Your mood improves immediately after reporting on social media that you feel bad. These are the findings of an analysis of millions of tweets, published in the journal Nature Human Behaviour.
A team of researchers at Indiana University and Wageningen University & Research, headed by Prof. Johan Bollen, analysed the mood of almost 75,000 people, as reflected in their tweets. They saw a steady decline in mood before people posted tweets such as ‘I feel really bad’. However, the tweeters started to feel better as soon as this fall in mood prompted an articulation of that mood (the tweet).
In other words, negative emotions built up slowly until they were expressed in a tweet, after which they then disappeared again very quickly. The progression was different for positive tweets (‘I feel really happy’). ‘We find a marked difference between the course of positive and negative emotions,’ says Wageningen researcher Ingrid van de Leemput. The expression of a very positive emotion on Twitter was preceded by a brief increase in positivity in the tweets. Once the emotion was articulated, it gradually faded again. ‘We therefore see that as soon as emotions – either negative or positive – are expressed, they become neutralised.’
The expression of negative feelings is shown to be most helpful: emotions were neutralised much more quickly after a very negative tweet. We suspect that this occurs through the act of articulation, as well as because of the responses to the “I-feel-really-bad” tweet’, Van de Leemput explains. ‘In general, we also saw that tweets from women were on average somewhat more positive than tweets from men’.
Labelling your emotions reduces their impact
The research emphasises the power of ‘affect labelling’, or naming your emotions. By giving it a name, you reduce its impact. ‘So you feel awful, say it out loud! Or write it “out of your system” and post it on Twitter.’
The study compared the timelines of 74,478 tweeters who used specific expressions in English (such as, ‘I feel really bad’). Using a computer algorithm that took account of word use, emoticons, slang and sentence structure, a mood value was then assigned to every tweet in the timeline of that Twitter user. ‘Tweets that are about something other than your own feelings also tend to have an undertone, which says something about your mood’, says Ingrid van de Leemput. In total, more than a billion tweets were analysed. The tweets expressing the emotion were not included in the analysis but served as the ‘mid-point’. For each minute in the timeline, the team mapped out the average mood of all these Twitter users, from two hours before to two hours after the tweet expressing the emotion. They found that an emotion lasted for about one and a half hours on average.
The minute-scale dynamics of online emotions reveal the effects of affect labeling. Rui Fan, Onur Varol, Ali Varamesh, Alexander Barron, Ingrid A. van de Leemput, Marten Scheffer & Johan Bollen. Nature Human Research, December 17. 2018.