显著性(神经科学)
突出
好奇心
心理学
价(化学)
信息搜寻
社会心理学
认知心理学
计算机科学
人工智能
量子力学
物理
图书馆学
作者
Russell Golman,George Loewenstein,András Molnár,Silvia Saccardo
出处
期刊:Management Science
[Institute for Operations Research and the Management Sciences]
日期:2022-09-01
卷期号:68 (9): 6454-6476
被引量:55
标识
DOI:10.1287/mnsc.2021.4244
摘要
Management scientists recognize that decision making depends on the information people have but lack a unified behavioral theory of the demand for (and avoidance of) information. Drawing on an existing theoretical framework in which utility depends on beliefs and the attention paid to them, we develop and test a theory of the demand for information encompassing instrumental considerations, curiosity, and desire to direct attention to beliefs one feels good about. We decompose an individual’s demand for information into the desire to refine beliefs, holding attention constant, and the desire to focus attention on anticipated beliefs, holding these beliefs constant. Because the utility of resolving uncertainty (i.e., refining beliefs) depends on the attention paid to it and more important or salient questions capture more attention, demand for information depends on the importance and salience of the question(s) it addresses. In addition, because getting new information focuses attention on one’s beliefs and people want to savor good news and ignore bad news, the desire to obtain or avoid information depends on the valence (i.e., goodness or badness) of anticipated beliefs. Five experiments (n = 2,361) test and find support for these hypotheses, looking at neutrally valenced as well as ego-relevant information. People are indeed more inclined to acquire information (a) when it feels more important, even if it cannot aid decision making (Experiments 1A and 2A); (b) when a question is more salient, manipulated through time lag (Experiments 1B and 2B); and (c) when anticipated beliefs have higher valence (Experiment 2C). This paper was accepted by Yan Chen, behavioral economics and decision analysis.
科研通智能强力驱动
Strongly Powered by AbleSci AI