In Defence of Liberty

Driven by data; ridden with liberty.

How Misinformation Spreads Online


Share like you just don’t care.

How does misinformation spread online? In a world where everyone with an internet connection can publish, and share with their friends and followers, the capacity for information, news, falsehoods and rumours to diffuse within a social network is evident.

In the United States, a military training exercise, Jade Helm 15, generated conspiracy theories about the US government preparing for martial law, or civil war [1]. In 2013, the World Economic Forum declared that “massive digital misinformation sits at the centre of a constellation of technological and geopolitical risks” [2].


A paper published in the Proceedings of the National Academy of Sciences considers this question, studying five years of user behaviour on the popular social network Facebook [3].

The researchers, led by Walter Quattrociocchi, looked at three categories of posts: conspiracy theories, scientific news, and “trolling” or satire. The first category is defined as pages that “disseminate alternative, controversial information, often lacking supporting evidence and frequently advancing conspiracy theories”. The latter category is used to check the data-driven percolation model.

What they found is that “homogeneity is clearly the driver of information diffusion”, meaning that Facebook users for these types of content share and like with each other. This is often in concordance with pre-existing beliefs — confirmation bias — as “users mostly tend to select and share content according to a specific narrative and to ignore the rest”. Consequently, this creates a polarised and homogeneous clusters, commonly called echo chambers.


Indeed, the consumption patterns for both conspiracy theories and scientific information on Facebook were similar, exhibiting “a probability peak in the first 2h, and then in the following hours they rapidly decrease”.


The probability distribution functions are similar. (Source: PNAS)

It is the cascade pattern, how conspiracy theories and scientific news are shared, that differs. In the paper’s conclusion:

Science news is usually assimilated, i.e., it reaches a higher level of diffusion, quickly, and a longer lifetime does not correspond to a higher level of interest. Conversely, conspiracy rumors are assimilated more slowly and show a positive relation between lifetime and size.


The researchers found that, by simulating users sharing information if it is close to their own opinion, an accurate model for the observed data can be created.

Whilst science news is shared quickly, and then dwindles, conspiracies burn in the tinder and inflame over time. This is how misinformation spreads online: rebounding around the echo chamber, before it diffuses into the wider world. The mutating whispers have become digital.


[1] Lamothe, D., 2015. Remember Jade Helm 15, the controversial military exercise? It’s over. Washington Post. Available from: [Accessed: 10th January 2016]

[2] WEF, 2013. Digital Wildfires in a Hyperconnected World. Available from: [Accessed: 10th January 2016]

[3] Quaatrociocchi, W., et al, 2015. The spreading of misinformation online. PNAS. Available from: [Accessed: 10th January 2016]



This entry was posted on January 13, 2016 by in Social Media and tagged , .
%d bloggers like this: