The power of pictures: Visual bias in the news

A- A A+

Imagery within media stories has a powerful effect on the way a reader thinks about a particular topic. This column highlights the strength of the effect, presenting evidence from the United States. Readers’ views on a range of topics – from police budgets to pandemic management – are highly sensitive to the pictures that accompany news stories. People also tend to react more extremely to images that directly confirm or challenge their pre-existing opinions, shining new light on the risk of political polarisation.


A picture paints a thousand words. But they can also influence the way a reader interprets a story in the media. New research shows that the images used to illustrate written news pieces can have a powerful influence on the way in which a reader understands the contents of a story. Media outlets exploit this phenomenon to sway their audience towards a particular agenda. 

Newsmakers, the study shows, choose leading pictures whose content reflects their own political preferences. As a result, liberal and conservative newsmakers adopt systematically different visual languages. This visual bias of news coverage, often less apparent than biases expressed in text, affects the way readers form opinions on issues covered in the news. This may reinforce the polarisation of public opinion on a range of political issues. Through both words and pictures, people are being fed very different perspectives, deepening divides.

The media landscape in the United States provides a useful case study. To conduct this latest piece of work, 300,000 leading images from news published in 2020 by the main US news outlets are analysed. By feeding the pictures through a series of vision algorithms to extract information on their content (such as the subjects and objects depicted, their pictorial attributes, or contextual aspects of each photo), it is possible to categorise the different pictures. This information is then gathered to form a ‘visual vocabulary’, and used to study the overall visual language of the different news sources. This process shows that liberal and conservative news outlets adopt very different visual jargon. Their visual language also appears politically divided, particularly in the coverage of politics, security or health topics. 

To investigate the effects of this partisan visual language on readers’ opinions, the research also includes results from an experiment. In the test, a nationally representative sample of 2,000 US residents are shown news pieces on different topics and asked to express their opinion on each topic (as in a regular opinion survey). All the written content (texts, headlines, bylines, sources, dates, etc.) is left identical for every participant. Instead, the leading picture accompanying each article varies randomly, among three alternatives: its visual language could resemble that of a Republican-leaning news source, a Democrat-leaning news source, or a politically neutral source. This means that the textual information associated with the story is held constant, but the imagery varies based on political categories.

The results are striking. The experiment reveals that by changing the pictures in the story, it is possible to influence respondents’ opinions on topics such as politics, the economy, security and public health. In fact, readers who saw stories illustrated by a Republican-leaning picture on average expressed a significantly more conservative opinion than readers whose news contained a Democrat-leaning one, despite reading the same written content. Even when the words in the articles are identical, a reader’s stance can be moulded through imagery.

These differences in opinion are measurable. For example, relative to the Republican-leaning readers, the Democrat-leaning readers preferred a much lower budget for police forces (with a $4.5 billion difference), they were more dissatisfied with the pandemic management during Trump’s administration, they were more confident about lower future inflation, and they were more certain about Biden’s chances to revive the US-Iran nuclear deal. These are all extremely complex topics, and it is therefore notable that a simple change in an article’s accompanying photo can affect a person’s view on such issues. 

There is also the factor of confirmation bias. By considering participants’ political leanings, the research shows that readers react more acutely to pictures aligned with their pre-existing political viewpoint. For example, using Republican affiliates who were exposed to neutral leading images as a reference point, Republicans who were shown Republican-leaning images would be more distant (i.e., have a more extreme opinion) than Republicans exposed to Democrat-leaning pictures. The same dynamic applied also to the Democrat participants in the study. This suggests that consuming media from across the political spectrum can have a polarising effect. 

So, images are incredibly influential. These results indicate that news-leading pictures are more powerful in reinforcing prior beliefs than in promoting a different viewpoint. Given this, the presence of visual bias in news coverage is likely to cause public opinion to be more divided and politically polarised. This effect will be further amplified if people receive their news exclusively from like-minded sources, as often occurs when content is recommended by a social media algorithm. 

Policy-makers should think carefully about how media outlets can shape public opinion. It is no secret that news organisations are closely involved in political processes around the world, and that regulation can only go far in controlling editorial policy. A free press is a vital cornerstone of democratic systems, but policy-makers must at least remain mindful of the power that media outlets hold. Beyond the content of their headlines, articles and columns, the imagery these companies use can shape opinions, and change the world.