Online curiosity explored: 'We are more likely to accept information uncritically if it answers a question'
What do people wonder about on social media? University lecturer Matthijs Westera is the recipient of an NWO grant to investigate what people are curious about online.
If you want to tell someone something that is not quite right, your best plan is to answer questions that the person probably has anyway. 'We are more likely to accept information uncritically if it answers a question we already had,' Westera explains. 'We want information so badly that truthfulness seems to count for less. This is known as the need for closure bias. It makes questions a potential gateway for disinformation.'
Search of social media
This makes it all the more striking that until now there has been very little research into what questions people ask on social media, even though it is a breeding ground for disinformation. 'There is a lot of research on the spread of information on social media,' says Westera. 'Sentiment on social media has also often been studied, but there is little research on curiosity on social media. So, a year ago in the Research Traineeship, I did a project with students in which they investigated whether you find a different kind of question in disinformation than in true information. Indeed, we discovered some general trends in the wording of questions, such as more conjunctions, more 'cognitive' words like think, and fewer negations like not. Now I want to go a step further and also start looking at the topics about which questions are asked, such as a politician or a measure.'
Wave of questions, wave of information?
Westera has received an Open XS grant from NWO to carry out this research. The money will enable him to use AI techniques to comb Twitter for six months. 'I want to know, for example, whether a wave of questions is followed by a wave of information or whether it happens more simultaneously,' he says. 'I can do that on a collective level, but I can also focus on an individual level: if someone posted a question in the past, are they more likely to share an answer later? Next, I want to see if I can also chart more general changes in curiosity. Can we say, for example, "On 13 December, we see a shift from causal questions to responsibility questions?"'
Give it our best shot
Westera has not yet decided the topic on which he is going to answer these questions. 'I’ve worked on COVID disinformation before, but my main interest is in the underlying mechanisms, such as the need for closure bias. I'm looking forward to getting my teeth into that for a year and, together with a research assistant, we’re going to give it our best shot.'