Game Politics (gamepolitics) wrote,
Game Politics

Play Games, Advance Science

"A person is smart. People are dumb, panicky, dangerous animals, and you know it." - Agent K, Men in Black

And, as Rob Goldstone, director of Indiana University's Group Experiment Environments, might add, "Collective behavior is potentially more controllable than isolated individual behavior because of the strong influences among the individuals' behavior." Goldstone's research uses video games to "observe, explain, and computationally model how groups of people behave."

The game experiments are available online and open to the public. Goldstone's design goal is to have more than five people playing at any given time. But until the project gains more popularity, AI bots fill in for missing human players. At game's end, players can read a thorough explanation of the group behavior modeled and what the experiment results can mean to the real world.

Group Path Formation is one experiment/game that rewards participants for reaching target destinations but subtracts points for each step. Fewer points are deducted if a previously established path is followed so players must strike a balance between the shortest distance and following in the footsteps of others.

"At a broad level, the best grounds I can see for being optimistic about the future of humanity is that sometimes people will shun well-trodden roads and forge their own paths," said Goldstone in an email to GamePolitics. "But once they do, it turns out that they are attractive paths for other people to follow. In this way, humanity can be both flexible and efficient."

"The alternative method of crowd control suggested by our work is to change the structure of the environment such that certain navigational behaviors are facilitated while others are indirectly hindered. Even without instituting physical or abstract barriers, it may be possible to indirectly control collective behavior with considerable efficacy..."

"To this end, the Active Walker can be used to guide policy decisions. For example... if you want a group of people to create path networks that minimize the total amount of trail needed but still connect a set of destinations, it is best to have people's steps wear away quickly... because rapid decay of paths allows the group to create new paths without being strongly constrained by the original..."

While a web version of Group Path Formation is not yet available, there are four other experiments you can occupy yourself with while benefiting science at the same time. Bonus!

Forager explores how participants allocate themselves in a world with scarce resources.

Picture Game asks participants to guess an answer to a problem. Participants are shown other participants' answers and how well they scored, which can be used to influence guesses.

Group Sum ask participants to enter numbers which are added to a single sum. The goal is to make that sum equal to a hidden number. Participants are given feedback about the relationship between the sum of their guesses and the number that is hidden.

In Musical Choice participants listen to and rate music. Players are shown others' ratings; this information can shape their own.

"I hope that the combination of serious experiment and somewhat interesting game will make the experiments attractive," said Goldstone. "People spend an awful lot of time playing Solitaire and society really doesn't benefit all that much. I'm hoping that some people will decide to play our games instead so that science can benefit."

"The result I find most interesting is that precisely because people share the desire to avoid crowds, migratory crowds emerge!"

-Reporting from San Diego, GP Correspondent Andrew Eisen

Tags: indiana university, rob goldstone

  • Post a new comment


    Anonymous comments are disabled in this journal

    default userpic

    Your reply will be screened

    Your IP address will be recorded