set. What's your interpretation?

[brandrinn]

No, the images circulated around quite a bit. Most people saw each creature at least once if not twice or three times. I'll upload the table of who saw what in which round tomorrow. But for now, just know that Each person saw two named images, then named two images, and received two different named images, repeat.

[brandrinn]

Here is a segment of the table I used to organize the data. Each person has two columns. The first shows which picture they were shown and which they named. The second column shows the names they gave. Some of the numbers in this picture are inaccurate, and three participants are not listed, but you get the idea about how it works. I hope this can be helpful for you Terra, even if it only serves to highlight problems to overcome.