Skip to Content
Uncategorized

An Emerging Science of Clickbait

Researchers are teasing apart the complex set of links between the virality of a Web story and the emotions it generates.

In the world of Internet marketing and clickbait, the secret of virality is analogous to the elixir of life or the alchemy that turns lead into gold. It exists as a kind of Holy Grail that many search for and few, if any, find.

The key question is this: what is the difference between stories that become viral and those that don’t?

One idea is that the answer lies in the emotions stories generate for the people that read them. But what quality of emotion causes somebody to comment on a story or share it on social media?

Today, we can get an insight into this question thanks to the work of Marco Guerini at Trento Rise in Italy and Jacopo Staiano at Sorbonne Université in Paris. These guys have studied the data from two websites that allow readers to rate news stories according to the emotion each generates. That opens a fascinating window into the relationship between virality and emotion for the first time.

Psychologists have long categorized emotion using a three-dimensional scale known as the Valence-Arousal-Dominance model. The idea is that each emotion has a valence, whether positive or negative and a level of arousal, which is high for emotions such as anger and low for emotions like sadness.

Dominance is the level of control a person has over the emotion. At one end of this spectrum are overwhelming emotions like fear and at the other, emotions that people can choose to experience, such as feeling inspired.

Every emotion occupies a point in this Valence-Arousal-Dominance parameter space.

Guerini and Staiano’s idea is that it is not an emotion itself that determines virality but its position in this parameter space.

It turns out that two news-based websites have recently begun to collect data that throws light on exactly this problem. Rappler.com is a social news site that allows each user to rate the emotional value of each story using a “mood meter.” The Italian newspaper site Corriere.it offers a similar function.

Together, these sites have some 65,000 stories rated by emotional quality. That’s a significant database to explore the link between emotion and virality, which they measure by counting the number of comments each story generates as well as the number of votes it gets on social media sites such as Facebook and Google Plus.

Finally, they mine the data looking for patterns of emotion associated with the most viral content.

The results make for interesting reading. Guerini and Staiano argue there is a clear link between virality and particular configurations of valence, arousal and dominance. “These configurations indicate a clear connection with distinct phenomena underlying persuasive communication,” they say.

But there is a curious difference between the emotions that drive commenting behavior compared to voting behavior. Guerini and Staiano say that posts generate more comments when they are associated with emotions of high arousal, such as happiness and anger, and with emotions where people feel less in control, such as fear and sadness.

By contrast, posts generate more social votes when associated with emotions people feel more in control of, such as inspiration.

Curiously, the valence of an emotion does not influence virality at all. In other words, people are just as likely to comment or vote on a post regardless of whether it triggers a positive or negative emotion.

Of course, this is by no means a recipe for online success. But it should provide some food for thought for Internet marketers, bloggers and journalists alike.

Anybody who has spent some time trawling the Internet will have come across headlines designed to manipulate emotion in a pretty crude way. But that may only be the beginning.

Guerini and Staiano’s work provides some much more detailed insights into the fundamental emotional drivers of virality and, as such, could be thought of as laying the foundations for an emerging science of clickbait.

Ref: arxiv.org/abs/1503.04723 : Deep Feelings: A Massive Cross-Lingual Study on the Relation between Emotions and Virality

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.