信息样本的有偏采集如何导致决策偏见?——基于采样偏差的新视角述评

How Does Biased Information Sampling Lead to Biases in Judgment: A Review from a Biased Sampling Perspective

马丹丹,岑咏华,吴承尧

2017 第 39 卷 第 12 期 页码 23-37

DOI: 10.16538/j.cnki.fem.2017.12.002

PDF全文下载 HTML全文

摘要

判断和决策中的许多偏见现象都源自于决策者无意识或者有意识地将有偏差的不完全观测样本当成有代表性的整体。近些年以Fiedler、Juslin、Hertwig、Denrell、Le Mens等为代表的学者重点关注了这一信息样本的有偏采集问题,为研究决策偏见的形成机理提供了新的采样偏差视角。本文对这一视角进行了系统性述评,厘清了“采样”的概念内涵,阐述了采样偏差导致决策偏见的基本机制,并对采样偏差引起的一种典型偏见效应——“描述—经验”差异进行了重点剖析,最后归纳了几个融入采样偏差思想的重要决策模型。作为结论,本文指出,决策偏见研究需要关注信息采样问题,尤其关注决策信息环境的哪些特征更易导致个体对信息样本的有偏采集。本文研究丰富了决策偏见的理论视角,对于洞悉和改进人类有限理性决策具有一定的实践指导意义。

Individual judgment often systematically deviates from the optimal beliefs and choices hypothesized by rational-agent models, leading to cognitive illusions and biases in judgment. Researchers have observed an enormous number of biased beliefs and behaviors in laboratory and field studies since 1970s, such as confirmatory bias, overconfidence, false correlation, base-rate neglect, herd behavior, etc. This biased way of judgment reveals the limitations of rational-agent models and has been an important concern in management & decision science, behavioral economics, psychology, and sociology. Pertaining studies concerning biased judgment in psychology attempt to model and understand the nature of these biases by un-riddling the mental processes of judgment, resulting in various influential perspectives. In particular, individuals’ internal motivational constraints and cognitive deficits are widely concerned by researchers and are regarded as the main mechanisms for the formation of biased judgment. In terms of motivational perspective, researchers attribute selective exposure, information distortion and confirmatory bias to the motivation of cognitive coherence, and ascribe biases in social judgments, like obedience and herd behavior, to the motivation of affiliation. In terms of cognitive perspective, researchers attribute biases in judgment to heuristic search and attribution substitution induced by individuals’ limited cognitive resources in the process of information processing. Previous studies have shed substantial light on individuals’ motivation and cognition, but little on individuals’ interactions with the environment and the biasing impact of these interactions. However, as claimed by Simon, bounded rationality is the result of interplay of the mind and environment. In order to understand human behavior, one first has to understand the environment in which the behaviors take place. In 2000, Fiedler introduced the concept of " sampling” in an important paper published in Psychological Review, emphasizing the impact of distribution of information in the environment on the acquisition process of decision samples(information sampling)and judgment. Many biases in judgment can be attributed to biased sampling, namely decision-makers consciously or unconsciously take the biased and incomplete information samples they draw from the population as the representation of the environment. As the environment component of Simon’s bounded rationality, the biased sampling perspective plays a constructive role in providing alternative accounts for the existing decision biases. In the last decade, the biased sampling perspective has received much attention. Inspired by the perspective of information sampling, researchers discover a new bias effect " description-experience gap”. In addition, the idea of information sampling is integrated into the decision-making models in management and economics, e.g. query theory, decision field theory, leaky-competing accumulator model and decision by sampling. However, although a variety of researchers have attempted to explore the formation mechanism of judgment from the perspective of biased sampling, relatively little attention has been given to this perspective in China. In response to this situation, in current research, an alternative account for the mechanisms underlying biased judgment is presented, which highlights individuals’ information sampling from the environment in the spirit of the biased sampling approach launched by the influential figures like Fiedler, Juslin, Denrell, Le Mens, Hertwig, etc. This paper proceeds as follows: by introducing the conception of sampling proposed by Fiedler, the specific psychological and environmental mechanisms are induced, and resulting biases in judgment are elaborated. Then, the description-experience gap, a typical bias resulting from biased information sampling, is comprehensively delineated, followed by several enlightening sampling-based decision models. Based on a literature review, this paper appeals to the academia for future scholarly endeavors from the perspective of biased information sampling, especially the investigation of the characteristics of the environment which are more likely to lead to biased information sampling. It adds to the theoretical foundations of biased judgment, and provides practical implications for understanding and correcting the bounded rationality of human decision-making.

参考文献

1 李艾丽莎, 张庆林. 决策的选择偏好研究述评[J]. 心理科学进展, 2006, (4): 618–624.
2 刘腾飞, 徐富明, 马红宇, 等. 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, (7): 1068–1079.
5 Appelt K C, Hardisty D J, Weber E U. Asymmetric discounting of gains and losses: A query theory account[J]. Journal of Risk and Uncertainty, 2011, 43(2): 107–126. DOI:10.1007/s11166-011-9125-1
6 Asch S E. Opinions and social pressure[J]. Scientific American, 1955, 193(5): 31–35. DOI:10.1038/scientificamerican1155-31
7 Barron G, Ursino G. Underweighting rare events in experience based decisions: Beyond sample error[J]. Journal of Economic Psychology, 2013, 39: 278–286. DOI:10.1016/j.joep.2013.09.002
10 Brown G D A, Gardner J, Oswald A J, et al. Does wage rank affect employees’ well-being?[J]. Industrial Relations: A Journal of Economy and Society, 2008, 47(3): 355–389. DOI:10.1111/j.1468-232X.2008.00525.x
11 Busemeyer J R, Johnson J G. Micro-process models of decision making[A]. Sun R. The Cambridge handbook of computational psychology[M]. Cambridge, MA: Cambridge University Press, 2008: 302–321.
13 Camilleri A R, Newell B R. The long and short of it: Closing the description-experience " gap” by taking the long-run view[J]. Cognition, 2013, 126(1): 54–71. DOI:10.1016/j.cognition.2012.09.001
14 Chaiken S, Liberman A, Eagly A H. Heuristic and systematic information processing within and beyond the persuasion context[A]. Uleman J S, Bargh J A. Unintended thought[M]. New York: Guilford, 1989: 212–252.
15 Chambers J R, Windschitl P D. Biases in social comparative judgments: the role of nonmotivated factors in above-average and comparative-optimism effects[J]. Psychological Bulletin, 2004, 130(5): 813–838. DOI:10.1037/0033-2909.130.5.813
16 Cowan N. Metatheory of storage capacity limits[J]. Behavioral and Brain Sciences, 2001, 24(1): 154–176.
17 Davidai S, Gilovich T, Ross L D. The meaning of default options for potential organ donors[J]. Proceedings of the National Academy of Sciences of the United States of America, 2012, 109(38): 15201–15205. DOI:10.1073/pnas.1211695109
18 Denrell J. Why most people disapprove of me: Experience sampling in impression formation[J]. Psychological Review, 2005, 112(4): 951–978. DOI:10.1037/0033-295X.112.4.951
19 Denrell J, Le Mens G. Information sampling, conformity and collective mistaken beliefs[A]. Proceedings of the 35th annual conference of the cognitive science society[C]. Austin, TX: Cognitive Science Society, 2013: 2177–2182.
20 Denrell J, Le Mens G. Information sampling, belief synchronization, and collective illusions[J]. Management Science, 2016, 63(2): 528–547.
23 Ditto P H, Scepansky J A, Munro G D, et al. Motivated sensitivity to preference-inconsistent information[J]. Journal of Personality and Social Psychology, 1998, 75(1): 53–69. DOI:10.1037/0022-3514.75.1.53
24 Dufau S, Grainger J, Ziegler J C. How to say " no” to a nonword: A leaky competing accumulator model of lexical decision[J]. Journal of Experimental Psychology: Learning, Memory, and Cognition, 2012, 38(4): 1117–1128. DOI:10.1037/a0026948
25 Feiler D C, Tong J D, Larrick R P. Biased judgment in censored environments[J]. Management Science, 2013, 59(3): 573–591. DOI:10.1287/mnsc.1120.1612
26 Fiedler K. Beware of samples! A cognitive-ecological sampling approach to judgment biases[J]. Psychological Review, 2000, 107(4): 659–676. DOI:10.1037/0033-295X.107.4.659
27 Fiedler K. Information ecology and the explanation of social cognition and behavior[A]. Kruglanski A, Higgins E T. Social psychology: Handbook of basic principles[M]. 2nd ed. New York: Guilford, 2007: 176–200.
28 Fiedler K, Juslin P. Taking the interface between mind and environment seriously[A]. Fiedler K, Juslin P. Information sampling and adaptive cognition[M]. New York: Cambridge University Press, 2006: 1–29.
29 Fox C R, Hadar L. " Decisions from experience”=sampling error+prospect theory: Reconsidering Hertwig, Barron, Weber & Erev (2004)[J]. Judgment and Decision Making, 2006, 1(2): 159–161.
31 Glöckner A, Hilbig B E, Henninger F, et al. The reversed description-experience gap: Disentangling sources of presentation format effects in risky choice[J]. Journal of Experimental Psychology: General, 2016, 145(4): 486–508. DOI:10.1037/a0040103
32 Goldstein D G, Johnson E J, Herrmann A, et al. Nudge your customers toward better choices[J]. Harvard Business Review, 2008, 86(12): 99–105.
33 Hardisty D J, Johnson E J, Weber E U. A dirty word or a dirty world? Attribute framing, political affiliation, and query theory[J]. Psychological Science, 2010, 21(1): 86–92. DOI:10.1177/0956797609355572
34 Hau R, Pleskac T J, Kiefer J, et al. The description-experience gap in risky choice: The role of sample size and experienced probabilities[J]. Journal of Behavioral Decision Making, 2008, 21(5): 493–518. DOI:10.1002/bdm.v21:5
35 Hau R, Pleskac T J, Hertwig R. Decisions from experience and statistical probabilities: Why they trigger different choices than a priori probabilities[J]. Journal of Behavioral Decision Making, 2010, 23(1): 48–68. DOI:10.1002/bdm.v23:1
38 Hertwig R, Pleskac T J. The game of life: How small samples render choice simpler[A]. Chater N, Oaksford M. The probabilistic mind: Prospects for Bayesian cognitive science[M]. Oxford: Oxford University Press, 2008: 209–236.
39 Hertwig R, Pleskac T J. Decisions from experience: Why small samples?[J]. Cognition, 2010, 115(2): 225–237. DOI:10.1016/j.cognition.2009.12.009
40 Hills T T, Hertwig R. Information search in decisions from experience: Do our patterns of sampling foreshadow our decisions?[J]. Psychological Science, 2010, 21(12): 1787–1792. DOI:10.1177/0956797610387443
42 Johnson E J, Goldstein D. Do defaults save lives?[J]. Science, 2003, 302(5649): 1338–1339. DOI:10.1126/science.1091721
43 Johnson E J, Häubl G, Keinan A. Aspects of endowment: A query theory of value construction[J]. Journal of Experimental Psychology: Learning, Memory, and Cognition, 2007, 33(3): 461–474. DOI:10.1037/0278-7393.33.3.461
46 Kahneman D, Frederick S. Representativeness revisited: Attribute substitution in intuitive judgment[A]. Gilovich T, Griffin D, Kahneman D. Heuristics and biases: The psychology of intuitive judgment[M]. New York: Cambridge University Press, 2002: 49–81.
48 Kareev Y. Seven(indeed, plus or minus two)and the detection of correlations[J]. Psychological Review, 2000, 107(2): 397–402. DOI:10.1037/0033-295X.107.2.397
49 Kareev Y. Good sampling, distorted views[A]. Fielder K, Juslin P. Information sampling and adaptive cognition[M]. New York: Cambridge University Press, 2006: 33–52.
50 Khemlani S S, Oppenheimer D M. When one model casts doubt on another: A levels-of-analysis approach to causal discounting[J]. Psychological Bulletin, 2011, 137(2): 195–210. DOI:10.1037/a0021809
51 Koehler J J, Mercer M. Selection neglect in mutual fund advertisements[J]. Management Science, 2009, 55(7): 1107–1121. DOI:10.1287/mnsc.1090.1013
54 Lejarraga T, Hertwig R, Gonzalez C. How choice ecology influences search in decisions from experience[J]. Cognition, 2012, 124(3): 334–342. DOI:10.1016/j.cognition.2012.06.002
60 Olivola C Y, Sagara N. Distributions of observed death tolls govern sensitivity to human fatalities[J]. Proceedings of the National Academy of Sciences of the United States of America, 2009, 106(52): 22151–22156. DOI:10.1073/pnas.0908980106
61 Oppenheimer D M, Kelso E. Information processing as a paradigm for decision making[J]. Annual Review of Psychology, 2015, 66: 277–294. DOI:10.1146/annurev-psych-010814-015148
63 Petty R E, Cacioppo J T. Communication and persuasion: Central and peripheral routes to attitude change[M]. New York: Springer-Verlag, 1986.
64 Polman E, Russo J E. Commitment to a developing preference and predecisional distortion of information[J]. Organizational Behavior and Human Decision Processes, 2012, 119(1): 78–88. DOI:10.1016/j.obhdp.2012.05.004
65 Proeger T, Meub L. Overconfidence as a social bias: Experimental evidence[J]. Economics Letters, 2014, 122(2): 203–207. DOI:10.1016/j.econlet.2013.11.027
66 Rakow T, Demes K A, Newell B R. Biased samples not mode of presentation: Re-examining the apparent underweighting of rare events in experience-based choice[J]. Organizational Behavior and Human Decision Processes, 2008, 106(2): 168–179. DOI:10.1016/j.obhdp.2008.02.001
67 Rakow T, Newell B R. Degrees of uncertainty: An overview and framework for future research on experience-based choice[J]. Journal of Behavioral Decision Making, 2010, 23(1): 1–14. DOI:10.1002/bdm.v23:1
69 Russo J E, Carlson K A, Meloy M G, et al. The goal of consistency as a cause of information distortion[J]. Journal of Experimental Psychology: General, 2008, 137(3): 456–470. DOI:10.1037/a0012786
72 Simon H A. Models of bounded rationality: Empirically grounded economic reason[M]. Cambridge: MIT Press, 1982.
73 Simonson I. Choice based on reasons: The case of attraction and compromise effects[J]. Journal of Consumer Research, 1989, 16(2): 158–174. DOI:10.1086/jcr.1989.16.issue-2
74 Stewart N, Chater N, Brown G D A. Decision by sampling[J]. Cognitive Psychology, 2006, 53(1): 1–26. DOI:10.1016/j.cogpsych.2005.10.003
76 Thaler R H, Sunstein C R. Nudge: Improving decisions about health, wealth, and happiness[M]. New Haven, CT: Yale University Press, 2008.
78 Trueblood J S, Brown S D, Heathcote A. The multiattribute linear ballistic accumulator model of context effects in multialternative choice[J]. Psychological Review, 2014, 121(2): 179–205. DOI:10.1037/a0036137
79 Tsetsos K, Gao J, McClelland J L, et al. Using time-varying evidence to test models of decision dynamics: Bounded diffusion vs. the leaky competing accumulator model[J]. Frontiers in Neuroscience, 2012, 6: 79.
80 Tversky A. Elimination by aspects: A theory of choice[J]. Psychological Review, 1972, 79(4): 281–299. DOI:10.1037/h0032955
81 Tversky A. Features of similarity[J]. Psychological Review, 1977, 84(4): 327–352. DOI:10.1037/0033-295X.84.4.327
82 Tversky A, Kahneman D. Judgment under uncertainty: Heuristics and biases[J]. Science, 1974, 185(4157): 1124–1131. DOI:10.1126/science.185.4157.1124
87 Vul E, Goodman N, Griffiths T L, et al. One and done? Optimal decisions from very few samples[J]. Cognitive Science, 2014, 38(4): 599–637. DOI:10.1111/cogs.12101
88 Wang T, Wang D S. Why Amazon’s ratings might mislead you: The story of herding effects[J]. Big Data, 2014, 2(4): 196–204. DOI:10.1089/big.2014.0063
91 Wulff D U, Hills T T, Hertwig R. Online product reviews and the description-experience gap[J]. Journal of Behavioral Decision Making, 2015, 28(3): 214–223. DOI:10.1002/bdm.v28.3