The Birth of Memes: Modeling Memes as Information Cascades on Social Networks
Article: https://www.vice.com/en_us/article/kwx3jv/we-asked-an-expert-why-some-memes-work-better-than-others
Every young adult who uses social media knows what memes are. Although they aren’t keen to admit it, many spend significant portions of their day perusing memes. Thousands of sites and groups online specialize in catering memes to niche audiences. Indeed, most Cornell students are familiar with “Make Cornell Meme Again”, a Facebook page dedicated to memes for Cornell students.
The idea behind a meme is simple: a photo is captioned with humorous content and goes viral as it is shared across social platforms. Others adopt the same image with different captions, and a meme is born. New memes with diverse origins become popular on a monthly basis. However, the growth of these memes has some common qualities that resemble those of a decision making process that depends on an information cascade.
“Bad Luck Brian”, a popular meme from 2012
The rise in popularity of a meme is similar to an information cascade. In this case, the probability of a positive response is high for many of the first users, such that their action is to share the meme. The article describes several factors that decide the popularity of a meme. Two of these stand out: absurdity and relatability. The first makes sense- absurdity is closely linked to humor, so good memes must be funny. Relatability is more intimately connected to the information cascade. A popular meme must be understood by some audience in order to be shared. In other words, a joke will only be funny to people who get the joke. This will have an obvious contribution to the likelihood the meme is shared. While sharing a meme, a social media user in most cases consciously decides whether his/her friends will find the content meaningful or funny, which is like a private signal. Thus, like in other information cascades, a decision is made with rational inference.
“Y U No Guy”, a popular cartoon meme
A meme can be targeted towards a specific audience or it can become a generalized Internet phenomenon, depending on its origin. Some images are naturally easier to turn into memes for the general public than others. For example, an image featuring McGraw Tower will likely be meaningful only to the Ithaca and Cornell community. Therefore, the information cascade determining the longevity of a meme is limited by those users who satisfy some relatability factor that is inherent to the individual. More simply, someone who isn’t a part of the audience for the meme is unlikely to share it. Therefore the most successful memes have images that can be broadly interpreted and modified with unique captions, shared on platforms that are open to the public and easy to access, which is mentioned in the article. Thus, the model of information cascades is limited to this group of people who are at least somewhat familiar with the meme content and have access to it. In other words, the private signal probably carries a much higher weight than the signals from previous users in determining whether a particular user will share a meme.
The sharing of memes is a good modern demonstration of the propagation of information in information cascades. Though some limits exist, it can be seen that the growth of a meme is similar to the way information is shared in a cascade. Specifically, each user forms an opinion of the meme based on how visible it is to them (how many previous users have shared it) as well as how relevant it is to them and their audience. Like herding effects mentioned in lecture, the popularity of a meme is directly affected by how many previous members of a group have shared it, where the group can be a specific audience or the Internet at large. Lastly, because relatability is an important factor in the success of a meme, an individual’s personal opinion of the meme is likely to be more important than the opinions of the previous sharers of the meme, so the information cascade must be modified to some extent. Nevertheless, the birth and growth in popularity of a meme is very much like an information cascade.