According to Tim Berners Lee, the of the World Wide Web was to make communication easier, to bring knowledge to everyone, and to strengthen democracy and connections. It seems that it is dividing us into smaller and more angry splinter groupings. Why?
Online echo rooms – digital spaces with people with similar beliefs – or filter bubbles – the idea that algorithms show us content which we will likely agree with – are often blamed.
These concepts were both challenged by several studies. One of us, Dana, led a 2022 study that tracked the social media behaviors of ten respondents. The results showed people engage with content they don’t agree with. They even go so far to actively seek it out.
Social media platforms make money when users engage with an offensive post, whether it is “rage bait” (or something else) or something that makes them angry. On a larger scale, however, it can have antisocial consequences.
This is a form of “affective” polarisation, where we are attracted to people with similar views and dislike those who do not share them. Research, and Global Surveys show that this type of polarisation has been growing around the world.
Changes in the economics of platforms for social media would reduce online polarisation. This won’t happen without the government and us.
Our views are reinforced online
The use of social media has been linked to a growing polarisation of affect.
We can be influenced online by opinions we disagree or agree with, even on topics that we were previously neutral about. If you respect an influencer and they share their opinion on a law that you haven’t given much thought to, you are more likely adopt the viewpoint of this person.
This is known as “partisan-sorting”.
The research shows that our interactions on social media may lead us to develop new views about a subject. This shows that any search we do for more information can help solidify our emerging views. The results will likely contain the same language in the original post.
If you read a post in which incorrectly states that paracetamol taken during pregnancy can cause autism, you may find more posts with the keywords “paracetamol autism pregnancy”.
A high emotional state is associated with a greater susceptibility to believe false or “fake content.
Why do we get polarising content to read?
The economics of internet are at play here. Divisive, emotional posts will be more likely to get engagement, such as shares, likes and comments, especially from those who strongly agree with or disagree with the post, and also from provocateurs. The platforms will then display these posts to even more people and the engagement cycle continues.
According to a report from the Washington Post, Facebook’s ranking algorithm once treated emoji reactions (including anger) as five times more valuable than “likes”. This leads to them making more money through advertising. According to a report by the Washington Post from 2021, Facebook’s algorithm used to value emojis (including anger) five times as much as “likes”.
Simulations-based research has also shown how anger and division are driving online engagement. One simulation used bots (in a paper that has not yet been peer reviewed) to show how platforms measuring their success and income based on engagement (currently, all of them) will be most successful if they boost divisive posts.
What is our future?
The current state of social networks does not necessarily represent its future.
The average person spends less time on social networks than in the past. According to a report by the Financial Times published in recent months, social media time peaked in 2022. Since then it has been decreasing. Users aged 16 and over spent 10% less on social media platforms by the end of 2024 than they did in 2012.
Many users also leave the “mainstream” platforms in favor of smaller ones that reflect their political views, like the leftist BlueSky or the right-wing Truth Social. This may not be a good thing for polarisation but it does show that many people aren’t satisfied with social media as it is.
Internet polarisation also results in real costs for the government in terms of mental health care and police expenditure. In Australia, misinformation and online hate have been a factor in neo Nazi marches and the cancellation of LGBTQIA+ events due to threats.
We can all work together to change the status-quo. Researchers have found that online users who are more tolerant of other views can reduce polarisation. By not sharing or promoting content likely to enrage others, we can give social media companies fewer signals from which to work.
This is fundamentally a structural issue. It will require a re-frame of the economics behind online activity in order to encourage more respectful and balanced conversations and reduce the rewards for creating and/or engaging in rage bait. This will almost certainly need government intervention.
The government has taxed and regulated companies that have produced harmful products. The taxation and regulation of social media platforms is also possible. It’s not impossible, even if it is difficult. It’s worth it if you want to live in a world without outcasts.
Dana McKay received funding from Australian Research Council, Australian Digital Health Agency and Google.
George Buchanan has not disclosed any relevant affiliations other than their academic appointment. He does not work, consult, own or receive funding from companies or organisations that would benefit from the article.
