佛蒙特大学研究团队在《科学进展》杂志发表的研究对长期主导语言科学的理论提出了质疑1。研究人员通过分析超过20,000个词汇在真实文本中数十亿次的使用情况，提出了一个新的语言意义组织框架1。这一框架由权力、危险和秩序三个维度构成，用于解释人类如何组织和理解语言意义1。
与传统的价、激发度和主导性模型相比，新框架在解释词义变异方面表现出了显著优势1。新模型能够解释超过90%的词义变异，而传统模型约为72%1。研究最重要的发现是人类语言中存在对安全的一致性偏好，词汇使用中安全相关词汇的出现频率明显高于危险相关词汇1。由彼得·多德斯和克里斯·丹福斯领导的研究团队得到了美国国家科学基金会、谷歌和MassMutual等机构的资助1。
Researchers at the University of Vermont have published findings in Science Advances that fundamentally challenge 70 years of established science on how human language conveys meaning. 1 By analyzing billions of word uses across more than 20,000 vocabulary items in real-world texts, the team led by Peter Dodds and Chris Danforth identified a new organizational framework for semantic meaning based on three dimensions: power (weak versus strong), danger (safe versus dangerous), and structure (orderly versus chaotic). 1 This framework accounts for over 90 percent of variation in word meanings, substantially outperforming the traditional Valence-Arousal-Dominance (VAD) model, which explains approximately 72 percent. 1
A striking discovery emerged from the analysis: human language displays a consistent bias toward safety. 1 Words related to safety appear significantly more frequently than those conveying danger across the vocabulary studied. 1 This systematic preference provides insight into how people naturally organize and express meaning through language. 1 The research was supported by the National Science Foundation, Google, and MassMutual, among other institutions. 1
评论
还没有评论，欢迎留下第一条。