Identifying Fast/Slow Thinking without Prior

ArXiv(2021)

引用 0|浏览3
暂无评分
摘要
System 1 vs. 2 theory [6, 19] describes two modes of thought, a fast, instinctive one and a slow, logical one. When we ask a question (e.g. A bat and ball cost $1.10. The bat costs $1 more than the ball. How much does the ball cost?), with prior, we can identify fast/slow thinking ($.10/$.05). But what if we do not have prior? A very clever method, surprisingly popular [13], additionally asks what percentage of other people answer $.10/$.05 and selects the answer that is more popular than people predict. However, the distribution report is non-minimal for many people especially for non-binary choices, the choices design requires prior and only the best answer is selected. Here we propose a simple minimal paradigm that elicits the full hierarchy of the collected answers: we ask a single open response question and elicit each respondent’s answer (e.g. $.05) and guess(es) for other people’s answers (e.g. $.10). We record the number of people who report a specific answer-guess pair (e.g. 10 people answer $.05 and guess $.10) by an answer-guess matrix. By ranking the answers to maximize the sum of the upper triangular area of the matrix, we obtain and visualize the hierarchy of the answers without any prior. Our paradigm has minimal requirement for both the respondent (no distribution report) and the requester (no choices design; check the hierarchy visually) and can be also used to research how people reason about other people’s minds [14]. We conduct four studies including math problems, Life-and-death Go problems, general knowledge and pronunciation questions. The empirical results not only validate our approach but also show that more sophisticated people can reason about less sophisticated people’s minds.
更多
查看译文
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