Fact checking and group think

People fact-check less often when they are in a collective setting.

Researchers at Columbia Business School conducted eight experiments to test how the context in which people access information affects their fact-checking behaviour.

The researchers surveyed over 2200 U.S. adults via Amazon Mechanical Turk. Participants logged onto a simulated website and evaluated a series of statements. These statements consisted of ambiguous claims (of which half were true and half were false) on a range of topics.

Example statements provided for review

“Scientists have officially declared the Great Barrier Reef to be dead”

“Undocumented immigrants pay $12 billion a year into Social Security”

Key finding

The experiments show that people fact-check less often when they are in a collective setting. Simply perceiving that others are present reduces people’s vigilance when they are processing information.

The experiments

Participants could mark the statements as ‘true’, ‘false’ or flag them for fact-checking.  The researchers tested a range of incentive models including giving participants points for flagging; penalties for flagging; points for performance.  They found that different incentive models did not change the overall patterns.

In one experiment, participants gave their responses to 36 statements described as news headlines published by a U.S. media organisation. Throughout the task, half the participants saw their own username displayed alone on the side of the screen, while the other half also saw those of 102 respondents described as currently logged on, presumably completing the same task. People flagged fewer statements when they perceived that others were present.

In another, the researchers displayed the statements as headlines appearing on a news website and on a news organisation’s Facebook feed.  On the traditional news site, people flagged less often if they perceived others undertaking the same task. But on Facebook, people flagged less often irrespective of whether they saw other people undertaking the same task.

Even when participants were told that user names showing on the screen may belong to people who completed the task weeks ago, they were less likely to flag content.

Browsing information on social media, an inherently social context, seemed to make individuals behave as if they were in a group.

A number of behavioural factors might be in play here, including the bystander effect (if other people are checking this why should I?) or the fact that simply being in a group may lead to people lowering their guard (safety in numbers).

The original research is published in the Proceedings of the National Academy of Sciences (PNAS).

Source: Harvard Business Review