How Feeds Reward Reaction Before Reflection

Introduction

Shareability incentives are the rewards built around a social post: likes, comments, reposts, stitches, quote posts, follower growth and algorithmic boosts. They matter because they change the question a user is implicitly answering. Instead of “Is this accurate?”, the feed often asks, “Will this get a reaction?” In the age of social media and AI, that distinction is central to critical thinking. A fluent AI-generated claim, a cropped clip or a dramatic screenshot can travel quickly when the platform’s visible rewards make speed, emotion and social approval feel more valuable than verification.

The strongest evidence does not suggest that people are simply indifferent to truth. It points to a mechanism: social platforms repeatedly reward content that attracts attention, and users learn from those rewards. Experiments show that shifting attention towards accuracy improves the quality of what people share, while studies of platform ranking suggest that engagement-based systems can amplify emotionally charged or divisive material. [Nature+2PNAS]nature.comShifting attention to accuracy can reduce misinformation…by G Pennycook · 2021 · Cited by 1605 — The results show that subtly sh…

Visible rewards teach people what “works”

A social feed does not just display information. It displays social judgement. Like counts, repost numbers, replies and creator metrics make popularity visible before evidence has been checked. That changes how a post is read: a claim with thousands of shares can feel socially validated even when the underlying evidence is weak.

Research on misinformation habits helps explain why this matters. A 2023 PNAS study by Gizem Ceylan, Ian Anderson and Wendy Wood argued that misinformation sharing is not only a problem of laziness or political bias; it can become habitual when users are repeatedly rewarded for sharing attention-grabbing content. The study also found that changing rewards so that they favoured accurate sharing could improve discernment, meaning users shared more true information relative to false information. [PNAS]pnas.orgSharing of misinformation is habitual, not just lazy or biasedby G Ceylan · 2023 · Cited by 224 — Habitual sharing of misinformation…

This is a subtle but important point. The reward does not have to be money. A like, a quote post, a reply from a prominent account, a repost by a friend or a spike in followers can all act as feedback. Over time, the user learns which styles of posting produce attention: sharper wording, faster reaction, stronger emotion, simpler villains, more confident claims and less visible uncertainty.

That reward structure can make even careful people behave less carefully. Pennycook and Rand’s research on “accuracy nudges” found that people often share misinformation because their attention is focused on factors other than accuracy; when prompted to think about accuracy, they become more discerning about what they share. In a field experiment on Twitter and several survey experiments, a small shift in attention improved the quality of subsequently shared news. [Nature]nature.comShifting attention to accuracy can reduce misinformation…by G Pennycook · 2021 · Cited by 1605 — The results show that subtly sh…

The critical-thinking lesson is practical: a post’s popularity is not the same as its reliability. Popularity may tell us that content is emotionally resonant, identity-relevant, funny, alarming or well-timed. It does not, by itself, tell us whether the claim has been tested against evidence.

Algorithmic amplification turns reactions into reach

Visible rewards are only half the mechanism. The other half is ranking. Social platforms do not show every post to every follower in chronological order. They use ranking systems that estimate what each user is likely to engage with, often drawing on signals such as clicks, likes, shares, replies, watch time and past behaviour.

A 2025 PNAS Nexus study on engagement and user satisfaction found that social media ranking algorithms typically optimise for “revealed preferences” such as clicks, shares and likes. The researchers’ audit of Twitter/X-style ranking found that engagement-based ranking amplified emotionally charged and out-group hostile political content compared with a reverse-chronological baseline, even though users did not necessarily prefer what the algorithm selected when asked directly. [PMC]pmc.ncbi.nlm.nih.govby S Milli · 2025 · Cited by 259 — Abstract. Social media ranking algorithms typically optimize for users' revealed preferences, i.e…

That finding captures the difference between what people click and what they value. A user may pause on an angry post, reply to correct it, quote it in disbelief or watch a misleading clip to the end. To an engagement system, all of these behaviours can look like interest. Unless the platform has better signals, reaction can be converted into distribution.

A separate line of modelling and empirical work on engagement ranking reaches a similar warning. Research on “Ranking for Engagement” argues that giving greater weight to online social interactions such as likes and shares can increase platform engagement while also increasing misinformation and polarisation. Its authors describe this as a trade-off: the same design choices that make a feed lively can also crowd out truth-oriented signals. [ifo Institut]ifo.deifo InstitutRanking for Engagement: How Social Media Algorithms…by F Germano · Cited by 20 — This paper investigates the dynamic feedb…

The Facebook “Meaningful Social Interactions” change in 2018 is a useful concrete example. Facebook said it would prioritise posts that generated interaction among friends and family, aiming to make time on the platform feel more meaningful. Coverage at the time noted that the change would reduce the reach of some publisher and brand content while favouring posts that prompted conversation. Later research has used that shift as evidence for studying how stronger weighting of interaction can affect polarisation and misinformation. [Time]time.comOpen source on time.com.

The problem is not that conversation is bad. Replies, corrections and personal testimony can be valuable. The problem is that “conversation” is an ambiguous metric. A careful explanation, an inflammatory rumour and a misleading AI-generated image can all generate comments. If the system rewards the reaction before assessing the reason for the reaction, the feed can become a machine for promoting whatever makes people respond fastest.

Reaction-first formats narrow judgement

Shareability incentives also shape what a post looks like. Platforms reward posts that are easy to recognise, remix and respond to. That favours short claims, visual proof, emotional framing and clear in-group or out-group cues. In this environment, the most shareable version of an issue is often not the most accurate one.

A misleading post can be designed for low-friction spread. It may use a screenshot rather than a link, because screenshots travel across platforms and detach a claim from its source. It may use a cropped clip, because the missing context is invisible at the point of sharing. It may use a dramatic caption, because the caption supplies the interpretation before the viewer has time to ask what happened before or after the clip.

AI makes this more powerful because it lowers the cost of producing polished, emotionally legible material. Synthetic images, fake quotes, generated summaries and confident-looking explainers can be produced quickly and then adapted to the style of each platform. The shareability incentive remains the same: the post that is easiest to understand and react to may beat the post that is most careful.

This is why critical thinking on social platforms has to include the structure around the post. A user should ask not only “What does this claim say?” but also:

None of this means a viral post is false. It means virality is a separate property from truth. A claim can be true and viral, false and viral, or true but ignored because it is complex, slow or emotionally unsatisfying.

Design choices can change what users notice

The incentive structure is not fixed. Experiments suggest that small design changes can shift behaviour, though no single intervention solves the problem.

One approach is to redirect attention to accuracy. Pennycook and colleagues’ accuracy-nudge studies show that asking users to consider whether a headline is accurate can improve the quality of what they later share. The point is not that people need a lecture; it is that the sharing moment can be redesigned so that truth becomes more salient. [Nature]nature.comShifting attention to accuracy can reduce misinformation…by G Pennycook · 2021 · Cited by 1605 — The results show that subtly sh…

A second approach is to change the reward itself. Globig and colleagues tested whether social rewards and punishments could be made contingent on information veracity. Across six experiments with 951 participants, they found that adjusting the incentive structure in this way increased sharing discernment: participants shared a higher proportion of true information relative to false information. [PMC]pmc.ncbi.nlm.nih.govOpen source on nih.gov.

A third approach is to add corrective context to misleading posts. Research on X’s Community Notes is mixed but useful. An earlier large-scale study found no evidence that the rollout of Community Notes significantly reduced engagement with misleading tweets, partly because notes may arrive too late in the viral cycle. A later causal study of 40,074 posts found that once notes were attached, reposts, likes, replies and views fell, with average reposts down by about 45.7% and likes down by about 43.5% after attachment. [arXiv]arxiv.orgOpen source on arxiv.org.

That contrast is revealing. Corrections can work, but timing matters. If a misleading post receives most of its attention before a note appears, the platform has already rewarded the content and taught the poster that the tactic can succeed. A correction that arrives late may reduce further spread without undoing the original incentive.

Design changes around visible metrics are another contested area. Instagram and Facebook introduced options to hide public like counts, partly in response to concerns about social pressure and competitive posting. However, hiding a number from users does not necessarily remove the metric from ranking, creator analytics or advertiser systems. Vogue Business reported that Instagram’s like-hiding option did not affect the algorithm and that creators could still access performance metrics. [Vogue]voguebusiness.comVogue Instagram will let users hide the "like". Should brands care?Vogue Instagram will let users hide the "like". Should brands care?

This distinction matters for critical thinking. A cosmetic change to the interface may reduce some public pressure, but the deeper incentive remains if the platform still ranks and rewards content by engagement. The feed may look less numerical while still being governed by numbers behind the scenes.

Why “just think harder” is an incomplete answer

Individual critical thinking is necessary, but shareability incentives show why it is not sufficient. The user is making judgements inside a designed environment. That environment can reward speed over patience, confidence over uncertainty, emotional clarity over nuance and visibility over accuracy.

This does not remove personal responsibility. People still choose whether to share, quote, mock, correct or pause. But the platform can make some choices feel natural and others feel costly. Checking a source takes time. Withholding a repost gives no public reward. Saying “I am not sure yet” rarely travels as far as a confident claim. In a feed built around reaction, restraint is often invisible.

The more useful habit is to treat shareability as a signal to inspect rather than obey. A highly shareable post deserves a brief pause precisely because it has been optimised for movement. Before sharing, the reader can ask: Is the original source visible? Is the date clear? Would the claim still seem strong without the caption? Is the post asking me to verify, or only to react?

For AI-generated content, the same questions become sharper. Fluency, visual polish and confident wording are not proof. A generated post can be designed to fit the emotional and stylistic patterns that platforms reward. Critical thinking therefore means separating presentation quality from evidential quality.

What better incentives would reward

A healthier platform would not need to make social media slow, dull or sterile. The goal is not to eliminate sharing, humour, emotion or public conversation. The goal is to stop treating raw reaction as the closest available substitute for value.

Better incentives would reward signals that are closer to judgement:

Research on stated preferences suggests this is not merely idealistic. The PNAS Nexus study found that ranking by users’ stated preferences could reduce angry, partisan and out-group hostile content, although it also raised trade-offs such as potentially reinforcing content that users already agree with. [PMC]pmc.ncbi.nlm.nih.govby S Milli · 2025 · Cited by 259 — Abstract. Social media ranking algorithms typically optimize for users' revealed preferences, i.e…

That is the central tension. Platforms cannot simply optimise for “truth” as if it were always easy to measure. But they can decide whether the default system rewards the fastest reaction or creates more room for reflection. For readers, recognising that incentive structure is a core part of modern critical thinking: before asking whether a post is true, ask why this particular post was made so easy to see, reward and share.

Amazon book picks

Further Reading

Books and field guides related to How Feeds Reward Reaction Before Reflection. Use these as the next step if you want deeper reading beyond the article.

Book

The Chaos Machine

By Max Fisher

Directly explains how platform incentives and engagement systems shape behaviour and information sharing.

See on Amazon

Book

The Attention Merchants

By Tim Wu

Explains how businesses compete for attention and build systems around engagement.

See on Amazon

Book

Stolen Focus

By Johann Hari

Connects attention economics and digital environments to reduced reflection and judgement.

See on Amazon

Book

The Shallows

By Nicholas Carr

Examines how online environments affect thinking, concentration and judgement.

See on Amazon

Browse more on Amazon: The Chaos Machine The Attention Merchants Stolen Focus

As an Amazon Associate I earn from qualifying purchases.

eBay marketplace picks

Marketplace Samples

Example marketplace items related to this page. Use the search link to explore similar finds on eBay.

Shop location

Using USA USA USAUKCanadaAustraliaIreland

Example eBay listing

Disobey Mug Anarchist Stuff Vendetta Self Ownership Critical Thinking Question

Search eBay.co.uk: critical thinking mug

Browse similar on eBay.co.uk

Example eBay listing

Realist Mug Black Coffee Cup Truth Logic Critical Thinking Objectivism

Search eBay.co.uk: critical thinking mug

Browse similar on eBay.co.uk

Example eBay listing

Critical Thinking Mug Homeschool Mom Ayn Rand Question Everything Free Thinker

Search eBay.co.uk: critical thinking mug

Browse similar on eBay.co.uk

Example eBay listing

Critical thinking meme mug coffee tea office gift white elephant

Search eBay.co.uk: critical thinking mug

Browse similar on eBay.co.uk

Browse more on eBay.co.uk

Example items shown for inspiration; availability and pricing can change. Branchoria may earn a commission if you purchase through outbound eBay links.

Endnotes

1. Source: nature.com
   Link: https://www.nature.com/articles/s41586-021-03344-2
   Source snippet

   Shifting attention to accuracy can reduce misinformation...by G Pennycook · 2021 · Cited by 1605 — The results show that subtly sh...

2. Source: pnas.org
   Link: https://www.pnas.org/doi/10.1073/pnas.2216614120
   Source snippet

   Sharing of misinformation is habitual, not just lazy or biasedby G Ceylan · 2023 · Cited by 224 — Habitual sharing of misinformation...

3. Source: pmc.ncbi.nlm.nih.gov
   Link: https://pmc.ncbi.nlm.nih.gov/articles/PMC11894805/
   Source snippet

   by S Milli · 2025 · Cited by 259 — Abstract. Social media ranking algorithms typically optimize for users' revealed preferences, i.e...

4. Source: pmc.ncbi.nlm.nih.gov
   Title: PMCNudging Social Media toward Accuracy
   Link: https://pmc.ncbi.nlm.nih.gov/articles/PMC9082967/
   Source snippet

   NIHby G Pennycook · 2022 · Cited by 131 — We review research that shows how a simple nudge or prompt that shifts attention to accur...

5. Source: arxiv.org
   Link: https://arxiv.org/abs/2305.16941

6. Source: ifo.de
   Link: https://www.ifo.de/en/cesifo/publications/2026/working-paper/ranking-engagement-how-social-media-algorithms-fuel-misinformation
   Source snippet

   ifo InstitutRanking for Engagement: How Social Media Algorithms...by F Germano · Cited by 20 — This paper investigates the dynamic feedb...

7. Source: time.com
   Link: https://time.com/5100245/facebook-newsfeed-update-posts-friends-family/

8. Source: pmc.ncbi.nlm.nih.gov
   Link: https://pmc.ncbi.nlm.nih.gov/articles/PMC10259455/

9. Source: arxiv.org
   Link: https://arxiv.org/abs/2307.07960

10. Source: arxiv.org
    Link: https://arxiv.org/abs/2502.13322

11. Source: arxiv.org
    Link: https://arxiv.org/html/2305.16941v6

12. Source: arxiv.org
    Link: https://arxiv.org/html/2601.11140v1

13. Source: arxiv.org
    Link: https://arxiv.org/html/2502.13322v1

14. Source: pnas.org
    Link: https://www.pnas.org/doi/abs/10.1073/pnas.2216614120

15. Source: nature.com
    Link: https://www.nature.com/articles/s41467-026-72597-0

16. Source: instagram.com
    Title: Cn6U5f Gu3Oj
    Link: https://www.instagram.com/p/Cn6U5fGu3Oj/

17. Source: facebook.com
    Link: https://www.facebook.com/groups/3741333852807065/posts/4348137645460013/

18. Source: youtube.com
    Link: https://www.youtube.com/watch?v=QqQkjAOIbAo
    Source snippet

    Reclaiming Your Attention: How the Attention Economy Controls Your Focus...

19. Source: pubmed.ncbi.nlm.nih.gov
    Link: https://pubmed.ncbi.nlm.nih.gov/37278047/

20. Source: voguebusiness.com
    Title: Vogue Instagram will let users hide the “like”. Should brands care?
    Link: https://www.voguebusiness.com/technology/instagram-will-let-users-hide-the-like-should-brands-care

21. Source: pmc.ncbi.nlm.nih.gov
    Link: https://pmc.ncbi.nlm.nih.gov/articles/PMC11373151/

22. Source: pmc.ncbi.nlm.nih.gov
    Link: https://pmc.ncbi.nlm.nih.gov/articles/PMC12478135/

23. Source: pmc.ncbi.nlm.nih.gov
    Link: https://pmc.ncbi.nlm.nih.gov/articles/PMC10631661/

24. Source: pubmed.ncbi.nlm.nih.gov
    Link: https://pubmed.ncbi.nlm.nih.gov/30036748/

25. Source: frontiersin.org
    Link: https://www.frontiersin.org/journals/communication/articles/10.3389/fcomm.2025.1667471/full

26. Source: communitynotes.x.com
    Link: https://communitynotes.x.com/guide/en/about/introduction

27. Source: assets.publishing.service.gov.uk
    Title: Facebook Response to CMA Algorithms Consultation
    Link: https://assets.publishing.service.gov.uk/media/60c9d224d3bf7f4bca6e7379/Facebook_Response_to_CMA_Algorithms_Consultation.pdf

Additional References

1. Source: insights.som.yale.edu
   Title: how social media rewards misinformation
   Link: https://insights.som.yale.edu/insights/how-social-media-rewards-misinformation
   Source snippet

   Ceylan found that the reward systems of social media platforms are inadvertently encouraging users to spread misinformation. By constantl...

2. Source: youtube.com
   Title: Reclaiming Your Attention: How the Attention Economy Controls Your Focus
   Link: https://www.youtube.com/watch?v=6mBnOnOrqhI
   Source snippet

   The Attention Economy: Why Truth is Becoming Irrelevant...

3. Source: youtube.com
   Title: The Attention Economy: Why Truth is Becoming Irrelevant
   Link: https://www.youtube.com/watch?v=AAPK3fJI8sY
   Source snippet

   Designed for Outrage: How Algorithms Profit from Hate...

4. Source: researchgate.net
   Link: https://www.researchgate.net/publication/342579856_Fighting_COVID-19_Misinformation_on_Social_Media_Experimental_Evidence_for_a_Scalable_Accuracy-Nudge_Intervention

5. Source: researchgate.net
   Link: https://www.researchgate.net/publication/371339281_Changing_the_incentive_structure_of_social_media_platforms_to_halt_the_spread_of_misinformation

6. Source: researchgate.net
   Link: https://www.researchgate.net/publication/395572441_THE_VIRALITY_TRAP_HOW_SOCIAL_MEDIA_ALGORITHMS_PRIORITIZE_ENGAGEMENT_OVER_TRUTH_IN_SHAPING_PUBLIC_OPINION

7. Source: researchgate.net
   Link: https://www.researchgate.net/publication/397763733_The_Role_of_Social_Media_Algorithms_in_Amplifying_Misinformation_in_Political_Elections

8. Source: researchgate.net
   Link: https://www.researchgate.net/publication/401405020_Ranking_for_engagement_How_social_media_algorithms_fuel_misinformation_and_polarization

9. Source: knightcolumbia.org
   Link: https://knightcolumbia.org/content/engagement-user-satisfaction-and-the-amplification-of-divisive-content-on-social-media

10. Source: cambridge.org
    Link: https://www.cambridge.org/core/journals/behavioural-public-policy/article/how-does-the-design-of-social-media-content-controls-shape-users-choice-evidence-from-an-online-experiment/4FF2B810CB0E88EC2DF28AFF536B9C97