Dislike, unsubscribe: the origins of aggression on the Internet

2024 Author: Seth Attwood | [email protected]. Last modified: 2023-12-16 15:55

The desire for cooperation and the inner assumption that there are good people around us helped humanity to survive in the harsh times of the struggle for a place in the sun. Now it is quite possible to survive, remaining an ardent individualist, therefore, not only the desire to help, but also simply a benevolent attitude towards each other goes into the background. And especially on the internet and social networks.

We are publishing an abbreviated and adapted translation of an article that analyzes the origins of aggression on the Internet, to which everyone can be exposed. Both as a victim and as an immediate source.

The constant barrage of online abuse, including death threats and sexual assault, silences people, pushes them out of online platforms, and further diminishes the diversity of online voices and opinions. And there is no reason to believe that this situation is changing somehow. A survey last year found that 40% of adults have personally experienced online violence, with nearly half of them experiencing violent forms of harassment, including physical threats and harassment.

The business models of platforms like YouTube and Facebook are promoting content that is more likely to resonate with other users, because more engagement means better advertising opportunities. But the consequence of this approach is a preference for controversial and highly emotional content, which in turn can generate online groups of people that reflect and reinforce each other's opinions, fostering more extreme content and providing a niche for fake news to emerge.

Our human ability to communicate ideas through human networks has enabled us to build the modern world. The Internet offers unprecedented prospects for collaboration and communication between all members of humanity. But instead of taking advantage of the massive expansion of our social circles on the Internet, we seem to be returning to tribalism and conflict, and the belief in the potential of the Internet as a means of networking humanity for cooperation is beginning to seem naive.

While we usually communicate with strangers in a polite and respectful manner in real life, we can behave horribly online. Can we re-examine the collaborative techniques that once enabled us to find common ground and flourish as a species?

"Don't think too much, just press the button!"

I choose the amount and quickly move on to the next question, knowing that we are playing against the clock. My teammates are far away and unknown to me. I have no idea if we are striving for the common good or if I am being deceived, but I continue, knowing that others depend on me. I participate in the so-called "public goods game" at the Yale University Human Collaboration Laboratory. Researchers use it as a tool to help understand how and why we collaborate and whether we can improve our prosocial behavior.

Over the years, scholars have proposed various theories about why people interact so well with each other that they form strong societies. Most researchers now believe that the evolutionary roots of our shared grace can be found in the individual survival benefit that humans experience when we work together as a group. I came to New Haven to visit a group of laboratories where researchers are conducting experiments to further explore our extraordinary tendency to be kind to others, even at our own expense.

The game I'm playing is one of the lab's ongoing experiments. I am on a team of four, each in a different location and given the same amount of money to play. We are asked to choose how much money we will deposit into a common bank, given that this bank will then be doubled and divided equally among all of us. This social dilemma, like any collaboration, is based on a certain level of confidence that the other people in the group will be good. If everyone in the group contributes all their money, the entire amount is doubled, redistributed to four, and everyone gets twice as much. A win-win!

Nicholas Christakis, director of the Yale Human Nature Lab, thinks a lot about the design of our social media interactions. His team is exploring how our position on the social network affects our behavior, and even how some influential people can dramatically change the culture of the entire network.

The team is exploring ways to identify these people and include them in public health programs that can benefit the community.

Online, they can turn a culture of bullying into a culture of support.

Corporations are already using a crude Instagram influencer detection system to advertise their brands through them. But Christakis looks not only at how popular a person is, but also at his position in the network and the shape of that network.

Much of the antisocial behavior on the Internet stems from the anonymity of interactions - here the reputation costs associated with bad behavior are much lower than offline. One way to reduce the reputational costs of bad online behavior is to apply some form of social punishment.

One gaming company, League of Legends, did just that by introducing a Tribunal feature in which players could punish each other for negative play. The company reported that 280,000 players were "re-educated" in one year, which means that after being punished by the tribunal, they changed their behavior and then achieved a positive reputation in the community. Developers can also embed social rewards for good behavior by encouraging more collaborative elements that help build relationships.

Researchers are already beginning to learn to predict when a situation is about to go bad - the point at which it can benefit from proactive intervention. “You might think that the sociopaths we call trolls on the net are a minority, who do all this harm,” says Cristian Danescu-Niculescu-Mizil of the Information Science Department at Cornell University. “But in our work, we actually find that ordinary people like you and me can behave antisocial. For a certain period of time, you too can become a troll. And this is amazing."

It's also alarming. At the end of the day, it may be tempting to offend someone distant you don't know if you think it will impress your close social group. Danescu-Niculescu-Mizil studies the comment sections under the online articles. He identifies two main triggers for trolling: the context of the exchange, that is, the behavior of other users, and your mood. “If you’ve had a bad day, you’re much more likely to start trolling in the same situation,” he says.

After collecting data, including from people who were trolling in the past, Danescu-Niculescu-Mizil built an algorithm that predicts with 80% accuracy when someone is about to start behaving abusively on the Internet. And this makes it possible, for example, to introduce a delay in the publication time. If people have to think twice before they write something, it will improve the context of the exchange for everyone at once: you are less likely to witness other people behaving badly, and therefore less likely to misbehave yourself.

The good news is that, despite the horrific behavior that many of us have encountered online, most of the time we communicate in a pleasant and cooperative manner. Even more, well-founded moral outrage is useful in challenging hateful tweets. A recent UK study of anti-Semitism on Twitter found that messages challenging anti-Semitic tweets spread far more than the anti-Semitic tweets themselves.

As Danescu-Niculescu-Mizil notes, we have had thousands of years to hone our personal contact mechanisms, but only 20 years for social media.

As our online behavior evolves, we may well begin to introduce subtle cues, digital equivalents of facial expressions and other bodily cues, to facilitate online discussions. In the meantime, we advise you to deal with insults on the Internet, staying calm - this is not your fault.

Don't retaliate, but block and ignore the bullies or, if you feel it's right, tell them to stop. Talk to family or friends about what's going on and ask them to help you. Finally, take screenshots and report problems to the social network tech support, and if they involve physical threats, report it to the police.

If social media as we know it survives, companies operating on these platforms will have to continue to manage their algorithms, perhaps relying on behavioral sciences to encourage collaboration rather than separation, positive online experiences rather than abuse. But as users, we too can learn to adapt to this new communication environment so that productive interaction remains the norm, both online and offline.