Next month, I will give two talks – or two versions of the same talk – on “Cognitive attraction and online misinformation”. One will be in Den Bosch at the Jheronimus Academy of Data Science (where I hope to convince data scientists that cultural evolution and cognitive anthropology can be useful to understand online diffusion dynamics) and one, shortly after, at a Conference on Cultural Evolution organised by The Cognition, Behavior & Evolution Network at the University of Antwerp (where I will do the opposite, hoping to convince cultural evolutionists that studying online diffusion dynamics can be useful for us).
Here is the abstract:
“The spread of online misinformation has gained mainstream attention in recent years. Here I approach this phenomenon from a cultural evolution and cognitive anthropology perspective, focusing on the idea that some cultural traits can be successful because their content taps into general cognitive biases. I analyse 260 articles from media outlets included in two authoritative lists of websites known for publishing hoaxes and ‘fake news’, tracking the presence of negative emotional content, threat-related information, elements associated to disgust, presence of sexually related material, minimally counterintuitive elements (and a particular category of them, i.e. violations of essentialist beliefs), and social information. The analysis shows that these features are, to a different degree, present in most texts, and thus that general cognitive inclinations may contribute to explain the success of online misinformation. I conclude discussing how this account can elucidate questions such as whether and why misinformation online is thriving more than accurate information, whether misinformation spreads better online than offline, or the role of ‘fake news’ as a weapon of political propaganda.”
Hopefully, I will have soon a manuscript on the above, to submit to Jamie Therani’s Cultural Evolution Research Collection at Palgrave Communications. In the meantime (this is the reason why I am writing this short post), I put the material – the links plus the text files of 260 ‘suspect’ news plus the coding – in an OSF repository, so that it is freely accessible. It is a small dataset (I wanted to be able to read the actual news) but if anyone is interested it can be downloaded from there (perhaps contact me if you have ideas/feedbacks!)
update November 2018: preprint available on PsyArXiv
[Yes, Den Bosch is where Hieronymus Bosch spent most of his life]