Brief exposure to short video clips of misinformation strategies improves awareness of online falsehoods. The corresponding study was published in Science Advances.
'Inoculation theory' has been suggested as a means to curb susceptibility to ‘fake news’. Just as a vaccine works by introducing a 'sample pathogen' to the body's immune system to train against future intruders, 'inoculation theory' works by informing people of different misinformation tactics so they can better recognize and resist them in the future. While theoretically effective, until now, the theory has been studied in limited settings.
In the present study, researchers from Cambridge University and the University of Bristol worked in partnership with Google’s Jigsaw to test five short videos aiming to ‘inoculate’ people against some of the most commonly-used misinformation techniques. These include emotionally manipulative language, incoherence, false dichotomies, scapegoating, and ad hominem attacks.
The researchers initially conducted six controlled experiments including 6,464 participants. Across these studies, participants were randomized to watch either a 1.5-minute inoculation video or a neutral control video of roughly equal length. They were then asked to rate ten synthetic social media posts as either ‘manipulative’ or ‘neutral’.
From these studies, the researchers found that participants in the inoculation group were significantly better at discerning manipulative content from neutral content.
The researchers next tested the inoculation videos on 5.4 million US YouTubers. Around a million of these viewers watched the videos for at least 30 seconds. 30% of these users were then given a random test question within 24 hours to see if they could identify misinformation. YouTube also gave a random control group of users who had not viewed an inoculation video the same test question. Altogether, 22,632 users answered this question.
From this experiment, the researchers found that those who had 30 seconds or more exposure to the inoculation video were 5% more likely to recognize misinformation techniques that controls.
The researchers noted that such an intervention could be ‘game-changing’ if scaled up across platforms. They also noted that it could be relatively cheap to deliver- at just US$ 0.05 per 30 seconds or more of viewing.
They wrote, however, that while their findings advance misinformation research, their findings are limited as they were not able to study the long-term effects of inoculation. They added that as inoculation videos were tailored to US audiences and only tested on US audiences, their results may not translate to other populations.