Human Computation, Game Design, and Behavioral Economics

April 24, 2007 § 1 Comment

The philosophies generally associated with the University of Chicago usually orbit around free markets, rational actors, and economic efficiencies. These were generally borne out of the Friedman & Stigler cohort and the Chicago School of Economics which reached its peak during the mid-80’s still is a dominant force in economics, politics, and law today.

An interesting spin-off of the classic Chicago school of thinking here has been the Behavioral Finance movement piloted by Richard Thaler in the GSB and by Cass Sunstein in the School of Law. The behavioralists generally think that free markets are all well and good, but that people simply don’t always act in perfectly rational ways. They try to and categorize ways in which people generally deviate from rational behavior so that modeling and predictive techniques can be formed around how people DO act, as opposed to how they SHOULD act. One of the classic models that has emerged from this school of thinking is the Prospect Theory, which models the fact that humans are generally more adverse to losses than they are to gains, i.e., a normal person would feel more pain from loosing $100 than joy they would feel from gaining $100, while a perfectly rational actor would give the same slope coefficient to losses as they would to gains.

Another interesting model that has come from this school of thought is the impact the framing of a question or situation has on the response it generates. For example, people are likely to not be neutral between a 25% chance to win $50 and a 50% chance to win $25, even though the expected payouts are the same ($12.50). Similarly, studies have shown that people tend to anchor expectations to numbers they have recently heard or seen. Smart attorneys attempt to use this heuristic in their closing arguments by referencing numbers near where they would like the jury award to fall, e.g. a attorney who wants a hundred million dollar reward in a pharmaceutical trial would be smart to talk about the hundreds of millions of people who could have been potentially hurt by a drug, thus anchoring the jurors’ minds around numbers of that size.

Anyways, what got me thinking about this whole subject was a reference to the Mechanical Turk on Guy Kawasaki’s blog. Now Guy was a little late to the Mechanical Turk party, as the comments to his post point out, but one thing I noticed in the comments were several references to the Mechanical Turk being a failure. I hadn’t given it much thought before I watched the video below. The classic Mechanical Turk task is paying users $0.005 to tag , or describe, a photo with text. Until a photo has been tagged, a computers currently have no way of telling if the picture is of a boy, a dog, or a airplane. There are some people that will tag photos for pennies, but it isn’t terribly rewarding and it’s not a great way to make any money. But, as you’ll hear in the video, if you frame the same task in the context of a game, you end up having to cut people’s playing time off after fifteen hours because they like it so much.

The video is long, but basically this guy Luis Von Ahn designed the ESP Game and Peekaboom which tagged more photos for free than the Mechanical Turk will ever even come close to seeing. Some recent posts on the Lightspeed Ventures blog have also piqued my interest about game design and its place in social networking and web applications in general. One of the posts is about Yelp harnessing game design to get users to do what they want them too (create listings, contribute reviews, etc.). I think LinkedIn has done a fabulous job of this as well. I still love making connections on LinkedIn but never once have I really gotten any real benefit from it…I just like watching my counter go up. Yahoo! Answers and Amazon’s Askeville are also great examples of game design getting people to do things. Every day on thee sites thousands of people answer questions posed by complete strangers to accrue millions of points that have no more value than a video game score. Nutty.

Edit: Thanks to Jeremy Liew at Lightspeed for the shout out about this post.


§ One Response to Human Computation, Game Design, and Behavioral Economics

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

What’s this?

You are currently reading Human Computation, Game Design, and Behavioral Economics at robwebb2k.


%d bloggers like this: