What is negativity bias?
What a large set of headline tests, a 17-country experiment and two reviews of development say about why bad news pulls harder, and for whom.
Answer to the question
What is negativity bias?
Negativity bias is the tendency for negative information to carry more weight than positive information of similar strength, in attention, learning and judgement. In a large set of randomised headline tests, each added negative word raised clicks by about 2.3 per cent. Reactions to news vary between people, and the pattern changes with age.
Key takeaways
- In a large set of randomised headline tests, each added negative word raised the click-through rate by about 2.3 per cent (Robertson and colleagues, 2023).
- A 17-country experiment found negativity biases in bodily reactions to real news video, with considerable differences between individuals and some people responding more to positive stories (Soroka and colleagues, 2019).
- A 2008 review argued that the bias is present from early in development, with infant social referencing as the clearest evidence (Vaish and colleagues).
- A 2019 narrative review found a positivity advantage in children for words and faces that fades with age, and described the overall question as still debated (Kauschke and colleagues).
- Johann Hari's Stolen Focus treats the bias as one reason news and feeds hold attention; the headline experiment measured clicks, not belief or satisfaction.
Each negative word added to a news headline raised its click rate by about 2.3 per cent in a large set of randomised headline tests. The data were real choices, roughly 5.7 million clicks on about 105,000 headline variations compared across 22,743 tests, which is behaviour at scale. That gap between how bad and good information land is what researchers call negativity bias: negative information tends to carry more weight than positive information of similar strength, in attention, learning and judgement. It is related to loss aversion, which concerns how people weigh losing against gaining, but negativity bias is wider. It covers what people notice, what they remember and what they rely on when they form a view.
How strong is negativity bias in news headlines?
The headline experiment matters because it separates wording from topic. Each variation was shown to different readers at random, so a difference in clicks can be traced to the words and not to who happened to see the story. Claire Robertson and five co-authors reported that headlines with more negative words earned higher click-through rates, and the figure of 2.3 per cent applies to each added negative word. The result describes what people click. It does not say what they believe, or what they would prefer to read if asked, and whether the extra clicks came with trust or satisfaction was not measured.
When does negativity bias appear, and when does it fade?
A 2008 review by Amrisha Vaish, Tobias Grossmann and Amanda Woodward asked where the bias comes from. Adults, they note, process positive and negative information asymmetrically across many tasks: they attend to, learn from and rely on negative information more. The review argues that the bias is present from early in development, and that the clearest evidence comes from infant social referencing, which means checking a caregiver's reaction before responding to something unfamiliar.
The authors discuss evolutionary roles the bias may serve, along with its consequences, and they set out testable predictions for work with infants and older children. The review is an argument assembled from many studies, and its main contribution is a developmental account that later researchers could test. That framing matters for how the bias is read. If negativity bias appears in the first year of life, it is unlikely to be only a product of newsrooms, advertising or social platforms, which means the headline result is better understood as a modern place where an old asymmetry shows up than as its cause.
The picture is not uniform across ages and materials. A 2019 narrative review led by Christina Kauschke brought together three overviews of how people process emotional words and faces. Studies of words showed no consistent advantage for either positive or negative items. Studies of faces varied with the experimental method. Children showed a clear positivity advantage for both words and faces, which faded with age and in some cases reversed into a negativity bias. The authors describe the question of which kind of emotion is processed more efficiently overall as still debated.
The later review narrows what the earlier one claims. Shoppers meet the same question when one poor rating sits beside many good ones, and Why Does One Bad Review Hurt So Much works through that case with sales data. Negativity bias is a strong pattern in many tasks, and age and materials change its size and sometimes its direction.
| Setting | What was measured | What the paper reported |
|---|---|---|
| Online news headlines | Clicks on randomised headline variations | About 2.3 per cent more clicks per added negative word |
| News video in 17 countries | Bodily reactions to real news clips | A negativity bias in reactions across countries, with wide differences between people |
| Early development | Review of social referencing and related work | A bias present from early in development |
| Words and faces across ages | Narrative review of three overviews | A positivity advantage in children that fades with age; mixed results for words and faces in adults |
Do our bodies really react more strongly to bad news?
The three kinds of evidence differ in what they can show. Clicks record a choice, bodily reactions record a response that people may not control or report, and developmental reviews trace when the pattern starts. Each answers a different question about negativity bias, and none of them settles how large it is in any single person on any single day. That is why the careful claim is about averages, ages and settings, and why the spread between individuals deserves as much attention as the central tendency.
A 17-country experiment measured how people's bodies responded to real news video, not what they said about it afterwards. Stuart Soroka, Patrick Fournier and Lilach Nir reported negativity biases in these psychophysiological reactions across the countries, a result they connect to the preference of news audiences around the world for mostly negative coverage.
The same paper found considerable variation between individuals, and some people responded more to positive stories. The bias describes an average tendency with a spread around it, and the spread is part of the finding.
The spread matters for the wider argument about attention. Johann Hari argues in Stolen Focus that a calm item has to compete with an alarming one for the same few seconds, which is the contest traced in What is the attention economy and lived out in Why You Cannot Stop Scrolling. An average pull towards bad news is the condition that argument needs, and readers who lean the other way are the ones it fits least. In a room of readers the pull towards bad news is real on average, and some lean the other way.
Common questions
What is negativity bias?
Negativity bias is the tendency for negative information to have more impact than positive information of similar strength. Researchers have documented it in attention, learning and judgement, and a large headline test found about 2.3 per cent more clicks per added negative word.
Is negativity bias the same as loss aversion?
No. Loss aversion concerns how people weigh losing against gaining. Negativity bias is wider and covers what people notice, remember and rely on.
Does everyone show negativity bias?
Not to the same degree. A 17-country experiment found negativity biases in bodily reactions to news on average, with considerable differences between individuals, and a review of words and faces found a positivity advantage in children that fades with age.
Does negativity bias explain why bad news dominates feeds?
It is one candidate explanation. Headline tests show that negative wording can raise clicks, and Johann Hari argues in Stolen Focus that the bias helps news and feeds hold attention.
Sources
Hari, J. Stolen Focus. Seed book; vault hub Books/Stolen Focus.
Robertson, C. E., Pröllochs, N., Schwarzenegger, K., Pärnamets, P., Van Bavel, J. J., & Feuerriegel, S. (2023). Negativity drives online news consumption. Nature Human Behaviour, 7(5), 812-822. doi:10.1038/s41562-023-01538-4 https://doi.org/10.1038/s41562-023-01538-4
Soroka, S., Fournier, P., & Nir, L. (2019). Cross-national evidence of a negativity bias in psychophysiological reactions to news. Proceedings of the National Academy of Sciences, 116(38), 18888-18892. doi:10.1073/pnas.1908369116 https://doi.org/10.1073/pnas.1908369116
Vaish, A., Grossmann, T., & Woodward, A. (2008). Not all emotions are created equal: The negativity bias in social-emotional development. Psychological Bulletin, 134(3), 383-403. doi:10.1037/0033-2909.134.3.383 https://doi.org/10.1037/0033-2909.134.3.383
Kauschke, C., Bahn, D., Vesker, M., & Schwarzer, G. (2019). The role of emotional valence for the processing of facial and verbal stimuli: Positivity or negativity bias? Frontiers in Psychology, 10, 1654. doi:10.3389/fpsyg.2019.01654 https://doi.org/10.3389/fpsyg.2019.01654