Content uploaded by Baruch Fischhoff
Author content
All content in this area was uploaded by Baruch Fischhoff on Dec 23, 2013
Content may be subject to copyright.
Hindsight is not equal to foresight: The effect of outcome knowledge on judgment under uncertainty
Abstract
Notes that a major difference between historical and nonhistorical judgment is that the historical judge typically knows how things turned out. 3 experiments are described with a total of 479 college students. In Exp I, receipt of such outcome knowledge was found to increase the postdicted likelihood of reported events and change the perceived relevance of event-descriptive data, regardless of the likelihood of the outcome and the truth of the report. Ss were, however, largely unaware of the effect that outcome knowledge had on their perceptions. As a result, they overestimated what they would have known without outcome knowledge (Exp II), as well as what others (Exp III) actually did know without outcome knowledge. It is argued that this lack of awareness can seriously restrict one's ability to judge or learn from the past. (16 ref)
... First, when asked to recall their original estimates, people's responses tend to be closer to the actual value than their original estimates were. This well-established phenomenon is known as hindsight bias (e.g., Christensen-Szalanski & Willham, 1991;Groß & Pachur, 2019;Hawkins & Hastie, 1990), and was initially introduced by Fischhoff (1975). It has been shown for estimates of realworld quantities (e.g., lengths of rivers, heights of towers; see Bayen et al., 2006;Bernstein et al., 2011;Erdfelder & Buchner, 1998;Groß & Bayen, 2015;Pohl & Hell, 1996), but also for two-alternative-choice tasks, judgments of event outcomes (e.g., historical and medical), and other tasks and materials (see Pohl, 2007, for an overview). ...
... In this article, we focused on hindsight bias in the context of realworld quantitative estimation, where metric knowledge plays an important role for judgment performance and can be updated as a result of exposure to numerical facts. However, hindsight bias has also been observed in domains and tasks where metric knowledge plays no central role, for instance, when judging event probabilities (e.g., Fischhoff, 1975), or identifying visual stimuli (e.g., Giroux et al., 2022). Does this mean that knowledge updating can be expected to be less relevant in these other contexts? ...
... Historically, it has been argued that hindsight bias-reflecting a lack of awareness of one's limited prior knowledge-restricts people's ability to learn (Fischhoff, 1975). Our research shows that, on ...
When people estimate the quantities of objects (e.g., country populations), are then presented with the objects' actual quantities, and subsequently asked to remember their initial estimates, responses are often distorted towards the actual quantities. This hindsight bias — traditionally considered to reflect a cognitive error — has more recently been proposed to result from adaptive knowledge updating. But how to conceptualize such knowledge-updating processes and their potentially beneficial consequences? Here, we provide a framework that conceptualizes knowledge updating in the context of hindsight bias in real-world estimation by connecting it with research on seeding effects — improvements in people's estimation accuracy after exposure to numerical facts. This integrative perspective highlights a previously neglected facet of knowledge updating, namely, recalibration of metric domain knowledge, which can be expected to lead to transfer learning and thus improve estimation for objects from a domain more generally. We develop an experimental paradigm to investigate the association of hindsight bias with improved estimation accuracy. In Experiment 1, we demonstrate that the classical approach to induce hindsight bias indeed produces transfer learning. In Experiment 2, we provide evidence for the novel prediction that hindsight bias can be triggered via transfer learning; this establishes a direct link from knowledge updating to hindsight bias. Our work integrates two prominent but previously unconnected research programs on the effects of knowledge updating in real-world estimation and supports the notion that hindsight bias is driven by adaptive learning processes.
... Investigations of outcome bias (building on Fischhoff, 1975) cover a wide range of fields including not only financial, but also ethical, judicial or medical decisions (e.g., Arkes et al., 1981;Kamin & Rachlinski, 1995;Gino et al., 2010). Hershey, 1988, 1992) were the first to conceptualize the effects of outcomes on evaluations. ...
In a sequence of experiments, this study investigates how people evaluate others who make risky decisions on their behalf, and how such evaluations affect delegated risk-taking. A decision maker acts on behalf of a client who holds the decision maker accountable by way of a subjective evaluation after observing a risky decision’s outcome. If evaluation is biased towards the outcome, it may have dysfunctional effects with respect to delegated risk-taking in that decision makers’ risk choices are increasingly misaligned with their clients’ risk preferences. We find evidence giving support to this conjecture. Across and within three experiments, we test for the effects of different types and degrees of accountability in that we manipulate the information available to clients as well as the consequences which evaluations have for decision makers. Evaluations are biased towards outcomes in all experiments. When evaluations affect decision maker’s compensations, a stronger outcome bias in evaluations translates into risk-taking decisions being less frequently aligned with clients’ risk preferences. In the same situation, giving clients the opportunity to make peer comparisons increases outcome bias. We further find that clients do not hold decision makers accountable for their risk choices when they cannot observe the risk-taking decision, but have to infer it from observing the outcome. Theoretical and practical implications of the results are discussed.
... This can be problematic because humans show cognitive biases that can lead to erroneous inferences (Tversky & Kahneman, 1974). For example, humans see coherent patterns in randomness (Brugger, 2001), convince themselves of the validity of prior expectations ("I knew it"; Nickerson, 1998), and perceive events as being plausible in hindsight ("I knew it all along"; Fischhoff, 1975). In conjunction with an academic incentive system that rewards certain discovery processes more than others (Koole & Lakens, 2012;Sterling, 1959), we often find ourselves exploring many possible analytic pipelines but reporting only a selected few. ...
Recent empirical studies have highlighted the large degree of analytic flexibility in data analysis that can lead to substantially different conclusions based on the same data set. Thus, researchers have expressed their concerns that these researcher degrees of freedom might facilitate bias and can lead to claims that do not stand the test of time. Even greater flexibility is to be expected in fields in which the primary data lend themselves to a variety of possible operationalizations. The multidimensional, temporally extended nature of speech constitutes an ideal testing ground for assessing the variability in analytic approaches, which derives not only from aspects of statistical modeling but also from decisions regarding the quantification of the measured behavior. In this study, we gave the same speech-production data set to 46 teams of researchers and asked them to answer the same research question, resulting in substantial variability in reported effect sizes and their interpretation. Using Bayesian meta-analytic tools, we further found little to no evidence that the observed variability can be explained by analysts’ prior beliefs, expertise, or the perceived quality of their analyses. In light of this idiosyncratic variability, we recommend that researchers more transparently share details of their analysis, strengthen the link between theoretical construct and quantitative system, and calibrate their (un)certainty in their conclusions.
... The central hypothesis to this research question being the existence of irrational exuberance on what could be the present and future potentials of AI and robotics. This hypothesis was motivated by documented biases in judgement and predictions in the presence of uncertainty (Tversky and Kahneman (1974), Fischhoff (1975), Kahneman et al. (1982 and Kahneman (2003)) as it is the case with innovations, by a somewhat dense literature on irrational exuberance in the domain of financial assets (Shiller (2015), Tetlock (2007) and finally, to some extend, the Tesla's Model 3 case. I tapped into a deep archive from the The New York Times, to extract a set of articles covering those labor-substituting technologies to build a dataset that constituted my corpus. ...
This paper investigates how the news industry perceives the potentials of AI and robotics technologies. From a well defined set of articles in The New York Times archive, I constructed an index of the news industry tone on AI and robotics potentials, from 1980 to 2020. Firstly, the index displays a decreasing trend overtime, suggesting the existence of irrational exuberance in the valuation of labor-substituting technologies' relative potentials. Secondly, for comparison purposes, I found that a similar tone index for the microprocessor technology displays an upwards trend during the same period. This points to an upwards updating of the valuation of the potentials of the microprocessor discovery in contrary to AI and robotics technologies. Finally, I found that the constructed news tone index on AI and robotics is correlated with the investment flow from venture capital going to AI and robotics projects in the US.
A growing part of the literature on crises, disasters, and policy failures focuses on the design, conduct, and impact of postincident reviews or inquiries, particularly whether the right lessons are identified and subsequently learned. However, such accounts underappreciate the specific challenge posed by epistemic puzzles, under what conditions their difficulty may vary, and which strategies could help to solve them. Drawing on insights from a wide range of cases, the article identifies hindsight bias, counterfactual reasoning, and root‐cause analysis as core components creating an epistemic triangle of inquiry puzzling. It advances four propositions about the conditions that help or hinder investigators' capacity to produce sound knowledge and concludes by setting out potential strategies that investigators can use to fully address or at least mitigate these epistemic challenges.
When people learn that an event has occurred, they often report that they expected it all along. The present study explored this “hindsight bias” for the outcome of a national referendum on abortion. Participants rated the probability of the referendum passing one week before polling day, and then recalled their original estimates one week and one year post‐referendum. The magnitude and direction of bias was predicted by voting choice: People on the losing side exhibited a classic hindsight bias, and remembered believing the outcome to be roughly 16% more likely than they originally reported. In contrast, people on the winning side exhibited a weak reversal of the typical bias. Hindsight bias was moderated by feelings of surprise in both groups. This study suggests that individuals on the losing side of political campaigns may be especially likely to misremember their prior expectations if the outcome is disappointing but not unexpected.
During rapid eye movement (REM) sleep, newly consolidated memories can be distorted to adjust the existing memory base in memory integration. However, only a few studies have demonstrated the role of REM sleep in memory distortion. The present study aims to clarify the role of REM sleep in the facilitation of memory distortion, that is, hindsight bias, compared to non-rapid eye movement (NREM) sleep and wake states. The split-night paradigm was used to segregate REM and NREM sleep. The hypotheses are (1) hindsight bias-memory distortion-is more substantial during REM-rich sleep (late-night sleep) than during NREM-rich sleep (early-night sleep); (2) memory stabilization is more substantial during NREM-rich sleep (early-night sleep) than during REM-rich sleep (late-night sleep); and (3) memory distortion takes longer time than memory stabilization. The results of the hindsight bias test show that more memory distortions were observed after the REM condition in comparison to the NREM condition. Contrary to the hindsight bias, the correct response in the word-pair association test was observed more in the NREM than in the REM condition. The difference in the hindsight bias index between the REM and NREM conditions was identified only one week later. Comparatively, the difference in correct responses in the word-pair association task between the conditions appeared three hours later and one week later. The present study found that (1) memory distortion occurs more during REM-rich sleep than during NREM-rich sleep, while memory stabilization occurs more during NREM-rich sleep than during REM-rich sleep. Moreover, (2) the newly encoded memory could be stabilized immediately after encoding, but memory distortion occurs over several days. These results suggest that the roles of NREM and REM sleep in memory processes could be different.
The institution of liability serves to mitigate the lack of care in almost all areas, whether private or business. However, we have not yet found such an institution in political decision-making. Surprisingly, the literature has not discussed a specific institution that subjects political actors who fail to exercise due diligence in their decision-making regarding personal liability. Hence, this paper aims to fill this gap and derive the necessity of internalizing the negative effects resulting from the imperfections of the market for political services in general and the democratic process, particularly by a liability rule. To design the new institution, we draw on the findings of corporate governance, combining economic thinking in incentives and legal knowledge expressed in the law of the corporation. In this respect, this paper is the first to make a concrete proposal for political liability accompanied by a political judgment rule. However, it is important to emphasize that the aim is not to punish a wrong decision but to provide strong incentives to prevent it ex ante. Political liability must be understood as a process-oriented institution that considers uncertainty and decision-making complexities. By proposing and analyzing this new institution, this work contributes to a broader discussion of incentive structures in the political process of modern democracies and shows how the political sphere can learn from the corporate world.
This paper investigates the time-trend of the capital-labor elasticity of substitution into the US non-manufacturing sectors. The central question here is whether the effects of labor-substituting technologies on the easiness to substitute labor for capital over time is different across non-manufacturing sectors. Using the data mainly from the Bureau of Economic Analysis (BEA) for the period going from 1955 to 2015, I found that the capital-labor elasticity of substitution does have a heterogeneous long-run time trend across US non-manufacturing sectors. This result, with the existing evidence on the sectoral heterogeneity of the effects of labor-substituting innovations suggests that sectors with the highest levels of penetration of automation technologies have experienced an increase in their easiness to substitute labor for capital whereas those with a high level of new created tasks where humans have comparative advantage, have experienced a decrease in their easiness to substitute labor for capital. With regards to the effects of IPPs capitalization, two results emerged: IPPs capitalization does not reverse the trend in the labor share of value-added in any sector while leaving the nature and the trend of the easiness to substitute labor for capital for the majority of sectors unaltered.
This article described three heuristics that are employed in making judgements under uncertainty: (i) representativeness, which is usually employed when people are asked to judge the probability that an object or event A belongs to class or process B; (ii) availability of instances or scenarios, which is often employed when people are asked to assess the frequency of a class or the plausibility of a particular development; and (iii) adjustment from an anchor, which is usually employed in numerical prediction when a relevant value is available. These heuristics are highly economical and usually effective, but they lead to systematic and predictable errors. A better understanding of these heuristics and of the biases to which they lead could improve judgements and decisions in situations of uncertainty.
This article described three heuristics that are employed in making judgements under uncertainty: (i) representativeness,
which is usually employed when people are asked to judge the probability that an object or event A belongs to class or process
B; (ii) availability of instances or scenarios, which is often employed when people are asked to assess the frequency of a
class or the plausibility of a particular development; and (iii) adjustment from an anchor, which is usually employed in numerical
prediction when a relevant value is available. These heuristics are highly economical and usually effective, but they lead
to systematic and predictable errors. A better understanding of these heuristics and of the biases to which they lead could
improve judgements and decisions in situations of uncertainty.
It is clear that we cannot distinguish the sane from the insane in psychiatric hospitals. The hospital itself imposes a special environment in which the meanings of behavior can easily be misunderstood. The consequences to patients hospitalized in such an environment-the powerlessness, depersonalization, segregation, mortification, and self-labeling-seem undoubtedly countertherapeutic. I do not, even now, understand this problem well enough to perceive solutions. But two matters seem to have some promise. The first concerns the proliferation of community mental health facilities, of crisis intervention centers, of the human potential movement, and of behavior therapies that, for all of their own problems, tend to avoid psychiatric labels, to focus on specific problems and behaviors, and to retain the individual in a relatively non-pejorative environment. Clearly, to the extent that we refrain from sending the distressed to insane places, our impressions of them are less likely to be distorted. (The risk of distorted perceptions, it seems to me, is always present, since we are much more sensitive to an individual's behaviors and verbalizations than we are to the subtle contextual stimuli that often promote them. At issue here is a matter of magnitude. And, as I have shown, the magnitude of distortion is exceedingly high in the extreme context that is a psychiatric hospital.) The second matter that might prove promising speaks to the need to increase the sensitivity of mental health workers and researchers to the Catch 22 position of psychiatric patients. Simply reading materials in this area will be of help to some such workers and researchers. For others, directly experiencing the impact of psychiatric hospitalization will be of enormous use. Clearly, further research into the social psychology of such total institutions will both facilitate treatment and deepen understanding. I and the other pseudopatients in the psychiatric setting had distinctly negative reactions. We do not pretend to describe the subjective experiences of true patients. Theirs may be different from ours, particularly with the passage of time and the necessary process of adaptation to one's environment. But we can and do speak to the relatively more objective indices of treatment within the hospital. It could be a mistake, and a very unfortunate one, to consider that what happened to us derived from malice or stupidity on the part of the staff. Quite the contrary, our overwhelming impression of them was of people who really cared, who were committed and who were uncommonly intelligent. Where they failed, as they sometimes did painfully, it would be more accurate to attribute those failures to the environment in which they, too, found themselves than to personal callousness. Their perceptions and behavior were controlled by the situation, rather than being motivated by a malicious disposition. In a more benign environment, one that was less attached to global diagnosis, their behaviors and judgments might have been more benign and effective.