40 — Competition and Success in the Meme Pool: a Case Study on Quickmeme.com

This paper explores the data available from Quickmeme.com, an internet-meme creation website. Coscia likely chose this resource because of the strong time-stamping and usage metadata associated with each creation: This allows some degree of use-tracking/

Perhaps one of the most delicious parts of this paper is when Coscia explains appropriate usage of different memes. A real excerpt:

“Ridiculously Photogenic Guy” (RFG), instead, has an immediate peak larger than BLB, but it then fades into oblivion as an ephemeral trend.

Coscia downloaded around 179,000 meme “implementations” (unique text and creation-date) and tracked them over usage for several months. The popularity data follow a power law curve: There are a handful of extremely popular memes, with a quick falloff. Memes that make it to the front page of the site have a greater-than-expected response, due to the heightened exposure — in what Coscia calls the “front page effect”.

Coscia’s model of meme spread points out a few intuitive but interesting features of the data: Different memes compete for attention; or similar memes boost each others’ popularity by mutual exposure. (The math is not too hard to follow, so if this sounds like a thing you want to read more about, this is a great paper to check out.)

There were further time- and neighbor-dependent patterns: Higher popularity peaks appear to paradoxically reduce the long-term success rate of memes: High-peak memes have a 13% success-rate; low-peak memes have a 55% success-rate. Memes with similar neighbors (similar text or imagery) tend to collaborate in popularity-level, and survive for longer, with greater resilience to becoming outdated.