The second full day at the ECREA 2026 conference in Brno starts with a session on news avoidance, and a paper by Krisztina Burai that focusses on Facebook. She begins by saying that citizens have agency in navigating political information, but that this has changed over time: through the choice of newspapers they read; through the choice of broadcast TV channels they watched; through the greater choice of channels in the network age; and finally through the abundance of information sources they can follow in the social media age, and the news feed curation tools available to them.
Social media are now a very important source of news; in Hungary, for instance, Facebook is a key news source for 43% of the population, though this has also declined slightly in recent years. Here, users can curate their news feeds: they can follow and unfollow, mute, and hide content from news pages, politicians, and other sources, but their experience here is thus also also shaped by the selections made by those sources, as well as by platform algorithms, and the personal curation practices of their social ties.
One key question here is to what extent social media algorithms override user choices – and to what extent users can push back at this. Krisztina drew on a face-to-face survey of 1,000 Hungarians to examine this: it explored political interest, curation (both to add and remove political content) and avoidance practices, and exposure to political campaigns and opposition primary elections.
Users who actively engaged with political content also reported higher levels of political exposure on Facebook; this is relatively unsurprising and suggests that Facebook’s curation environment responds to these interests. But limiting and avoidance strategies seemed much less effective: such users still reported substantive political exposure, and in the campaign contexts even higher political exposure – this might be the result of political advertising breaking through such avoidance, or of the algorithm interpreting the removal of unwanted content from one side as an indication of political preferences.
All of this is based on self-reporting, of course, which may also skew results. These results can be tested further through data donations and digital trace data, though, and these might point more clearly to the consequences of avoidance structures on subsequent exposure to political content. Overall, though, avoidance may serve more as a coping strategy than as a meaningful intervention to curate and regulate one’s own feed.
