Correcting misinformation may be ineffective because people continue using it

Evidence

This is what participants were asked after reading the above story in a seminal misinformation experiment conducted by two psychologists at the University of Michigan in 1994. Having read the story in thirteen staggered messages, like a breaking news report, participants were invited to freely write out their thoughts to answer these as well as some other factual questions. However, in just one of the conditions – the correction group, who were presented with the story exactly as it appears above – the participants learned that the ‘closet reportedly containing the cans of oil and paint gas cylinders had actually been empty before the fire’. No reference whatsoever was made to this incorrect claim in the control group; it was simply stated that the closet in the warehouse hall was empty. In other words, only the correction group were provided with updated and new information about the oil and gas cans. — Sander van der Linden, Foolproof, 249

Categorization

  • XREF: The continued influence effect of misinformation studied by Ecker/Lewandowsky and parallels work on belief updating failure in the trust literature. — connects to: Doctored videos can be shallowfakes, simple out-of-context cuts reused as misinformation — Sander van der Linden, Foolproof: Why Misinformation Infects…, 355; Fixing misinformation requires changing revenue incentives, not just content exposure — Sander van der Linden, Foolproof: Why Misinformation Infects…, 470; Fact-check warnings fail to stop repeated fake headlines gaining perceived truth — Sander van der Linden, Foolproof: Why Misinformation Infects…, 70 (+10 more)

Broader Connections