Happiness and surprise are associated with worse truth discernment of COVID-19 headlines among social media users in Nigeria

Leah Rosenzweig; Bence Bago; Adam Berinsky; David Rand; Vicky Gabriel

In brief

Among 1,341 Facebook users in Nigeria, experiencing emotions—especially happiness and surprise—was associated with worse ability to distinguish true from false COVID-19 headlines. Respondents who felt emotions were more likely to believe and share false headlines relative to true ones, and emotional reactions were associated with greater overall engagement (belief, clicking, sharing) regardless of veracity. Older, more reflective respondents and those without ruling-party affiliation showed better discernment.

Key claims

Emotional reactions impair discernment: Experiencing any emotion after reading a headline was associated with greater belief in false headlines relative to true ones. Respondents showed the best discernment when they felt no emotion.
Happiness and surprise amplify false-news sharing: Of six measured emotions (fear, happiness, surprise, sadness, anger, disgust), only happiness and surprise were significantly associated with worse truth discernment and greater sharing of false relative to true headlines.
Emotions increase overall engagement: Feeling any emotion was associated with greater likelihood of believing, wanting to click, and wanting to share headlines—both true and false.
Individual differences in discernment: Older respondents, those with higher cognitive reflection (analytic thinking), and those not supporting the ruling party showed significantly better truth discernment of COVID-19 headlines.
Limited effectiveness of religiosity and partisanship as moderators: The study found negative associations between religiosity and discernment, and between ruling-party support and discernment, suggesting partisan and religious factors influence COVID-19 misinformation susceptibility even in contexts with weaker party identification than the U.S.

Sources cited in the article

The article references prior work on emotions and misinformation:

Martel et al. (2020) on reliance on emotion promoting fake news belief
Pennycook & Rand research on analytic thinking and discernment
Lerner & Tiedens (2006) on emotion and cognition
Holland et al. (2012) on mood and attitude-behavior links

Notes

This is the first study to examine specific emotions (rather than general emotional arousal) in relation to COVID-19 misinformation susceptibility in a non-Western sample. The finding that happiness and surprise—rather than anger and fear—drive worse discernment differs from some prior work and may reflect COVID-19-specific emotional responses or differences in Nigerian vs. U.S. samples. The sample is more educated than the Nigerian population average (76% with some university education), raising generalizability questions but also suggesting even educated, urban populations remain vulnerable to emotionally driven misinformation. The study reports pre-registered hypotheses and provides open materials and data. Limitations include reliance on self-reported emotional states and stated sharing intentions rather than actual sharing behavior, and the focus on COVID-19 headlines limits generalization to other domains of misinformation.

How it connects

Builds on Reliance on emotion promotes belief in fake newsMartel et al. · Cognitive Research: Principles and Implications · 2020 Related Beliefs about COVID-19 in Canada, the U.K., and the U.S.A.: A novel test of political polarization and motivated reasoningPennycook et al. · Personality and Social Psychology Bulletin · 2021