I've seen a lot of discussion of the 'filter bubble', particularly as a way for DuckDuckGo to differentiate itself from other search engines. But I have never seen any scientific data indicating it actually exists. I've seen the TED talk, and I've seen the oft-repeated example involving BP, but I've never seen anything more than a couple of examples about how it might exist.

Has anyone demonstrated in any scientific fashion whatsoever that there is a population of people that are insulated from news they don't want to hear because it is actively being kept from them on an ongoing basis by supposedly unbiased news sources?

Here's the data I do know about: we already have 'news' channels that pander to particular audience, and we've already seen that watching news you agree with activates the same pleasure centers in the brain that are stimulated by watching pornography.

Based on evidence I've seen, the problem is with the audience, not the news. The assertion in the article that the filter "can polarise populations creating potentially harmful divisions in society" seems, well, unfounded, at least so far.