Science: Tweets Reveal the Role of Circadian Rhythms on Mood

It seems that Twitter—the micro-blogging Web site that records brief, time-stamped public comments—is useful for so much more than just following celebrities and posting what you ate for lunch. Researchers have now used Twitter to study the moods of individuals from various cultures around the world, and they identify consistent variations in people’s moods, depending on the time of day and season.

Based on 509 million Twitter posts (tweets) from 2.4 million users in 84 different countries over a two-year period, Scott Golder and Michael Macy say that people across the globe display similar rhythms to their moods despite very different cultures, geographies, and religions.

Their study appears in the 29 September issue of the journal Science.

According to the analysis, people tend to be more positive on weekends and early in the morning. In general, individuals awaken in a good mood that slowly deteriorates as the day progresses, which the researchers say is consistent with the effects of sleep and circadian rhythms.

 


Scott Golder of Cornell University discusses the research on global mood tracking via Twitter.

[Video © Science/AAAS]

Golder and Macy noted that early-morning good moods tend to be delayed for two hours on weekends, suggesting that people sleep later on those days.

“We saw that the shape of the rhythm [of people’s moods] is exactly the same on Saturday and Sunday, when many people are not at work,” said Golder. “So, clearly something else is playing a role here, whether it’s sleep or biological rhythms.”

The two researchers even confirmed these patterns with tweets from the United Arab Emirates, where people work Sunday through Thursday—and sleep in on Friday and Saturday.

Golder and Macy relied on a popular text-analysis program, known as the Linguistic Inquiry and Word Count, to analyze the hundreds of millions of tweets and gauge the positive affect—emotions like “enthusiasm,” “delight,” “activeness,” and “alertness”—or negative affect—things like “distress,” “fear,” “anger,” “guilt,” and “disgust”—expressed in the text.

“Positive and negative often sound like they’re opposite ends of a single continuum… but it turns out psychologists do not view affect, or mood, in that way,” said Golder. “Instead, there are positive kinds of moods—things like delight or activeness or alertness—that are considered positive. And on the other hand, there are moods that are considered negative affect, which include things like disgust or fear or anxiety.

“You might be full of delight but also full of anxiety, and so you may have high positive and negative affect at once,” Golder continued. “So, we really can’t look at them as being opposite ends of a single spectrum.”

The researchers say that this new study differs from other previous studies on mood in that the text they collected was directly obtained from comments posted by individuals in real time and not prompted by an experimenter or recalled from memory.

“You could always survey really large numbers of people or you could study small numbers of people really closely,” Golder said. “But as we’ve done in this study, it’s now becoming possible to passively observe millions of people acting in their native environments for a long time without disturbing them or inducing what is known as ‘experimenter demand effects.’”

Links

Read the abstract, “Diurnal and Seasonal Mood Vary with Work, Sleep, and Daylength Across Diverse Cultures,” by Scott Golder and Michael Macy.