Skip to main content

Information Cascades and the Statler Auditorium Upper Deck Exit Plan

Every Monday, Wednesday, and Friday my friend and I sit in the upper deck of the Statler Auditorium for Networks. There’s always more seats if you get to class late, and the projection is just about at eye level if you sit in the right place. Our only issue with sitting in the upper deck is the difficulties we encounter when trying to exit the class. The major staircase leading up to the upper deck is always full of students coming out of class to the point where it takes a long time to actually get off of that floor and out of the building. Seeing this, my friend and I did a little exploring and found a smaller staircase immediately to the right (as you walk out of the lecture hall) that we could use to get down to the first floor. Since then, we only ever take the new staircase and find it to be significantly faster than the other. We assumed early in the semester that it was only a matter of time before the staircase we took became just as crowded as the other staircase but so far have been completely wrong. The congestion levels of our staircase haven’t changed since the first day of class.

When we first thought about what would happen to our route we thought of it in terms of common sense. We figured that more people would start to take it as they saw us and any others taking it and getting out faster. As the class progressed, it has become more of a traffic in networks problem. Each staircase has a different throughput, the main staircase being much larger than the other, so the travel time would be based off of the number of people choosing each route. We learned in class that we can reasonably expect an equilibrium to form such that the travel times for both routes would be equal. This equilibrium is something we’ve been expecting for weeks but it still hasn’t come. Some of the newer material covered in class has helped shed some light on the situation though, and I feel my friend and I may be witnessing an ongoing information cascade.

I believe what we have is a case of herding, whether it be directly influencing the actions of some of my classmates or indirectly. Each day we have class people start to get ready to leave slightly before the class has officially ended. These individuals are beating the rush of people that occurs when such a large lecture is let out. What these individuals are also doing however, is displaying their choice of route to all of the subsequent individuals leaving the auditorium. For the early individuals the path they choose is relatively empty so it makes rational sense to take.  Once others leave the auditorium they have their own private information (how long they think it will take to get downstairs versus how much of a rush they are in), and they also have the decisions made by all the people before them. I believe that in this situation, seeing everyone else going down the main stairway is outweighing any private information that anyone is holding. One could argue that people may simply not know about the alternate pathway, but there is a significant number of people that do take it that I would argue more people must know about it. The situation reminds me of the Milgram, Bickman, and Berkowitz experiment mentioned in the textbook:

“They found that with only one person looking up, very few passersby stopped. If
five people were staring up into the sky, then more passersby stopped, but most still ignored
them. Finally, with fifteen people looking up, they found that 45% of passersby stopped and
also stared up into the sky.”

In this case, there are simply not enough people taking the alternate route for the people taking the busier route to assume that those taking the alternate route know something that they (busier route people) don’t know. This leads to an information cascade and thus the main staircase becomes congested. If more people took the alternate path it stands to reason that, over time, and equilibrium would form and a matching would be reached that is globally optimal.

The result of all this is that the traffic shift my friend and I have been waiting for hasn’t happened and we still enjoy a much faster route out of the building. Now that I’ve posted this in the blog I’ll be interested to see if anything changes in the coming weeks. Regardless it seems in this case my friend and I have found our dominant strategy and now we know why.


Leave a Reply

Blogging Calendar

November 2014
« Oct   Dec »