Chapter 20 Further Reading: The Outrage Machine: Anger as Engagement

Foundational Research Papers

  1. 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. The foundational empirical study documenting the six-fold spread advantage of moral-emotional language on Twitter. The paper's analysis of 563,312 tweets about gun control, same-sex marriage, and climate change provides the most rigorously quantified evidence of outrage's virality advantage. Essential primary reading for anyone engaging seriously with the outrage amplification literature. The paper's methodology section rewards careful study for the tools it provides for analyzing political communication.

  2. 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. The controversial emotional contagion study demonstrating that Facebook's algorithmic curation actively shapes users' emotional states. Despite ongoing ethical critique, the paper's core finding — that manipulating News Feed emotional valence changed users' emotional expression — is one of the most significant results in social media psychology. The ethical controversy is itself worth studying as a case in research ethics and platform governance.

  3. Berger, J., & Milkman, K. L. (2012). "What makes online content viral?" Journal of Marketing Research, 49(2), 192-205. The landmark study establishing the relationship between emotional arousal and content virality, based on analysis of nearly 7,000 New York Times articles. Berger and Milkman's finding that high-arousal emotions (anger, anxiety, awe) predict sharing while low-arousal sadness does not provides the emotional science foundation for understanding why engagement-maximizing algorithms converge on outrage content.

  4. Ribeiro, M. H., Ottoni, R., West, R., Almeida, V. A. F., & Meira, W. (2019). "Auditing Radicalization Pathways on YouTube." arXiv:1908.08313. (Published in Proceedings of ACM FAT* 2020.) The systematic academic documentation of recommendation-driven radicalization pathways on YouTube, mapping the Alternative Influence Network and analyzing audience migration patterns across channels of increasing political extremity. The paper's methodology and limitations are as important as its findings, illustrating both what is and isn't knowable through external research with limited data access.

Books

  1. Haidt, J. (2012). The Righteous Mind: Why Good People Are Divided by Politics and Religion. Pantheon. The fullest statement of Haidt's moral foundations theory, explaining how different political communities are organized around different fundamental moral intuitions. The book provides essential theoretical scaffolding for understanding why outrage amplification follows tribal patterns — why content that outrages one moral community may be invisible to another, and how algorithms learn to customize outrage delivery to each community's moral profile.

  2. Haidt, J. (2023). The Anxious Generation: How the Great Rewiring of Childhood Is Causing an Epidemic of Mental Illness. Penguin Press. Haidt's most recent book broadening his critique of social media's social effects to mental health, particularly for adolescents. While the book's claims have been contested by some researchers, it provides a readable synthesis of evidence for social media's emotional and psychological effects. The chapters on social contagion and algorithm dynamics are directly relevant to the outrage amplification discussion.

  3. Vosoughi, S., Roy, D., & Aral, S. (2018). "The spread of true and false news online." Science, 359(6380), 1146-1151. This Science paper found that false news spreads faster than true news on Twitter, and that human behavior rather than bots accounts for this difference. False news tends to be more novel and more emotionally activating — and thus more shareable — than true news. The paper's findings complement Brady et al. by suggesting that the moral-emotional language that drives outrage sharing may also select for less accurate content.

  4. Sunstein, C. R. (2017). #Republic: Divided Democracy in the Age of Social Media. Princeton University Press. An updated version of Sunstein's long-running argument about social media and democratic deliberation. While some of Sunstein's filter bubble claims have been challenged by subsequent research, the book provides a thoughtful legal and political theory perspective on how algorithmic curation of outrage content relates to the social prerequisites for democratic self-governance. Useful for the democratic stakes section of the chapter.

Journalism and Investigative Reporting

  1. The Wall Street Journal. (September-October 2021). "The Facebook Files." Investigative series by Jeff Horwitz, Keach Hagey, and others. The definitive journalistic account of Facebook's internal research on platform harms, drawing on thousands of pages of documents shared by Frances Haugen. The series is essential reading for anyone studying the gap between platform internal knowledge and external accountability. Individual installments cover the outrage algorithm, teen mental health, and political content, each providing specific documented cases of known harm and inadequate response.

  2. Lewis, P. (February 2, 2018). "'Fiction is outperforming reality': how YouTube's algorithm distorts truth." The Guardian. The Guardian investigation drawing on Chaslot's AlgoTransparency data that brought YouTube's recommendation dynamics to broad public attention. The article documents specific recommendation pathways from mainstream to extreme content and quotes YouTube's official response. A historically important piece in the development of public and regulatory understanding of recommendation algorithm radicalization.

  3. Horwitz, J., & Seetharaman, D. (May 26, 2020). "Facebook Executives Shut Down Efforts to Make the Site Less Divisive." The Wall Street Journal. A key early piece in the Facebook accountability journalism series, reporting that Facebook researchers developed and proposed changes to reduce divisiveness in the News Feed, and that these proposals were rejected or significantly modified by executives concerned about engagement metrics. Provides crucial context for understanding the organizational dynamics behind the Facebook Files revelations.

  1. Bail, C. A., Argyle, L. P., Brown, T. W., Bumpus, J. P., Chen, H., Hunzaker, M. B. F., et al. (2018). "Exposure to opposing views on social media can increase political polarization." Proceedings of the National Academy of Sciences, 115(37), 9216-9221. An important counterintuitive finding: exposure to opposing political views on social media increased polarization rather than reducing it. This finding complicates simple "echo chamber" models of outrage dynamics and suggests that cross-cutting exposure (having outrage-inducing content from the other side appear in your feed) may be as harmful as homogeneous exposure.

  2. Carver, C. S., & Harmon-Jones, E. (2009). "Anger is an approach-related affect: Evidence and implications." Psychological Bulletin, 135(2), 183-204. The foundational psychological research establishing anger as an approach-motivating emotion, providing the neurological and behavioral science basis for understanding why anger drives action (sharing, commenting, engaging) rather than withdrawal. Essential for the neuroscientific foundation of the chapter's opening section.

  3. Huszár, F., Ktena, S. I., O'Brien, C., Belli, L., Schlaikjer, A., & Hardt, M. (2022). "Algorithmic amplification of politics on Twitter." Proceedings of the National Academy of Sciences, 119(1), e2025334119. A rare case of platform researchers publishing research on their own platform's political content dynamics. The study found that Twitter's algorithm amplified content from right-leaning political accounts more than left-leaning accounts across most countries studied. The paper demonstrates what research with full platform data access can achieve, and illustrates the gap between this and what external researchers can do.

  4. Starbird, K., Arif, A., & Wilson, T. (2019). "Disinformation as Collaborative Work: Surfacing the Participatory Nature of Strategic Information Operations." Proceedings of the ACM on Human-Computer Interaction, 3(CSCW), 1-26. Research on how outrage content and disinformation spread through collaborative social media behavior — showing that algorithmically amplified outrage and strategically produced disinformation often operate through similar participatory mechanisms. Relevant to understanding how the outrage machine intersects with deliberate disinformation operations.

Policy and Regulatory Analysis

  1. Acemoglu, D. (2021). "Harms of AI." NBER Working Paper No. w29247. National Bureau of Economic Research. An economist's analysis of how optimization systems cause social harms through misaligned objectives — directly relevant to the algorithmic outrage amplification problem. Acemoglu's framework of "directed technological change" and its social effects provides an economic theory foundation for understanding why outrage amplification is a structural rather than incidental feature of engagement-optimized platforms.

  2. Roth, Y. (2022). "What Happened at Twitter." Wired. (Interview and related publication.) Perspective from a former Head of Trust and Safety at Twitter on the internal dynamics of managing platform harms, including outrage content. Provides insider context on the organizational and technical challenges of moderating and reducing outrage amplification within platforms constrained by business and free speech considerations.

  3. European Parliament. (2022). "Digital Services Act." Regulation (EU) 2022/2065. The EU's landmark regulation on online platform accountability, which includes requirements for large platforms to assess and mitigate systemic risks from their systems — including recommendation algorithms. The DSA's provisions on algorithmic transparency and risk assessment represent the most developed current regulatory response to outrage amplification and related algorithm harms. Essential for the policy and regulatory analysis dimension.

Books for Broader Context

  1. Lanier, J. (2018). Ten Arguments for Deleting Your Social Media Accounts Right Now. Henry Holt. A provocative polemic from one of the internet's most prominent critical voices. Lanier's argument that the behavior modification business model of social media is fundamentally corrupting human discourse provides a useful foil for more nuanced academic analyses. The book is more useful as a readable articulation of a radical critique than as a rigorous argument.

  2. Settle, J. E. (2018). Frenemies: How Social Media Polarizes America. Cambridge University Press. A political scientist's rigorous empirical examination of how social media contributes to political polarization, with careful attention to the mechanisms that distinguish legitimate correlation effects from causal claims. Settle's work provides important methodological context for evaluating the outrage amplification research discussed in this chapter, including its limitations and what it would take to establish stronger causal claims.