The Wisdom of the Crowd is not a Forgone Conclusion. Effects of Self-Selection on (Collaborative) Knowledge Construction

TOPICS IN COGNITIVE SCIENCE(2024)

引用 0|浏览9
暂无评分
摘要
Web 2.0 has elevated the possibilities of collaboration to unprecedented levels. Therein lies great potential, as the aptly coined phenomenon "Wisdom of the Crowd" implies. When it comes to controversial topics, however, there is no safety in numbers alone. On the contrary, collaboration among only like-minded people may even exacerbate biases (e.g., Echo Chambers). Yet, it is human nature to seek out like-minded others. Consequently, the process of self-selection is crucial if the heterogeneity of opinions serves as a safeguard against undesirable effects of group processes (e.g., attitude polarization). Accordingly, online environments that invite more heterogeneous (vs. homogeneous) users should produce less biased content. We tested this hypothesis in a field study, comparing articles on the same 20 controversial topics from the online encyclopedias Conservapedia and RationalWiki with Wikipedia (and Britannica serving as a gold standard) and exploring the opinions of discussants in the three online encyclopedias. As expected, articles from Conservapedia and RationalWiki were significantly less balanced than articles from Wikipedia and Britannica. We replicated this finding in a lab study with 257 participants who self-selected to one of three online wikis (Vegan Love, Nutrition, Meat & Fish) and individually as well as collaboratively wrote an encyclopedia-like article about "Diets." As expected, Wikis with a specific focus (Vegan Love, Meat & Fish) predominantly attracted authors with a positive attitude toward this focus and, as a consequence, resulted in more biased content than in the Nutrition Wiki. Overall, our results suggest that crowds alone do not guarantee wisdom-self-selection is a crucial process that needs to be taken into account.
更多
查看译文
关键词
Bias,Information processing,Self-selection,Collaboration,Wisdom of the crowds
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要