The mnemonic effect of central and peripheral misinformation on social media

ABSTRACT

The increasing use of social media has amplified the spread of false information. Yet little is known about the mnemonic consequences associated with exposure to different types of false information online. In two studies, we examined in a simulated online context how exposure to false information either central or peripheral in events affected memory. European American and Asian/Asian American college students (Study 1 N = 200; Study 2 N = 225) were presented with GIFs of daily life events and read tweets about the events that included four types of information: central true information, central false information, peripheral true information, and peripheral false information. They then took a True/False recognition test that included tweeted and untweeted true and false information and indicated how confident they were in their responses. Regardless of cultural background, participants in both studies demonstrated the misinformation effect, whereby they falsely recognised more and resisted less tweeted than untweeted false information. Furthermore, they showed higher susceptibility to peripheral than central false information exposed via tweets. Asian participants were less influenced by misinformation than European Americans in Study 2. These findings have important implications to combat misinformation in online environments.

KEYWORDS:

• Misinformation effect; misinformation resistance; information centrality; online; culture

Open Scholarship

This article has earned the Center for Open Science badges for Open Data and Open Materials through Open Practices Disclosure. The data and materials are openly accessible at https://osf.io/2f9gx.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Research materials and data are available at https://osf.io/2f9gx.

Notes

1 GIFs are a looped sequence of static images that are extensively used for communication on social media platforms. They can convey messages quickly in a virtually engaging way.

2 Note that d’ scores were not used for the recognition test data because correct and incorrect responses are complementary in a forced-choice task, where analysing one type of responses provides information about the other (Mahé et al., 2015). We therefore measured memory accuracy based on correct responses following prior research Dalton & Daneman, 2006; Daneman et al., 2013; Mahé et al., 2015).

Additional information

Funding

This research was supported by funding from the College of Human Ecology at Cornell University to Qi Wang.