描述信息和经验信息对风险决策的不同影响
发布时间:2018-09-10 09:31
【摘要】:以往关于风险决策的研究大多将每个备择方案的结果和相应概率以文字或图表形式直接呈现给决策者。通过上述方式呈现的信息被研究者们称为描述信息。但在现实生活中,人们并非或不仅仅依靠描述信息进行风险决策,在很多情境中,人们利用自身经验对结果发生的可能性进行判断,从而做出风险决策,即基于经验信息做风险决策。而相关研究显示基于描述信息和基于经验信息做出的风险决策是存在差异的。本文通过三个研究探讨了在单独或同时呈现描述、经验信息时,两类信息对风险决策的影响存在何种差异。研究一单独呈现描述信息和经验信息,并沿用Hertwig等人使用的抽样范式呈现后者。结果显示基于描述信息的风险决策结果符合预期理论。基于经验信息的风险决策结果与前者不同,在中高概率获益情景中,相比于描述信息,被试在基于经验信息做决策时更冒险。此现象在群体和个体水平都得到了验证。研究二中实验组同时接受两类信息,对照组仅接受其中一类信息。结果表明实验组被试的风险决策结果与仅接受经验信息组被试的风险决策结果无显著差异,而与仅接受描述信息组的差异显著,被试的风险决策结果受经验信息的影响更大。主观估计概率和决策后报告的经验信息影响权重也都验证了上述结论。研究三采用更真实的抽球实验呈现经验信息,结果重复验证了在同时接受两类信息时,被试的风险决策和主观概率估计受经验信息的影响更大。此外,研究三也探究了对描述信息信任程度是否中介了两类信息提供概率的偏离程度对风险决策的影响。结果表明仅在经验信息对应的概率高于描述信息时,对描述信息信任程度起到了中介作用,即经验概率高于描述概率的程度越大,被试对描述信息信任度越低,从而更偏好于选择风险选项。以上结论为描述-经验差距提供了新的证据,且验证了两类信息同时呈现时,经验信息对人们的风险决策和主观概率估计影响更大,并初步探索了其中的机制。上述研究结论启示我们应关注经验信息在风险决策中的作用。
[Abstract]:Most of the previous studies on risk decision have presented the results of each alternative scheme and the corresponding probability directly to the decision makers in the form of text or chart. The information presented in this way is referred to by researchers as descriptive information. But in real life, people do not or not rely on descriptive information to make risk decisions. In many situations, people use their own experience to judge the possibility of the outcome and make risk decisions. That is, based on empirical information to make risk decisions. Relevant studies show that there are differences in risk decisions based on descriptive information and empirical information. Through three studies, this paper discusses the difference between the two kinds of information when presenting the description and experience information separately or simultaneously, and how the two kinds of information affect the risk decision making. A separate presentation description and empirical information was studied and the latter was presented using the sampling paradigm used by Hertwig et al. The results show that the risk decision results based on descriptive information are in line with the expected theory. The result of risk decision based on empirical information is different from that of the former. In the case of medium and high probability benefit scenario, the participants are more adventurous in making decisions based on empirical information than in describing information. This phenomenon has been verified at both group and individual levels. In study 2, the experimental group received two types of information, while the control group received only one type of information. The results showed that there was no significant difference between the results of risk decision in the experimental group and those in the group receiving only empirical information, but there was a significant difference between the results of risk decision and that of the group receiving only descriptive information, and the results of the risk decision of the subjects were more affected by the information of experience. The above conclusions are verified by subjective estimation probability and the weight of empirical information reported after decision making. In the third study, the empirical information was presented by a more real pumping experiment. The results showed that the risk decision and subjective probability estimation of the subjects were more affected by the empirical information when the two types of information were accepted simultaneously. In addition, the third study also explores whether the degree of confidence in describing information mediates the influence of the deviation degree of the two types of information providing probability on risk decision-making. The results show that only when the corresponding probability of empirical information is higher than that of descriptive information, it plays an intermediary role in the degree of confidence in descriptive information, that is, the greater the degree of empirical probability is higher than the degree of description probability, the lower the trust degree of descriptive information is. Thus prefer to choose the risk option. The above conclusions provide new evidence for the descriptive-empirical gap and verify that when two types of information are presented at the same time, the empirical information has a greater impact on people's risk decision and subjective probability estimation, and the mechanism is preliminarily explored. These conclusions suggest that we should pay attention to the role of empirical information in risk decision-making.
【学位授予单位】:浙江大学
【学位级别】:硕士
【学位授予年份】:2017
【分类号】:B848
本文编号:2234056
[Abstract]:Most of the previous studies on risk decision have presented the results of each alternative scheme and the corresponding probability directly to the decision makers in the form of text or chart. The information presented in this way is referred to by researchers as descriptive information. But in real life, people do not or not rely on descriptive information to make risk decisions. In many situations, people use their own experience to judge the possibility of the outcome and make risk decisions. That is, based on empirical information to make risk decisions. Relevant studies show that there are differences in risk decisions based on descriptive information and empirical information. Through three studies, this paper discusses the difference between the two kinds of information when presenting the description and experience information separately or simultaneously, and how the two kinds of information affect the risk decision making. A separate presentation description and empirical information was studied and the latter was presented using the sampling paradigm used by Hertwig et al. The results show that the risk decision results based on descriptive information are in line with the expected theory. The result of risk decision based on empirical information is different from that of the former. In the case of medium and high probability benefit scenario, the participants are more adventurous in making decisions based on empirical information than in describing information. This phenomenon has been verified at both group and individual levels. In study 2, the experimental group received two types of information, while the control group received only one type of information. The results showed that there was no significant difference between the results of risk decision in the experimental group and those in the group receiving only empirical information, but there was a significant difference between the results of risk decision and that of the group receiving only descriptive information, and the results of the risk decision of the subjects were more affected by the information of experience. The above conclusions are verified by subjective estimation probability and the weight of empirical information reported after decision making. In the third study, the empirical information was presented by a more real pumping experiment. The results showed that the risk decision and subjective probability estimation of the subjects were more affected by the empirical information when the two types of information were accepted simultaneously. In addition, the third study also explores whether the degree of confidence in describing information mediates the influence of the deviation degree of the two types of information providing probability on risk decision-making. The results show that only when the corresponding probability of empirical information is higher than that of descriptive information, it plays an intermediary role in the degree of confidence in descriptive information, that is, the greater the degree of empirical probability is higher than the degree of description probability, the lower the trust degree of descriptive information is. Thus prefer to choose the risk option. The above conclusions provide new evidence for the descriptive-empirical gap and verify that when two types of information are presented at the same time, the empirical information has a greater impact on people's risk decision and subjective probability estimation, and the mechanism is preliminarily explored. These conclusions suggest that we should pay attention to the role of empirical information in risk decision-making.
【学位授予单位】:浙江大学
【学位级别】:硕士
【学位授予年份】:2017
【分类号】:B848
【参考文献】
相关期刊论文 前10条
1 刘建民;李海垒;张文新;;行为决策中的描述 经验差距:信息加工角度的解释[J];心理科学进展;2016年10期
2 祝婧媛;何贵兵;;风险来源与决策:背信规避现象及人际联结需求的作用[J];心理学报;2016年06期
3 江程铭;李纾;;中介分析和自举(Bootstrap)程序应用[J];心理学探新;2015年05期
4 李爱梅;颜亮;王笑天;马学谦;李方君;;时间压力的双刃效应及其作用机制[J];心理科学进展;2015年09期
5 潘禄;钱秀莹;;动态重复决策中先前结果对风险偏好的影响[J];心理学报;2014年12期
6 张阳阳;饶俪琳;梁竹苑;周媛;李纾;;风险决策过程验证:补偿/非补偿模型之争的新认识与新证据[J];心理科学进展;2014年02期
7 刘翠翠;陈彬;刘磊鑫;原献学;汪祚军;;当局者迷,旁观者清? 自我-他人决策的理性差异及其机制[J];心理科学进展;2013年05期
8 赵秋荻;刘永芳;段婧;徐沙;;心理距离与决策者角色对风险决策的影响[J];应用心理学;2013年01期
9 王璐璐;李永娟;;心理疲劳与任务框架对风险决策的影响[J];心理科学进展;2012年10期
10 刘腾飞;徐富明;马红宇;马向阳;吴修良;;行为决策研究的新取向——基于经验的决策[J];心理科学进展;2012年07期
相关硕士学位论文 前1条
1 陈凡;在描述范式、经验范式下对风险决策的比较研究[D];复旦大学;2013年
,本文编号:2234056
本文链接:https://www.wllwen.com/shekelunwen/xinlixingwei/2234056.html