|
Hello there! I am very much into Starcraft and also passionate about scientific research. Today, I want to combine those two passions into one project. In my academic life I am studying in the M.Sc. program Neuro-cognitive Psychology at the LMU Munich (Germany). Recently, I finished an internship about economical decision-making and I was thinking that this field may also be a great opportunity to design a project around Starcraft. Video games are rarely (if ever) used in psychological research (Skillcraft with its expertise research being the exception). However, they have clear advantages. Especially in Starcraft, you can pretty easily get access to tons of data through replays without necessarily having to conduct expensive classical experiments. My personal goals are to make a first step in showing the scientific world of how valuable researching computer games is for various fields and to get an article published in a peer-reviewed journal in about one year.
I am now planning to put together a team of about five people. Fellow scientists who are also passionate about Starcraft and are able to dedicate some time to this side-project, please message me! I am explicitly looking for anyone in any field as long as you personally think that you can contribute to the success of the project. For example, an expert in game theory would be great to discuss what should even be treated as an decision within a game. I could also imagine someone from the field computer science to be a valuable member by developing a suitable means to extract the data from the replay files. Do you really want to contribute but do not know how? Just message me and I will happily discuss the possibilities with you.
Since I want to exert full collaboration from the start, any ideas are welcome! You think decision making is boring and have a cooler idea? I want to hear it! Even if you do not want to be on the actual team but have a great idea, please tell me!
What else would you like to know? Feel free to ask here or via direct message!
TL;DR: Research project about Starcraft. Possibly decision making. I want to put together a team. Interested? Read the whole text.
|
It may be a good idea to have a high-level player (preferably an articulate pro/semi-pro) to provide feedback on game-specific knowledge and expertise.
Also, it's probably better if you say that you want to explore the value of researching computer games is for valuable fields. I personally believe that this may be the case and certainly hope so, but I'd be very careful how I approached it.
|
Sounds like a really cool idea and something I would have liked to be involved in if I was smart enough haha
Have you tried posting in forums / boards for various disciplines you think may be useful? I think you may find some people who have an interest in gaming but may not know about TL or SC2 as well
Best of luck
|
You should probably start out by formulating some sort of hypothesis. Or retest an already formulated, generally accepted and well-known one using data from SC and then compare the results to the original as well as time spent on gathering the SC data showcasing the efficiency of SC as a model.
I sadly do not think I can be of any assistance with the actual project both due to my field of work and time constraints.
EDIT: I also think you are going to run into a bunch of problems depending on your project. I think the situations (playermentality, skill level, experience et.c.) might very well vary too much between a lot of games and can be very hard to adjust for resulting in it being impossible to really say anything. The field of decisionmaking is rather far outside of what I work with though, so do not take this as more than a complete laymans initial reaction.
|
Oh bugger, I don't think my electrical engineering background would be of much help, but I wish you the best in any case! As Day9 suggests, if you want to show your passion for Starcraft and ESports, don't just look for work, make some!
|
Sounds like an intriguing project. While I am an anthropologist I work in the sub field of biological anthropology so I am weaker on the cultural aspects of the field. Still I think having an anthropological perspective on this would be an interesting addition. I wonder if you will be using replays of players would you still need consent forms if you plan on doing any kind of analysis of their "work"?
|
Thank you for your input guys, I really appreciate it!
On April 15 2013 09:43 dicedicerevolution wrote:
It may be a good idea to have a high-level player (preferably an articulate pro/semi-pro) to provide feedback on game-specific knowledge and expertise.
Also, it's probably better if you say that you want to explore the value of researching computer games is for valuable fields. I personally believe that this may be the case and certainly hope so, but I'd be very careful how I approached it.
I am certain that it computer games like Starcraft are valuable in some kind, although I do not know the exact Ups and Downs yet. So I agree with you!
On April 15 2013 15:13 Meatex wrote:
Sounds like a really cool idea and something I would have liked to be involved in if I was smart enough haha
Have you tried posting in forums / boards for various disciplines you think may be useful? I think you may find some people who have an interest in gaming but may not know about TL or SC2 as well
Best of luck
This has almost nothing to do with smartness. Just a little background knowledge is helpful so you don't have to invest so much time. To your second point: I considered the idea but I think the most important thing of all is that you know the game. That's also why I did not ask any of my colleagues yet. I need people who love the game itself more importantly than having a big background knowledge in any scientific field. So I thought TL would be an ideal way to start.
On April 15 2013 16:01 Ghostcom wrote:
You should probably start out by formulating some sort of hypothesis. Or retest an already formulated, generally accepted and well-known one using data from SC and then compare the results to the original as well as time spent on gathering the SC data showcasing the efficiency of SC as a model.
I sadly do not think I can be of any assistance with the actual project both due to my field of work and time constraints.
If we stick to decision making under uncertainty I know some sweet studies which might be useful to try and replicate the results. I agree with you completely. To show that this method is able to produce is same results as classical methods is of highest value!
On April 15 2013 16:01 Ghostcom wrote:
EDIT: I also think you are going to run into a bunch of problems depending on your project. I think the situations (playermentality, skill level, experience et.c.) might very well vary too much between a lot of games and can be very hard to adjust for resulting in it being impossible to really say anything. The field of decisionmaking is rather far outside of what I work with though, so do not take this as more than a complete laymans initial reaction.
I experimented with motivation and computer games a little bit in the past. Your concern is correct. Such games are awfully complex and decisions may vary among players with different expertise. However the latter is not necessarily something bad. In classical reaction time measurements, interpersonal variance is also higher than one would initially suspect. If there is a suitable baseline which applies to everyone in the same way then this should not be a big problem. Finding the baseline and fixing decisions will be the biggest issue I imagine.
On April 15 2013 16:26 Slaughter wrote:
Sounds like an intriguing project. While I am an anthropologist I work in the sub field of biological anthropology so I am weaker on the cultural aspects of the field. Still I think having an anthropological perspective on this would be an interesting addition. I wonder if you will be using replays of players would you still need consent forms if you plan on doing any kind of analysis of their "work"?
I would love to have an anthropological perspective on this. What immediately came to my mind was: Various studies ahve shown cultural differences in decision making especially involving risk taking. It would be interesting if we could also show those differences or if they vanish within the Starcraft model.
To your second point: I am pretty sure that if just gather replays, we do not need a written consent, since consents are there for giving the permission to the experimenter to collect the date himself and use it. In our case the participants collected the data on their own and voluntarily sent them to us. Especially, as there are always two players in one game it is impossible to get consents from all players anyway.
|
sc2 is already us military program
|
Hi, I'm finishing my bachelor of science thesis (physics) and I plan on going on a PhD but I have no idea how can I help. Any your ideas?
|
669 Posts
This sounds like a really interesting idea,
I would have loved to be a part of this ( CS Ms ) but sadly I don't know anything about sc2 replay file formats or data mining in general. Also terribly busy with some papers and courses these days.
OK nevermind those,if you don't mind talking here about it; what kind of stuff you have in mind for CS guy?
I'm looking into replay file format as I'm posting this, I might be able to help to some extend
|
I just finished my BSc in management, also at LMU. I took one course in game theory, so I do not know if I would be of any help given my background. but you can PM me if you think so.
€ have experience in data mining w/ ProM & paper writing in general
|
Sounds like a nice initiative, but you have to start with some substance to get a project going. Basically, you've said "let's do a pysch paper using starcraft data." Taking data and seeing what kind of paper you can produce out of it generates poor research imo, and it's especially inadvisable to be looking for collaborators at this point. At the very least, formulate a hypothesis of some kind and do a little preliminary research first.
|
(I will reply to everyone who's interested in a couple of hours as I have to attend a meeting now.)
On April 15 2013 21:34 AcrossFiveJulys wrote:
Taking data and seeing what kind of paper you can produce out of it generates poor research imo, and it's especially inadvisable to be looking for collaborators at this point.
It seems that I did not bring my aims across the way I intended, so let me clarify. I am not planning on "just looking into the data and see what might be possible." What I personally want is to work in an interesting interdisciplinary project together with people who are as passionate about what they do as I am. That is why I want everyone to be fully on board with the project's / group's goal and define this goal together. I do not want to make a huge impact on matter Y in field X which is why I defined it so broadly.
|
You can PM me if you've done the background research and established hypotheses. Ideally, you should have some ideas about the design. I'm interested in discussing 3 questions:
1) What do we want to measure?
2) How do we measure x?
3) How do we analyze the results?
I got a BSc in Cog.-N.-psych and currently, I'm doing a Research MSc in Psychological Methods. + Show Spoiler +I also... advise on how to structure and write scientific articles, but not for free. 
Edit: All provided that this is in no way related to your own thesis, obviously.
|
On April 15 2013 22:02 Cyberonic wrote:(I will reply to everyone who's interested in a couple of hours as I have to attend a meeting now.) Show nested quote +On April 15 2013 21:34 AcrossFiveJulys wrote:
Taking data and seeing what kind of paper you can produce out of it generates poor research imo, and it's especially inadvisable to be looking for collaborators at this point. It seems that I did not bring my aims across the way I intended, so let me clarify. I am not planning on "just looking into the data and see what might be possible." What I personally want is to work in an interesting interdisciplinary project together with people who are as passionate about what they do as I am. That is why I want everyone to be fully on board with the project's / group's goal and define this goal together. I do not want to make a huge impact on matter Y in field X which is why I defined it so broadly.
Wish you the best with the project =]
Btw, an idea I had a few years ago was to start a (gently) peer-reviewed pub for starcraft (and other games?) related research. E.g., papers on ladder/tournament trends, including balance or success/proliferation of builds, supported by formal statistics. Or evidence-supported research on which countries are successful in starcraft and hypotheses/research on why (other countries besides SK would be interesting).
One clear idea I had was to determine at what league there was statistical significance in players choosing to play different styles on different maps or even different styles against different races (do bronze players?).
A lot of great material has been posted on this forum that would make good contributions with some revisions (e.g., that article someone posted on how gas mining varies by position of geysers).
Idk, I think it would be great if someone were to organize this. Perhaps starting by collecting previous articles and contacting their creators to see if they might be interested in submitting or helping in the peer-review process.
I'd be willing to co-organize something like this but not to lead the effort (I'm finishing up a phd and am doing "real" research lol)
|
Thank you for all your interests. I will come back to all interested ones until tomorrow.
Some of you really wrote long texts with many ideas already, that's great!
|
|
|
On April 15 2013 22:02 Cyberonic wrote:(I will reply to everyone who's interested in a couple of hours as I have to attend a meeting now.) Show nested quote +On April 15 2013 21:34 AcrossFiveJulys wrote:
Taking data and seeing what kind of paper you can produce out of it generates poor research imo, and it's especially inadvisable to be looking for collaborators at this point. It seems that I did not bring my aims across the way I intended, so let me clarify. I am not planning on "just looking into the data and see what might be possible." What I personally want is to work in an interesting interdisciplinary project together with people who are as passionate about what they do as I am. That is why I want everyone to be fully on board with the project's / group's goal and define this goal together. I do not want to make a huge impact on matter Y in field X which is why I defined it so broadly.
Hm. Well. My problem with all this is that you don't even know what you want to analyse. Until now your only goal is to have a paper published, which is basically nothing
I don't want to be a bummer, I think you can certainly put this up and have it produce results but you need to focus and define a goal for it. If I was a person with a background fitting to what you are looking for, I would never sign up for a project that has not even an idea about what they want to do.
Best of luck, still 
|
I am definitely interested, but before I could oblige I would need to know exactly what data you are looking for as well as what the commitment--on my end--would be like. I have a M.S. in Physiology (actual research degree, not a "plan b"), all coursework was neuroscience-based and nearly all research involved neuroendocrine systems.
I can say you chose a great topic to research. If your thesis involves a largely ignored area it is fertile grounds for pretty much anything you want to study.
Edit: I highly doubt my research would be directly relevant. I think it is highly illegal to homogenize the pituitaries, hearts, and other various control organs of starcraft players for the purposes of protein and mRNA extraction (or any other purpose, really). But I figured I have some understanding of the nervous system and that couldn't hurt.
|
A while ago I did look into the possibility of using SC2 replay data for my M.Sc. thesis in Psychology, but I didn't come up with any hypothesis that would be a) interesting or b) actually possible to test with the given data.
I would be quite interested in what you come up with. Gogo team science!
|
|
|
|
|
|
EPT
