Unverified50% confidenceFactExact time
Academic research indicates that algorithms optimizing for dwell time tend to push content that triggers emotional fluctuations, especially anger and anxiety.
1
Sources
50%
Confidence
Long-term
Relevance
9/1/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe core metrics for content distribution platforms are user dwell time and engagement rate, and algorithms favor content that triggers strong emotions such as fear, anger, and anxiety (negativity bias).77% similarUnverified当推荐模型的优化目标被设定为最大化engagement时，模型会系统性给予情绪化内容更高分发权重，这被研究者称为算法放大效应（algorithmic amplification）68% similarUnverifiedFacebook前数据科学家Frances Haugen披露的内部文件显示，平台明知带有愤怒情绪的错误信息传播效率更高，却因影响商业指标而迟迟未做有效干预68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/838722API
curl https://kongchang.com/api/v1/knowledge/claims/838722MCP
get_claim(id=838722)