How Living in a “Filter Bubble” Fuels Political Polarisation
If you are on social media, then chances are that you live in a “filter bubble” — your own unique information universe, personally tailored for you by filtering algorithms.
These algorithms constantly track what you — and many others like you — do online. Based on this, they decide what you see, for example, in your Facebook and Twitter feeds. But perhaps more importantly, they decide what you don’t see, and you can never really know what stays hidden from your sight.
In other words, filtering algorithms put you in a “bubble” where everything backs up your beliefs — and isolate you from any opposing views. However, something to keep in mind is that there are other people who live in their own individual “bubbles” which look nothing like yours.
However, the stakes get higher once we move from cute animal videos to politics. Algorithms push antagonistic, one-sided narratives into your social media feeds, search results and recommendations — the same is happening to the people on the other side of the ideological barricades.
Generally, people tend to favour information that supports their beliefs or biases and reject information that doesn’t fit into their belief system — this is known as “confirmation bias” [1]. So when we are only exposed to the information that we believe is true, it deepens our — possibly opposing — beliefs and increases the polarisation of opinions.
In the USA, where Facebook is now one of the key electoral battlegrounds, 60% of its users are unaware of filtering algorithms which present them with only one point of view [2]. This affects the voting public and might be influencing the results of elections. This is a big reason why it is necessary that we — the users — acknowledge that “filter bubbles” exist and get a better understanding of their nature. It is equally important that “filter bubbles” are further investigated by scientists.
But people being trapped in “filter bubbles” is not the only cause of the segregation of the internet into different political camps. The global internet is split into a number of nationally-administered internets aligned along physical borders between countries and regions — this is described by the term “cyberbalkanization” [3]. Yet, unlike filtering algorithms, cyberbalkanization only increases the polarisation of opinions on opposite sides of a border — but not within one country or region. Another big difference between these two phenomena is that cyberbalkanization is created deliberately via censorship of internet traffic — “filter bubbles” typically occur unintentionally due to our own actions. They are not the same thing and should be distinguished from each other.
It is debated whether the effects of the segregation of the internet on political polarisation are significant or not. There are studies that suggest that the segregation of the internet is actually lower than the segregation of traditional offline media. The same goes for interactions that we have outside of social media — in families, schools, and other communities [4]. Therefore, we might want to treat “filter bubbles” in the same way we would treat the segregation of information sources in the world of newspapers, magazines and television.
The New York Times — Here’s the Conversation We Really Need to Have About Bias at Google
Farnam Street — How Filter Bubbles Distort Reality: Everything You Need to Know
The Verge — The case that Russia is winning the cyberwar
Wired — Your Filter Bubble is Destroying Democracy
Greater Good Magazine — Are Social Media Driving Political Polarization?
