{
  "id": 16326466,
  "title": "A hidden bias in human language challenges 70 years of science",
  "url": "https://urgent.news/2026/10/24/a-hidden-bias-in-human-language-challenges-70-years-of-science",
  "topic": "science",
  "section": "Science",
  "published": "2026-10-24T13:01:22.000Z",
  "source": {
    "name": "ScienceDaily",
    "slug": "sciencedaily",
    "url": "https://www.sciencedaily.com/releases/2026/10/261023023738.htm"
  },
  "original_language": "en",
  "account": "Researchers at the University of Vermont have uncovered a hidden bias in human language that contradicts a theory widely accepted for over 70 years. Their groundbreaking study published in Science Advances introduces a new method called ousiometrics, which employs mathematical techniques to measure the fundamental meaning of words. The study's most significant finding is a consistent preference for safety over danger in the words people use. This safety bias could revolutionize how scientists study language, understand human behavior, and create artificial intelligence systems that interpret speech and writing. For decades, the dominant theory of word meaning, known as the VAD model, relied on three emotional dimensions: valence (positive vs. negative), arousal (excited vs. calm), and dominance (controlling vs. submissive). However, the University of Vermont researchers discovered that this long-standing model fails to account for essential aspects of word meaning. Their analysis of billions of word uses across various real-world texts revealed three distinct dimensions: power (weak vs. powerful), danger (safe vs. dangerous), and structure (ordered vs. chaotic). The new framework explained over 90% of the variation in word meaning, compared to 72% explained by the traditional VAD model. The researchers found that safety-related words appear far more frequently in everyday communication than danger-related words, providing a new explanation for the Pollyanna principle, which describes humans' tendency to favor positive over negative words. According to the study, this Pollyanna principle is actually a one-dimensional reflection of an underlying safety bias. The findings suggest that the structure of human language may reflect evolutionary pressures related to survival, as being able to distinguish between safety and danger is crucial for staying alive. The researchers propose that the emphasis on safety in language may also influence how meaning is organized, reflecting experiences fundamental to human life such as power, danger, and structure. These insights could have significant implications for artificial intelligence, particularly for systems that rely on sentiment analysis and other approaches to understanding human communication.",
  "summary": "A groundbreaking study challenges 70 years of assumptions about how human language works. By analyzing billions of word uses, scientists discovered that meaning is organized around power, danger, and order rather than emotion alone. Even more surprisingly, human language shows a powerful hidden preference for safety, suggesting that our words may reflect ancient survival instincts.",
  "key_points": [
    "Researchers discover hidden bias favoring safety over danger in human language",
    "New method called ousiometrics reveals safety bias contradicts 70-year-old VAD model",
    "Safety bias may impact AI systems interpreting speech and writing"
  ],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}