Skip to main content

The Ultimate Cascade Culprit

In this age of the internet, it is easier than ever before to find accurate and reliable information about almost any conceivable topic.  One should be surprised then, if some “viral” information turned out to be false, or a hoax.  If such information was untrue or unreliable, why would so many people spread the information when a 30 second Wikipedia trip might give clear verification?  The problem is that while checking Wikipedia or some other popular internet authority might take 30 seconds, hitting “Like” on Facebook or “Retweet” on twitter takes only two or three.  While information on the internet is more open to scrutiny than any information ever has been in history, it is also significantly easier to spread than verify.  Therefore the internet has, unfortunately, caused an increase in information cascades despite its ability to stop them.

In a blog post on, Albert Wenger discusses an example of such a cascade.  Shortly after the US raid that claimed the life of Osama bin Laden, Wenger noticed that an MLK quote was seeming to spread rapidly across the web among reactions to the raid.  It is fairly safe to assume that very few people initially used the quote, then slowly more and more people saw the quote and used it themselves.  The problem is, the quote was not an actual MLK quote, and anyone using it could have easily checked to see if it was.

This type of cascade is incredibly common among social media and thrives through network effects.  People who see information in online documents about topics they’re interested in or who see their friends post information to social media are likely to trust the information because they trust the source.  Once a cascade begins, there are enough “trustworthy” sources that most people do not feel a need to fact check the information.  This is similar to the homework problem in which people were attempting to decide whether to accept an invitation to a party.  Even if a person’s initial thought was not to go, if they saw that a lot of other people were going, they would assume that those people must have information they don’t that said the party would be good.  Similarly, if a person’s initial thought is to fact-check a quote or event or some other piece of information, if they see that a lot of people are talking about the information, they will be very likely to assume the information is safe and that they can talk about it themselves.

As shown in lecture and through the homework, information cascades can begin incredibly quickly.  Most of the examples in class and in the homework only dealt with small groups of people or a line of people going in a particular order.  On the internet, there are millions of individual people in no order.  Even if just one person throws out bad information, it is incredibly likely that at least multiple of the many people who see it will not take the time to verify the information and will simply “share” it or spread it in some form.  For each of those people the same process occurs.  On the internet, information cascades do not occur linearly, they occur exponentially.

As a result of these network effects and the simplicity of information distribution offered by the internet, it is far easier to start cascades on the internet than to stop them, and the only way not to fall into them, according to Wenger, is to make the decision to check any and all information found online.


Leave a Reply

Blogging Calendar

November 2014
« Oct   Dec »