What political uncertainty, algorithmic amplification, and nonstop news exposure do to our attention and emotional regulation.
The phone is in your hand before your feet touch the floor. It is the morning of November 4, and the screen has been busy overnight. One headline declares a winner. Another asks what happens next in a race nobody has called. Notifications stack up like unopened mail. One friend is posting confetti; another has written a paragraph that begins, "I genuinely cannot believe." A commentator is already explaining the consequences of decisions that have not yet been made.
You lock the screen, decide to make coffee, and unlock it again before the kettle is warm.
Wanting to know what happened is reasonable. But somewhere between the first check and the fifth, the purpose of checking quietly changes. The first look informs. The later ones are often less about learning than about quieting an uncertainty that refuses to settle.
The election may have an official ending. The information cycle does not.
Part 1 of this series examined how economic uncertainty occupies mental bandwidth before any loss arrives. This essay follows that thread into political uncertainty and an information environment that almost never goes quiet.
The Brain on an Unresolved Question
In a 2013 review in Nature Reviews Neuroscience, Dan Grupe and Jack Nitschke described anxiety as the anticipatory changes in feeling, thinking, and behavior that arise when a possible future threat is uncertain. Uncertainty is costly, they argued, because it undermines preparation.
Their Uncertainty and Anticipation Model of Anxiety identifies five processes that, when exaggerated, characterize clinical anxiety. Two are especially relevant here: heightened vigilance toward threat, and heightened reactivity to uncertainty itself, where not knowing becomes distressing in its own right.
The model describes clinical anxiety, and the authors stress that the same circuitry serves adaptive purposes. A voter refreshing results is not displaying a disorder, but the vocabulary is useful. A ballot count is exactly the kind of unresolved, consequential, largely uncontrollable question that anticipatory vigilance is built to monitor. Scanning feels like doing something. Often, it is the only thing available to do.
When the Coverage Becomes the Stressor
The most cited evidence that media exposure carries a psychological cost comes from crisis research. After the 2013 Boston Marathon bombings, E. Alison Holman, Dana Rose Garfin, and Roxane Cohen Silver surveyed a representative national sample of 4,675 U.S. adults two to four weeks after the attack. Adjusting for pre-bombing mental health and prior trauma exposure, they found that people reporting six or more daily hours of bombing-related media in the first week reported higher acute stress than people who had been at or near the event.
The study was cross-sectional and cannot establish that media caused the stress. And an election is not a terrorist attack; the finding offers a cautious signal about repeated exposure, not an equivalence.
There is also a complication. In a three-year longitudinal study (2019), Rebecca Thompson and colleagues found that the relationship runs in both directions: distress after one collective trauma predicted heavier media consumption around a later one, which in turn predicted more distress. Anxious people seek more news; more news can sustain anxiety.
Political news has been studied more directly. Brett Ford, Matthew Feinberg, and colleagues (2023) combined daily diary studies with experiments using political news clips. Daily politics reliably evoked negative emotions, which tracked with worse psychological and physical well-being, yet also predicted greater motivation to volunteer, donate, or protest. Participants who dampened those emotions through distraction or reappraisal felt better and were less motivated to act.
Political distress is not simply a malfunction. It is partly the fuel of participation.
The Feed Meets the Anxious Brain
Imagine two posts from accounts you follow: a plain update on certified results, and an emotionally charged prediction about what happens next. A chronological feed shows them in the order they were posted. A ranked feed weighs many signals, often including predicted engagement: how likely you are to click, reply, share, or linger. If the model expects the charged post to hold your attention, it can rise above the plain one, even if it arrived later. Objectives differ across platforms and change over time, so no single description fits every system.
Does emotional content travel further? Analyzing more than 560,000 tweets on polarizing issues, William Brady and colleagues reported in 2017 that each additional moral-emotional word was associated with roughly a 20 percent increase in retweets, mostly within like-minded networks. The size of that effect has been debated (Burton et al., 2021), and a 2025 preregistered meta-analysis by Brady's group, pooling 27 studies, put it closer to 12 percent per word. But retweets are human choices. More sharing is not proof that an algorithm amplified anything.
Direct tests of ranking are more mixed. A 2025 audit by Smitha Milli and colleagues found that, compared with a chronological feed of the same accounts, Twitter's engagement-based algorithm amplified emotionally charged, out-group hostile political content, the kind users said made them feel worse about the other side. Yet when Andrew Guess and colleagues, working with Meta, moved consenting Facebook and Instagram users to chronological feeds for three months during the 2020 election, time on the platforms fell substantially while polarization, political knowledge, and other key attitudes did not measurably change. In these studies, ranking changed what people saw and how long they stayed more clearly than it changed what they believed.
Emotions themselves can spread, modestly. In the 2014 Facebook "emotional contagion" study, Adam Kramer, Jamie Guillory, and Jeffrey Hancock reduced positive or negative posts in 689,003 users' feeds for a week; those shown fewer positive posts wrote slightly fewer positive words, and vice versa. The effects were tiny, the measure was word choice rather than felt emotion, and the journal issued an Editorial Expression of Concern about consent. My interpretation, not a finding, is that a feed's emotional texture may make anger look more universal than it is.
So the answer is conditional. When a system rewards engagement and a person is searching for certainty, the two can plausibly feed each other: uncertainty drives checking, checking produces engagement signals, and those signals can favor whatever is most arousing. The evidence supports pieces of that loop, not a complete causal chain.
The Developer's Paradox
Many DEV.to readers sit on both sides of this screen: we build notification pipelines and ranking services, then pick up phones running them.
We have become extraordinarily sophisticated at engineering information delivery without becoming equally sophisticated at protecting the attention that receives it.
This is not a story of villains. Engagement is often a reasonable proxy for relevance. The tension is that the easiest metrics to measure are not always what users would choose on reflection, the very gap Milli's team documented.
A few evidence-informed starting points:
Batch rather than drip. In a two-week field experiment with 237 smartphone users in India, Nicholas Fitz and colleagues found that delivering notifications three times a day lowered self-reported stress and inattention and lifted mood, though it also raised fear of missing out. Switching notifications off entirely raised anxiety without improving attention.
Rank by stated preferences, not only behavior. In Milli's analysis, ranking by what users said they valued reduced angry and hostile content, though it risked reinforcing existing views.
Make ranking legible and optional. A plain-language "why am I seeing this" and an easy chronological mode give people control. Their mental-health effects are unestablished; frame them as respect for agency, not therapy.
The Election Ends. The Uncertainty Changes Shape.
Back to November 4. Even when a race is called, its question gives way to new ones. Which policies will change? What happens to prices and jobs? How much of the commentary is evidence, and how much is forecasting dressed as fact?
Vigilance tries to resolve uncertainty by gathering information. But here lies the quiet mismatch at the center of the morning:
Much post-election uncertainty is not informational. It is temporal.
The answers depend on decisions not yet made and events not yet arrived. They will come on the schedule of legislatures, courts, and economies, not push alerts. Checking again cannot make them arrive sooner; it can only make the waiting louder.
None of this argues for looking away; Ford's research ties political discomfort to the willingness to act. Nor is any party or group of voters uniquely susceptible. These processes are human, not partisan. Responsibility is shared. Individuals choose when to check. Newsrooms choose how often to alert and how clearly to separate reporting from speculation. Platforms choose what their systems optimize for. Only the last two operate at scale.
The distinction worth protecting is between attention that serves understanding and attention that has become reflex. One informs a vote, a conversation, a decision. The other mostly keeps the question open.
Democracy asks citizens to pay attention. The attention economy asks them never to stop. Between those two demands lies a design problem, and many people reading this are equipped to work on it.
So the question for you: if you could change one default in the products you build, so that a person searching for certainty would be better served rather than merely more engaged, what would it be?
This is Part 2 of America Under Pressure: The Psychology of Uncertainty. Part 1 examined economic insecurity and anticipatory stress. Part 3 turns to the financial and emotional pressures of the holiday season.
References
Grupe, D. W., & Nitschke, J. B. (2013). Uncertainty and anticipation in anxiety: An integrated neurobiological and psychological perspective. Nature Reviews Neuroscience, 14(7), 488–501. https://doi.org/10.1038/nrn3524
Holman, E. A., Garfin, D. R., & Silver, R. C. (2014). Media's role in broadcasting acute stress following the Boston Marathon bombings. Proceedings of the National Academy of Sciences, 111(1), 93–98. https://doi.org/10.1073/pnas.1316265110
Thompson, R. R., Jones, N. M., Holman, E. A., & Silver, R. C. (2019). Media exposure to mass violence events can fuel a cycle of distress. Science Advances, 5(4), eaav3502. https://doi.org/10.1126/sciadv.aav3502
Ford, B. Q., Feinberg, M., Lassetter, B., Thai, S., & Gatchpazian, A. (2023). The political is personal: The costs of daily politics. Journal of Personality and Social Psychology, 125(1), 1–28. https://doi.org/10.1037/pspa0000335
Brady, W. J., Wills, J. A., Jost, J. T., Tucker, J. A., & Van Bavel, J. J. (2017). Emotion shapes the diffusion of moralized content in social networks. Proceedings of the National Academy of Sciences, 114(28), 7313–7318. https://doi.org/10.1073/pnas.1618923114
Burton, J. W., Cruz, N., & Hahn, U. (2021). Reconsidering evidence of moral contagion in online social networks. Nature Human Behaviour, 5(12), 1629–1635. https://doi.org/10.1038/s41562-021-01133-5
Brady, W. J., Rathje, S., Globig, L., & Van Bavel, J. J. (2025). Estimating the effect size of moral contagion in online networks: A pre-registered replication and meta-analysis. PNAS Nexus. https://doi.org/10.1093/pnasnexus/pgaf327
Milli, S., Carroll, M., Wang, Y., Pandey, S., Zhao, S., & Dragan, A. D. (2025). Engagement, user satisfaction, and the amplification of divisive content on social media. PNAS Nexus, 4(3), pgaf062. https://doi.org/10.1093/pnasnexus/pgaf062
Guess, A. M., Malhotra, N., Pan, J., Barberá, P., Allcott, H., et al. (2023). How do social media feed algorithms affect attitudes and behavior in an election campaign? Science, 381(6656), 398–404. https://doi.org/10.1126/science.abp9364
Kramer, A. D. I., Guillory, J. E., & Hancock, J. T. (2014). Experimental evidence of massive-scale emotional contagion through social networks. Proceedings of the National Academy of Sciences, 111(24), 8788–8790. https://doi.org/10.1073/pnas.1320040111
Fitz, N., Kushlev, K., Jagannathan, R., Lewis, T., Paliwal, D., & Ariely, D. (2019). Batching smartphone notifications can improve well-being. Computers in Human Behavior, 101, 84–94. https://doi.org/10.1016/j.chb.2019.07.016
Top comments (2)
The observation that "much post-election uncertainty is not informational, it is temporal" hits at the core of why modern news feeds feel so exhausting. We’re essentially applying rapid-response, high-frequency checking behaviors to slow-moving systemic shifts, creating a constant state of anticipatory vigilance that algorithmic feeds are only too happy to capture and monetize.
From an engineering and UX perspective, the disconnect between engagement optimization and user satisfaction (as Milli et al. highlighted) is one of the biggest design challenges we face today. When platforms treat ambient anxiety as a retention feature rather than a side effect, the default UX inevitably tilts toward urgency over clarity.
If I could change one default in notifications and feed architectures, it would be introducing threshold-based batching for non-time-critical updates. Instead of pushing every incremental commentary or speculative opinion as a real-time event, system defaults should favor digest views for ongoing news threads, saving immediate pushes exclusively for verified, structural state changes. Shifting the default from "stream everything instantly" to "summarize asynchronously unless critical" would respect human attention without cutting off information access. Looking forward to Part 3 on the holiday season overlay!
This is a fascinating look at the intersection of psychology, technology, and democracy. The distinction between seeking information and seeking relief from uncertainty is particularly powerful. I also appreciate the nuance that political distress isn't necessarily unhealthy; sometimes it's what motivates people to participate and take action. For developers, the challenge is recognizing that keeping users engaged isn't always the same as serving their interests. Perhaps the most important design principle is this: technology should help people become better informed, not make them feel that they must remain permanently on alert. Attention is a finite human resource, and protecting it should be part of responsible product design.