The minute-scale dynamics of online emotions reveal the effects of affect labeling

Rui Fan*, Onur Varol, Ali Varamesh, Alexander Barron, I.A. van de Leemput, M. Scheffer, J.L.T. Bollen

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

21 Citations (Scopus)


Putting one’s feelings into words (also called affect labeling) can attenuate positive and negative emotions. Here, we track the evolution of specific emotions for 74,487 Twitter users by analysing the emotional content of their tweets before and after they explicitly report experiencing a positive or negative emotion. Our results describe the evolution of emotions and their expression at the temporal resolution of one minute. The expression of positive emotions is preceded by a short, steep increase in positive valence and followed by short decay to normal levels. Negative emotions, however, build up more slowly and are followed by a sharp reversal to previous levels, consistent with previous studies demonstrating the attenuating effects of affect labeling. We estimate that positive and negative emotions last approximately 1.25 and 1.5 h, respectively, from onset to evanescence. A separate analysis for male and female individuals suggests the potential for gender-specific differences in emotional dynamics.
Original languageEnglish
Article number18911
Pages (from-to)92–100
Number of pages9
JournalNature Human Behaviour
Early online date17 Dec 2018
Publication statusPublished - 2019


Dive into the research topics of 'The minute-scale dynamics of online emotions reveal the effects of affect labeling'. Together they form a unique fingerprint.

Cite this