The Toxoplasma of Rage
Updated 2026-08-08
INTRODUCTION
English translation pending.
CORE DEFINITION
Coined by Scott Alexander in an essay. The core claim is that the content which travels furthest online is not the most correct or most consequential, but the most divisive, because it simultaneously gratifies one group and outrages another and therefore generates maximum engagement. Recommendation systems optimized for engagement then amplify whatever produces the most interaction, so the most inflammatory framings win visibility while broadly shared but unexciting consensus is filtered out. The qualifier matters: the argument concerns the selection pressure of the distribution channel, not the truth of the claims themselves.
SCAFFOLDING EFFECT
Reduce cognitive load
- Spread diagnosis: ask whether a trending fight is genuinely important or simply engineered to maximize outrage. - Algorithm audit: check whether a platform's engagement incentives reward anger over accuracy. - Consensus hunting: deliberately seek the boring shared ground that the feed hides from you.
Anchor fast decisions
Some claims carry a built-in controversy gradient: they delight one tribe and enrage another at the same time. That dual reaction roughly doubles the interaction volume, and recommendation systems that treat engagement as a quality signal read volume as merit. The most polarizing framing therefore outcompetes the most accurate one, and rage becomes the fuel that carries an idea from feed to feed. Content that offends nobody produces no reaction, so it never enters the competition at all.
MINIMUM ACTION
In progress 0/1Practice this model in one real situation:
account_treeGenealogyexpand_more
menu_bookReferencesexpand_more
Source support: Explicit
- slatestarcodex.comhttps://slatestarcodex.com/2014/12/17/the-toxoplasma-of-rage/verified
PRIVATE NOTES · Only visible to you
SAVED Q&A
ENTRY Q&A · Private saving available
Ask with a clear boundary
thinkingmodels answers from published entry context only.
Your question is sent to thinkingmodels. The answer uses public entry context only.
RELATED MODELS