Information Sorting
Information sorting is based on the idea that while no sources are entirely unbiased or reliable, some are significantly more reliable than others. Its focus isn’t on debunking false information but on discernment between sources of varying levels of reliability.[2]
It follows a two-step process: starting with companion reading to winnow out unreliable sources, followed by a close reading only of sources that are worth our critical attention. This combination prepares people to deal both with genuine misinformation and with information from legitimate sources that may nevertheless be biased, incomplete or framed in a misleading way.[3]
Companion reading is just about determining whether a source is basically reliable – whether or not you can count on them to give you accurate facts. To do that, we mostly don’t look at the source itself at all, but on what other sources that we know are reliable say about it. Because it’s a way of dealing with “too much information,” companion reading has to be quick and simple.
This technique, which is sometimes called lateral reading, works by using companion texts – things like search engines, encyclopedias and sources you already know are reliable – to find relevant and reliable sources and to find out if sources that come to us are worth our attention. Just as important, companion reading tells us which sources we’re better off ignoring. This is particularly helpful when it comes to AI-generated content because it really has no original source. Doing a reverse image search for a real photo will lead to the actual places that photo appeared, while searching for a fake one is likely only to lead to social media posts and articles telling you it’s a deepfake.
Companion reading also helps us avoid what information literacy expert Mike Caulfield describes as trust compression: rather than applying skepticism to distinguish between sources that are more or less trustworthy, or to recognize the ways in which different sources are biased, people may conclude that no sources are trustworthy and that all are equally – and fatally – biased.
MediaSmarts’ Break the Fake program covers this step in a lot more detail, but briefly, we teach four information sorting steps:
- Using fact-checking tools;
- Finding the original source of a claim;
- Verifying that source, if you don’t recognize it as being reliable or unreliable;
- And checking against other sources, particularly ones that you know are reliable, expert or authoritative.
Just one or two of these steps will typically be enough to find out whether a source is reliable enough for close reading. You can learn more about each one in the Companion Reading section.
Close reading is much more complicated. It’s about looking at how a source frames its content: things like bias and emphasis, what it includes and what it leaves out, what’s highlighted and what’s minimized, which photo is used to illustrate a story, and so on.
Another fundamental close reading skill is knowing how different genres and text forms work. This ranges from basic things like how to use the index and table of contents of a textbook to more specialized formats like the “inverted pyramid” structure of a news article.
To do any of that we have to look closely at the source itself, but to consider some questions – like what might have been left out – we might compare it to other sources, as well.
We only do a close reading of something that we know is worth our attention – either because we already know it’s reliable, or because we’ve verified it through companion reading. As Mike Caulfield and Sam Wineburg say in their book Verified, “Thinking critically about low-quality sources is a colossal waste of time.”[4]
[1] Hassoun, A., Beacock, I., Consolvo, S., Goldberg, B., Kelley, P. G., & Russell, D. M. (2023, April). Practicing Information Sensibility: How Gen Z Engages with Online Information. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (pp. 1-17).
[2] Kiili, C., Räikkönen, E., Bråten, I., Strømsø, H. I., & Hagerman, M. S. (2023). Examining the structure of credibility evaluation when sixth graders read online texts. Journal of Computer Assisted Learning, 39(3), 954-969.
[3] Hoes, E., Aitken, B., Zhang, J., Gackowski, T., & Wojcieszak, M. (2024). Prominent misinformation interventions reduce misperceptions but increase scepticism. Nature Human Behaviour, 1-9.
[4] Caulfield, M., & Wineburg, S. (2023). Verified: How to think straight, get duped less, and make better decisions about what to believe online. University of Chicago Press.