Skip to main content

Sentiment and Presentiment in Twitter: Do Trends in Collective Mood "Feel the Future"?

๐Ÿ“„ Original study โ†—
Radin, Dean I โ€ข 2023 Current Era โ€ข precognition

๐Ÿ“Œ Appears in:

Plain English Summary

Can the collective mood of millions of Twitter users sense bad things before they happen? This study analyzed 13 years of daily mood scores from tweets in 10 languages before 83 surprise tragedies like terrorist attacks and mass shootings. Across all 10 languages, mood consistently dipped in the two weeks before events. The combined result was highly significant (p = 0.001), meaning pure chance is very unlikely. Removing English (used to build the method) and retesting only the remaining 9 languages gave an even stronger result. First-ever test of "collective presentiment" (group-level gut feeling about the future) outside a lab, using publicly verifiable data.

Abstract

Meta-analyses of experiments investigating human behavioral and physiological reactions to unpredictable future events suggest the existence of a poorly understood ability to "feel the future." Is this effect reflected in sentiment metrics based on social media posts? To find out, analysis of 13 years of daily Twitter sentiment data in 10 languages was examined two weeks prior to events assessed as significantly negative and unpredictable, including acts of terrorism, mass shootings, unexpected deaths of celebrities, etc. Results of the analysis was statistically significant (p = 0.001), suggesting the existence of a form of collective presentiment.

Links

Related Papers

Also by these authors

More in Precognition

๐Ÿ“‹ Cite this paper
APA
Radin, Dean I (2023). Sentiment and Presentiment in Twitter: Do Trends in Collective Mood "Feel the Future"?. World Futures. https://doi.org/10.1080/02604027.2023.2216629
BibTeX
@article{radin_2023_sentiment,
  title = {Sentiment and Presentiment in Twitter: Do Trends in Collective Mood "Feel the Future"?},
  author = {Radin, Dean I},
  year = {2023},
  journal = {World Futures},
  doi = {10.1080/02604027.2023.2216629},
}