What is the hindsight bias What does it show us about our reliance on common sense?


During the White House’s daily coronavirus task force briefing on March 15, 2020, the NBC News reporter Peter Alexander cited the latest pandemic statistics and asked President Trump, “What do you say to Americans who are scared?” at the end of the exchange. The President replied, “…you are a terrible reporter!” and continued his rant by saying the question was “nasty” and that Alexander was doing “sensationalism” like NBC and Comcast, which he later referred to as “Con-Cast” (Darcy, 2020).

Afterward, many critics were appalled that the President did not address Americans’ fear for COVID-19 and failure to respond to others’ misfortune. This exchange between the President and Alexander highlighted the importance of perspective-taking, which is fundamental for social interaction. Perspective-taking can be critical in predicting and interpreting human behavior; it determines how we evaluate and describe others, as displayed in the exchange between the President and Alexander. Would the exchange have turned out differently if either of them was considerate of each other? Would perspective-taking have enhanced their ability to understand each other’s point of view?

Perspective-taking can function as a “curse.” Once we acquire knowledge, it hampers our judgment about what we knew beforehand or to adopt another person’s naïve perspective because the knowledge seems so obvious to us (Birch et al., 2017; Pezzo, 2011). This effect is readily seen in hindsight bias research related to outcome knowledge. The hindsight bias functions to influence our judgments about what we or others could have, should have, or knew about an outcome or before the outcome was known.

Hindsight Bias

Hindsight bias has garnered broad research interest such that in 2012, there were already over 800 scholarly works on the topic (Roese & Vohs, 2012). Not only is it basic and universal (Ghrear et al., 2019; Pohl et al., 2002), there exist developmental differences (Bernstein et al., 2011; Groß & Pachur, 2019). This bias has been documented in diverse domains such as terrorist attacks (Fischhoff et al., 2005), medical judgments (Arkes et al., 1988), athletic competitions (Roese & Maniar, 1997), legal profession for negligence and medical malpractice decisions (Giroux et al., 2016; Harley, 2007), problem-solving (Hoch & Loewenstein, 1989), and in judges and experts (Giroux et al., 2016; Wilson et al., 2020). Lately, researchers have examined the effect in evaluating the decisions of others like company directors (Strohmaier et al., 2021), gamer competition (Danz, 2020), pilots (Walmsley and Gilbey (2019), abduction (McDermott et al., 2020), financial forecasting (Chuan et al., 2019), rape victims (Felson and Palmore (2018) and cheaters (Stanley et al., 2021). Hindsight bias is seen in the mass media in Wikipedia articles (Oeberst et al., 2018) and exhibited in the way you write (Meuer et al., 2021).The phenomenon is robust and so prevalent that many investigators (e.g., Harley et al., 2004) believe that it is present in “all modalities” as shown in forms like taste (Pohl et al., 2003), faces (Bernstein & Harley, 2007) and pitch variation (Bernstein et al., 2012; Higham et al., 2017).

Typically, this bias effect is examined via the hypothetical or memory design. In the former, the difference is between different groups of individuals, one in hindsight (non-naïve) and the other in foresight (naïve). In the memory design, the judgments are made by the same individuals in foresight (naïve) and then after learning the outcome (non-naïve). In both instances, individuals in hindsight overestimate their ability to predict the outcome knowledge from the foresight perspective.

Illustrative Research on Perspective-Taking and Hindsight Bias

Central to the issue of perspective-taking is the notion of egocentrism. It is the belief that others share one’s own perspective (e.g., Epley et al., 2004; Nickerson, 1999). People tend to believe their behavior, knowledge, and beliefs are commonly shared (Mullen, 1983), readily accessible, and make an automatic inference about others (Epley et al., 2004). Also, this self-reliance is evidenced in the projection bias (Krueger & Clement, 1997) and the false consensus effect (Ross et al., 1977).

Perspective-taking in hindsight bias often follow semantic and perceptual domains. Regarding the former, Fischhoff (1975) revealed the effect using a series of independent experiments involving scenarios for a nineteenth-century battle or clinical cases. Inspired by his work, much hindsight bias research has employed case studies. In his research, participants overestimated the probability of the correct outcomes for themselves or their peers even when trying to ignore the correct outcomes in hindsight compared to foresight as they made their predictions. The implication is that people imagine their peers are much like themselves to reach the same conclusion. However, it is impossible to verify this inference because his participants did not make self or other predictions in the same experiment.

From the perceptual field is the use of faces (e.g., Harley et al., 2004), where the focus is on when participants can discriminate multiple distorted stimuli (e.g., celebrity faces). Initially, participants were presented with distorted stimulus ranging from a blurry to clear progression and asked to identify the stimulus. After identifying the stimulus, they were asked to repeat the blurry to the clear progression of the stimulus and indicate when they recognized the stimulus.

Participants showed hindsight bias by recognizing the stimulus at greater distortion in the second phase compared to the first phase of testing. This effect occurred when employing self-judgments for both testing phases or an initial self and then peer judgment. More evidence of shared perspective taking is seen in the auditory work of Bernstein et al. (2012). Employing the self-other procedure, participants in hindsight who clearly heard the targeted words or sentences misjudged naïve others’ ability to discriminate these stimuli when presented in degraded states. Because the self-self judgments or self-peer judgments occurred in different experiments (Harley et al., 2004) or the absence of an appropriate baseline comparison for the multiple judgments in a self-other comparison (Bernstein et al., 2012), it is not possible to reveal a self-peer distinction as to whether the degree of bias is the same or different for the self-peer comparison.

In a rare investigation of the self-peer distinction, Son et al. (2021) presented distorted pictures of objects and animals modeled after the work of Harley et al. (2004) to examine whether egocentrism could be altered to debias hindsight bias. In experiment 1, participants made self-self or self-peer judgments that resulted in the same outcome that supports the presence of egocentrism. In both instances, participants identified the stimuli at the same point of greater distortion when switched to the peer perspective or tested again from the self-perspective.

Regardless of the focus on the self or peer perspective, the expression of hindsight was the same. For experiment 2, participants made self-self judgments and then for 3rd and 5th graders. Again, hindsight bias was demonstrated by the participants in the self-self judgments portion of the study. Unlike experiment 1, perspective-taking for the younger targets showed a reverse effect. Participants thought the 3rd graders would take longer to recognize the objects with the same tendency for the 5th graders than themselves. While the authors focused on the debiasing of hindsight bias, their findings also suggest that egocentrism can be present or altered by the attributes of the targeted perspective-taker.

Not only do individuals believe that their perspective-taking is shared with others as reflected in the commonality of hindsight knowledge (Bernstein et al., 2018; Birch et al., 2017; Higham et al., 2017; Hoch & Loewenstein, 1989; Wood, 1978); they can overestimate it. Demonstrating the salience and influence of correct answers, Birch et al. (2017) found participants who knew the correct answers overestimated the percentage of peers naïve to the answers who would answer correctly. The difficulty with this research on shared perspective taking and commonality of outcome knowledge is that participants made singular judgments directed at others or consecutive judgments for self and then others. In either instance, it does not clearly reveal a self-other distinction showing the contribution of one’s own judgment in judging others.

There exist other situations implicating a self-other distinction resulting in individuals devaluing the view of others yet showed hindsight bias in doing so. Using the Clinton impeachment case as it unfolded, Bryant and Guilbault (2002) looked at perspective-taking involving judgments for oneself, followed by the average American and their best friend. Participants showed hindsight bias for acquittal for their self-predictions, but it was not present in their predictions for the average American or best friend. Similarly, Hom and Kaiser (2016) asked participants to make foreseeability and ethical judgments for an animal experiment for themselves and then a peer. Participants who knew the outcome claimed the experiment was more foreseeable and ethical and later judged the peer to be less foresightful and ethical than themselves.

Further research devaluing the views of another is seen in legal cases, particularly in medical malpractice cases (e.g., Harley, 2007). For example, LaBine and LaBine (1996) had participants review a hypothetical case in which a psychiatric patient became violent, or they did not know the outcome. Knowing the outcome led participants to perceive the therapist’s actions to be more foreseeable and less reasonable.

Other reactions altering the nature of egocentrism is present in situations involving self-relevant negative outcomes, namely ego-changing behavior that is defensive or reflecting retroactive pessimism. In the former, people will attempt to minimize their responsibility for the negative outcome as being unforeseeable in hindsight (Louie et al., 2000; Mark et al., 2003). If the negative outcome is foreseeable, then one is expected to act differently. Louie et al. (2000) had MBA students compete in a graded marketing study. Initially, participants made their predictions about a firm’s market performance from either a self or other perspective. Later, participants received a positive or negative outcome of the firm’s market performance. Louie et al. (2000) found that hindsight bias occurred when the participants’ own team succeeded but not when other teams succeeded. The bias was not observed when their team failed but was present in their judgments about the other teams’ failure. Similarly, Mark et al. (2003) using a stock trading game’s study found that students who received a negative outcome perceived it to be less foreseeable in competition with other students or neutral observers.

Instead of foreseeability, Tykocinski et al. (2002), in their retroactive pessimism approach argue that believing the outcome to be inevitable achieves the goal of reducing one’s culpability. In one experiment, participants could lose a large stipend for being late. Participants in hindsight viewed being late to be more inevitable than in foresight and this difference did not occur in their judgments for a friend.

In short, egocentrism and its role in perspective-taking can be subject to the attributes of the situation. The presence of situations involving blame and responsibility apparently trigger the alternation of shared perspective-taking in participants to include ego enhancing and protective mechanisms. What is lacking in this research literature is a clearer picture of a self-other distinction whereby one’s perspective-taking is shared or not shared. Past research aimed at singular judgments of others or consecutive judgments involving self and then others does not account fully for the contribution of one’s own perspective-taking in adopting the perspective of others. In the former, there is no assessment of one’s own judgment and in the latter, the multiple judgments confound an appropriate estimate of others’ judgement without an appropriate baseline comparison.

Rationale for the Present Study

In his review, Nickerson (1999) pointed out the importance of anchoring and adjustment (Tversky & Kahneman, 1974) in people’s egocentric judgments about others. Anchoring and adjustment effects can account for hindsight bias in self or other judgments using hypothetical or memory designs. Whether one is in the self or other’s perspective, the outcome information (hindsight) provides pertinent information to the person. People typically respond by adjusting their judgments in the direction of the outcome to sync with their current outcome knowledge. In contrast, individuals in foresight do not have obvious anchoring of the outcome information. In the latter, individuals make their original predictions about an outcome, anchor on the outcome’s receipt, and adjust their recall predictions in the outcome direction. This notion applies to situations involving self followed by judgments about others as well. Having people judge themselves and then others anchored one’s attention to their own knowledge in predicting for others. They can adjust their estimate based on the available information about the target or the situation itself.

In the absence of pertinent information, the person falls prey to seeing others much like themselves. Prior research (Tullis, 2018) shows that shying away from one’s perspective is effortful and time-consuming that necessitates purposeful thinking. People resort to their own perspective when distracted, have little incentive to change, and under time constraints.

For this investigation, the foreseeability-inevitability component model of Blank and Peters (2010) is utilized. Foreseeability refers to being able to predict the outcome as reflected in “I knew-it-all-along.” It is subjectively based on one’s past knowledge that can be susceptible to metacognitive inputs like familiarity and surprise. The information makes the outcome seem more predictable in hindsight than foresight. Inevitability refers to one’s certainty that is based on the situation and more objective, “It had to happen that way.” These 2 components can be independent of one another. For example, tornadoes in different regions are inevitable but not predictable. Unlike the bulk of hindsight bias research, the model encompasses both foreseeability and inevitability judgments that can show differences in these evaluations. Also, it allows for an examination of the self-other distinction across situations that foster egocentrism and its possible ego-changing characteristics of a defensive nature.

Participants should exhibit the traditional hindsight bias in their initial self-judgments when placed in foresight and hindsight conditions. The answers in hindsight participants should be seen to be predictable, certain, and less surprising compared to the foresight participants. When faced with making judgments for the peer, participants should rely on their own perspective for their subsequent foresight and hindsight judgments regarding the peer (e.g., Epley et al., 2004). In the problem-solving situation employed in the present research, participants’ egocentrism should not be altered by the activation of egocentric biases such as devaluing or ego defenses like unforeseeable and retroactive pessimism from the peer perspective. There should be no difference in participants’ predictions of hindsight bias in making self-peer estimates. Such an outcome is also consistent with the commonality of hindsight knowledge in judgments of others in that people think that others share one’s own perspective.

As noted earlier, a clear demonstration of egocentrism showing the sharing of one’s perspective with others was not possible (Fischhoff, 1975; Harley et al., 2004). This difficulty was overcome by the appropriate change in methodology by Son et al. (2021) to reveal that hindsight bias was the same from the self or peer perspective. Initially, participants made self-judgments from a naïve state, gained reorganization of the distorted stimulus, and then tested again from a self or peer perspective. In both instances, participants are faced with remembering their earlier recognition point in making another self or peer judgment and the resultant overestimation effect is dependent on participants’ initial self-judgments. From an anchoring and adjustment viewpoint, their participants anchored on their initial self-judgment and the resulting adjustment was not altered by the change in perspective-taking.

Prior to Son’s et al. (2021) publication, the current research was completed in 2019 with the same focus on the self to peer perspectives. Between the two studies there exist differences in methodology, domain (problem-solving) and dependent measures (foreseeability-inevitability). The goal of the current research was to examine for the presence of egocentrism and hindsight bias in participants’ initial self judgments via the hypothetical method and whether it carried over to their subsequent judgments from the peer perspective.

Also, the hypothetical method clearly demonstrates the influence of the correct answers on participants’ judgments in their initial and subsequent judgments moving from a self to peer perspective. As a result of the earlier research showing the commonality of hindsight knowledge (Bernstein et al., 2018; Birch et al., 2017; Higham et al., 2017; Hoch & Loewenstein, 1989; Wood, 1978) and the better basis for decision-making by participants (Bradfield & Wells, 2005), it is conceivable that any changes that occur in shifting from the self to peer to a peer perspective will be observed principally in participants knowing the answers compared to those who do not know the answers.

Like the research design of Son et al. (2021), participants are likely to anchor spontaneously on their own initial perspective due to its accessibility in adjusting to the peer perspective. Because of the repeated exposure to the testing stimuli involved in the self-peer comparison, it is important to employ an adequate baseline comparison condition for the self to peer perspectives.

To address the issue of egocentrism in perspective-taking for hindsight bias, the task of solving anagrams was employed. Anagrams are scrambled letters which unscramble to produce a meaningful word. Solutions to anagrams involve problem-solving skills of an “aha” nature. The solution occurs rapidly and participants in hindsight find the answers obvious and less difficult (Hom & Ciaramitaro, 2001). These features are also present in perceptual forms of hindsight bias involving vision and audition. Though the task is simpler than the decision-making involved in forecasting stock markets, weather, and the coronavirus pandemic, it shares many aspects of the same uncertainty for those predicting the correct answer.

Via the hypothetical method, participants were tested in two phases. In phase 1, participants were required to make self-estimates from a foresight (naïve) or hindsight (non-naïve) condition with or without being told to ignore the correct answers when making self-predictions and then for a peer. Ignoring the correct answers has been a key feature of much hindsight bias research to show the robustness of the effect (Roese & Vohs, 2012). To allow for a self-versus peer comparison, included was a condition where participants made their predictions for a peer but not themselves. Next, in phase 2, all participants were asked to make their judgments for the peer. This procedure allowed a self-peer repeated measure comparison for the foresight and hindsight conditions and the condition involving the repeated estimates for the peer. Multiple dependent measures typically used in hindsight bias research were employed in this study (e.g., surprise) to examine the hindsight bias in participants’ predictions for themselves and their judgments about how a peer would answer.

Earlier, Bradfield and Wells (2005) showed that participants who possessed outcome knowledge believed they had a superior basis for their decision-making and greater skills in understanding others. If so, participants in hindsight should believe their peer’s predictions are more accurate or realistic than participants in the foresight condition. This aspect is assessed by the same dependent measures employed to examine hindsight bias.

Finally, to examine participants’ hurdles in understanding the peer’s way of thinking, they were asked to evaluate their difficulty and confidence in judging themselves and the peer. Participants should find it more difficult and be less confident to access or estimate their knowledge of the peer because their own knowledge and beliefs are more readily accessible.



A total of 193 participants were recruited from Introductory Psychology classes and tested in three large group settings at a public university. Participants were tested and given two research participation credits following the APA ethical guideline after receiving their informed consent. Of the 193 participants, five were excluded from the research for failing to answer all questions. Based on multivariate and univariate analyses (± 3 SD) for outliers, five more participants did not receive consideration. The sample was comprised mostly of US citizens (99.5%), Caucasians (84.1%), more females (67.8%) than males, with an overall average age of 21.98 years (SD = 6.29).


This study involved the hypothetical method of investigating hindsight bias consisting of a 4 (foresight/three hindsight conditions) × 2 (male/female) × 2 (self-peer judgment) factorial design with repeated measures for the third factor. Proportionally balanced for gender, participants were randomly assigned to either the foresight or hindsight condition with the sample size varying from 44 to 48 participants for each condition. A problem-solving task (5-letter anagrams) was employed for this experiment.

For each condition, participants read information about the nature of predictions. They read those predictions are often based on their experience and can be a guess to make plans. Next, they read several examples including that the task dealt with solving moderately difficult anagrams (e.g., feukl-fluke) that are mixed-up letters that can be solved to produce a meaningful word (rapty-party), and the ability to solve them involved cognitive flexibility and intelligence. Finally, participants needed to imagine they had 10 s to solve each of the five anagrams. Then, they received a 10-s period to remind them of this time span. Also, participants were reminded to make their judgments quickly to complete testing.

In phase 1 of the foresight condition (F), naïve participants did not see anagrams with the correct answers and were asked to make their self-predictions for the correct answers. However, participants in the hindsight condition were not naïve as they saw anagrams with the correct answers. The hindsight condition was further varied with participants ignoring (HI, e.g., “what would you have predicted before knowing the answers”) or not ignoring (HNI) the answers as they made their self-predictions. Also, rather than focusing on their self-predictions, participants in another hindsight condition made perspective judgments for a peer by the name of Matthew at the university. He was someone they did not know and the same age as the participant. Participants were asked to ignore the answers as they made their peer predictions (HPI) and to imagine that he had the same experience as themselves.

During phase 2, participants who made self-predictions in phase 1 received their perspective-taking instructions directed at the pier and made their predictions for the same anagrams used in phase 1. In contrast, participants who had rendered their judgments for the peer in phase 1 were reminded of the perspective-taking instructions and asked to judge the anagrams again for phase 2. After completing phase 2, participants answered questions assessing hindsight bias. Next, they were queried about how accurate or realistic their predictions were for the peer, followed by difficulty and confident judgments for themselves and the peer.


Hindsight bias measures for self and peer predictions were assessed for each anagram (i.e., difficulty, obviousness, surprise, familiarity, and chance of solving anagrams) using a 7-point Likert scale. Difficulty, obviousness, and surprise were anchored by 1 = very little to 7 = very much; familiarity, 1 = slightly seen, heard, or used to 7 = often seen, heard, used every day; and chance of solving anagrams, 1 = very low to 7 = very great. These same questions were utilized to assess participants’ assertions of the accuracy or realism about their predictions for the peer. The Likert scale of 1 = very little and 7 = very much was also employed for participants’ judgments about their difficulty and confidence in self and peer’s judgments. The measures of difficulty and surprise were reversed coded for the statistical analyses.


An average score was computed across five anagrams for the dependent measures of obviousness, familiarity, chance of solving anagrams, surprise, and difficulty during phases 1 and 2 to examine for hindsight bias. Initially, the analyses included the gender factor, which only resulted in one main effect. Thus, the gender factor was excluded from subsequent analyses due to the unequal number of gender participants in the experiment with the lack of specific hypotheses regarding gender differences.

Based on the component view of hindsight bias (Blank et al., 2008), a composite measure of foreseeability was created using difficulty, obviousness, and familiarity with a single-item assessment of inevitability (chance) and surprise. Using a Cronbach’s alpha of .60 as the minimum score, these analyses showed that the foreseeability composites were satisfactory for phases 1 (α = .70) and 2 (α = .85). For the accuracy or realistic composite, difficulty was eliminated that resulted in a satisfactory composite (α = .65).

Visually, an examination of the means (See Table 1 for Means and Standard Deviations) for foreseeability, inevitability, and surprise showed the hindsight conditions of HI, HNI, and HPI clustered together separately from the F condition in phase 1 and 2. Each hindsight condition rose upward for foreseeability and inevitability, and surprise decreased across the phases.

Table 1 Means and Standard Deviations for Phases 1 and 2 for foresight or hindsight condition

Full size table

Judgments for Self or the Peer: Phase 1

A one-way MANOVA entailing self-judgments for foreseeability, inevitability, and surprise in phase 1 for F, HI, HNI, and judgments of Matthew (HPI) yielded only the main effect of conditions, F(9, 430.92) = 4.51, p < .001, η2 = .07. Significant univariate contributors to this effect were foreseeability, F(3, 179) = 9.94, p < .001, η2 = .14; and surprise, F(3, 179) = 3.74, p = .01, η2 = .06; but not inevitability (p > .05).

Initially, it was important to establish that the problem-solving situation resulted in hindsight bias for participants making judgments for themselves. Multiple comparisons relevant to the hypotheses were analyzed via Tukey’s HSD with a significance level of p < .05. F participants were compared to their counterparts in the HI and HNI conditions. These comparisons showed hindsight bias with participants in both hindsight conditions deeming the anagram solutions to be more foreseeability (ps < .001). This outcome was supported further by the HI participants who found the solutions to be less surprising (p = .008) and to a lesser extent by the HNI comparison (p = .090) to their F counterparts.

Asking participants to ignore (HI) or not ignore (HNI) the answers to the anagrams did not yield any differences for foreseeability or surprise, ps > .05. Importantly, participants hindsight judgments for these measures aimed at the peer (HPI) were the same when compared to HI and HNI participants, ps > .05. The findings of foreseeability and surprise for phase 1 showed hindsight bias in participants’ judgments for themselves and their expectations for the peer in hindsight were the same. Although inevitability was not a significant contributor to the main effect, the mean tendencies for the hindsight conditions followed the findings for foreseeability and inevitability.

Judgments for Self and the Peer: Phases 1 and 2

After making judgments for themselves or the peer in phase 1, participants assessed the anagrams again from the peer’s perspective. The issue was participants’ judgments for self or peer was altered by the focus on Matthew. A two-way MANOVA with conditions (F, HI, HNI, HPI) and the self-peer repeated measures as factors were conducted for foreseeability, surprise, and inevitability.

The multivariate analyses indicated a significant conditions effect for foreseeability, F(3, 179) = 14.78, p < .001, η2 = .20; inevitability, F(3, 179) = 4.69, p = .004, η2 = .07; and surprise, F(3, 179) = 11.49, p < .001, η2 = .16. The latter was qualified by an interaction between conditions and the self-peer factor, F(3, 179) = 4.51, p = .004, η2 = .07.

Examination of this interaction revealed the tendency of F participants to predict more surprise for the peer and less surprise for the HI, HNI, and HPI participants than themselves. These changes in the hindsight conditions also occurred for foreseeability and inevitability in the reverse order in that they increased their estimates for the peer. Unlike surprise, there was not a change tendency in the F condition for these measures.

Subsequently, the interactive effect negated the self-peer ratings across conditions for surprise (p > .05) but not foreseeability, F(1, 179) = 12.02, p = .001, η2 = .06, and inevitability, F(1, 179) = 14.07, p < .001, η2 = .07. Each hindsight condition shifted upward for foreseeability and inevitability, ps < .05, while foresight participants failed to change (p > .05. This change showed the consequence of participants making repeated judgments having knowledge of the answers in hindsight compared to the foresight participants without the answers.

Further examination of the conditions effect revealed the following results when comparing the F condition to each hindsight condition (HI, HNI) for signs that the hindsight bias observed in phase 1 carried over to participants’ expectations for the peer (see Table 1 for Means and Standard Deviations). F participants differed from each hindsight condition for foreseeability, (ps < .001), surprise (ps < .001), and inevitability (p = .03 or .003). Hindsight participants saw the anagram solutions to be more foreseeable, inevitable, and less surprising than their foresight counterparts over the self-peer phases. The findings for foreseeable and surprising were the same as in phase 1 for participants’ self-judgments. The inclusion of inevitability shows the consistency of participants’ judgments in these measures from the self to the peer perspective.

Including HPI in the analyzes revealed that the hindsight bias in participants’ judgments for themselves in phase 1 was the same when the target became the peer. Except for inevitability, HI and HPI participants deemed the solutions to be more foreseeable (ps = .001) and less surprising (ps = .001) than F participants without a difference between HI and HPI, ps > .05. This same outcome was seen in inevitability except the difference between F and HI participants was marginal (p = .08). Hence, making multiple judgments in hindsight (HPI) or a single judgment following one’s own judgment (HI) was the same.

Furthermore, participants’ estimates did not differ among the conditions of HI, HNI and HPI, for foreseeability, inevitability, and surprise, ps > .05. Regardless of the self to peer perspective or making judgments again more than once for the peer, their responses for the peer did not differ significantly from their own. The hindsight changes that occurred for these measures with repeated judgments reflects the familiarity of processing the answers a second time.

Accuracy or Realism of Predictions about the Peer

The following issue dealt with how accurate or realistic was participants’ judgments about the peer. A one-way MANOVA with conditions as the factor was conducted for the dependent measures. There was a significant main effect of conditions, F(9, 430.92) = 4.13, p < .001, η2 = .07, with significant contributions from the factors of foreseeability, F(3, 179) = 5.75, p = .001, η2 = .09; surprise, F(3, 179) = 5.43, p = .001, η2 = .08; and inevitability, F(3, 179) = 4.77, p = .003, η2 = .07.

No differences (see Table 2 for Means and Standard Deviations) were found when comparing participants’ judgments among the hindsight perspectives (HNI, HI, HPI) for any of the dependent measures, ps > .05. For foreseeability, participants felt more accurate or realistic in their hindsight judgments in the HI (p = .001) and HNI (p = .024) conditions involving a single assessment or multiple assessments for the peer (p = .030) than from the foresight perspective. Unlike these outcomes for foreseeability, there was not the same consistency in the surprise and inevitability measures. For example, participants in the HNI (p = .006) and the HPI (p = .002) but not HI (p > .05) conditions believed they were more accurate or realistic in their surprise estimates than the foresight participants. For inevitability, the accuracy or realistic judgments were greater for HI (p = .002) than F participants but not when the latter comparison was to HNI and HPI (ps > .05).

Table 2 Means and Standard Deviations on accuracy/realism measures as a function of foresight or hindsight condition for the peer

Full size table

Difficulty and Confidence in Predictions about the Peer

Finally, participants made difficult and confident judgments about themselves and the peer’s predictions. A two-way MANOVA with all conditions and the self-peer factor repeated measures showed only a significant main effect for the repeated measures, F(1, 179) = 75.63, p < .001, η2 = .30. Participants reported more difficulty in making predictions for the peer than for themselves (Self M = 4.38, SD = 1.91; Peer M = 2.78, SD = 1.72). Similarly, the only significant outcome for confidence was the self-peer factor, F(1, 179) = 93.07, p < .001, η2 = .34. Participants felt more confident in making judgments for themselves than the peer (Self M = 4.75, SD = 1.33; Peer M = 3.07, SD = 1.81). In making these decisions, participants were less confident and found it more difficult to judge the peer.


Participants exhibited egocentrism by perceiving the peer to essentially share the same hindsight perspective as themselves in problem-solving. In their initial judgments for themselves showing hindsight bias, participants judged the anagram solutions to be more predictable and less surprising than the foresight participants. Participants who made their initial judgments from a self-perspective when shifted to the peer perspective showed hindsight bias across the self-peer phases for foreseeability and surprise with the addition of inevitability. The inclusion of an appropriate baseline indicated that the hindsight bias observed in the self-perspective carried over to the peer-perspective. Ignoring (HI and HPI) or not ignoring (HNI) the answers in hindsight were inconsequential. Although post-hoc, individuals claimed their predictions were more accurate or realistic for the peer in hindsight than foresight for foreseeability, yet it was more challenging to predict for the peer than themselves.

Consistent with Nickerson’s (1999) supposition, participants in their self and/or peer judgments projected their estimations onto the peer by anchoring and adjusting on their own knowledge. Their hindsight determinations were similar for the comparison between themselves, and the peer and this effect was observed in their self to peer assessments. The correlations for each hindsight condition involving the measures of foreseeability, surprise, and inevitability were all significant (ps ranging from < .000–.029) with rs ranging from .62 to .33. Furthermore, the projective determinations of participants dovetail with earlier research showing that knowledge in hindsight is common in the greater population in judgments of others (Birch et al., 2017; Hoch & Loewenstein, 1989; Wood, 1978).

Believing that the peer’s hindsight bias is the same as for oneself and that outcome knowledge is widespread could be tied to participants’ accuracy or realistic judgments about the peer. Participants indicated their foreseeability judgments were more accurate or realistic in hindsight than foresight suggesting that they understood the peer’s way of responding with the following caveat. Roese and Vohs (2012) indicate that such a perception can falsely promote one’s understanding of others to produce negative outcomes.

The present research builds on the earlier findings of Son’s et al., (2021) research. One, it provides a replication showing egocentrism via the hypothetical method using a problem-solving task, and the dependent measures of foreseeability, inevitability, and surprise. Two, the findings shows that the expression of egocentrism and hindsight bias need not involve the multiple testing procedure of Son’s et al. (2021) procedure comparing self-self to self-peer judgments. Signs of hindsight bias and egocentrism occurred in participants’ initial judgments that was maintained in shifting to the peer perspective. Three, due to repeated testing, participants showed significant changes in the shifting of perspectives that did not alter the expression of egocentrism or hindsight bias. Four, showing further significance of knowing the answers in hindsight triggered participants’ changes for foreseeability, inevitability, and surprise in the same fashion. Five, the answers in hindsight resulted in participants believing that their foreseeability predictions for the peer were more accurate or realistic. Six, it was more challenging in general to predict for the peer than themselves. Seven, the showing of egocentrism and hindsight bias in the current research involved different groups of participants rather than the occurrence of these events within the participants.

Common to Son’s et al. (2021) and the current research, the situations did not involve responsibility and blame or self-relevant negative outcomes. The current research and Son’s et al. (2021), experiment 1 lacked an incentive for participants to question their evaluation to show egocentrism. However, in the latter for experiment 2, with the target being children, the expression of egocentrism was interrupted. Participants may use the same strategy for themselves and the peer because it is easy and effortless. To adjust a forecast by viewing the peer to be different, requires purposeful thinking. The work of Son et al. (2021) and the current findings, expands our understanding of how egocentrism and perspective-taking play out in hindsight bias.

Despite the proposed importance of the current findings, there exist possible limitations. The target for perspective-taking was an unfamiliar male, order effects of participants making judgments for themselves and then the peer, ecological validity of judging for a “hypothetical” peer instead of a real person and the interactive effect found for the surprise measure. Regarding ecological validity, Harley et al. (2004) reported the existence of hindsight bias in children when they recognized ambiguous pictures of objects from the perspective of “Ernie” from Sesame Street. Their research raises issues about broadening targets for perspective-taking to include non-human figures like dolls and avatars. The interactive effect for surprise indicated a tendency for all the hindsight conditions to exhibit less surprise in shifting to the peer-perspective, there was a tendency for greater surprise in the foresight participants. This tendency for the foresight participants was not present for the other measures of foreseeability-inevitability. While these latter measures are consistent with a cognitive account, it suggests the presence of a motivation to view the peer to be more surprised than themselves and that the alternation of one’s egocentrism is more susceptible to emotions.

The hindsight bias effect consistently occurs even when individuals are directed to ignore the outcome information as they make their predictions in hindsight (Roese & Vohs, 2012). In separate experiments, Fischhoff (1975) showed hindsight bias with or without the direction to ignore the outcome information. In the present research, asking participants to ignore or not ignore the correct answers to the anagrams failed to alter the expression of hindsight bias.

Much research has demonstrated that hindsight bias is exceedingly difficult of mitigate or to eliminate. Passive manipulations like warning again the bias (Fischhoff, 1975), to avoid the bias (Kamin & Rachlinski (1995), providing incentives for correct judgments (Hell et al., 1988) and forcing individuals to try harder (Fischhoff, 1975), fail to reduce the bias substantially. Generally, these outcomes indicate participants’ inability to ignore the correct answers in hindsight. In 2019, Dietvorst and Simonsohn proposed the novel idea that individuals choose to use the outcome information rather than to ignore it. This tendency can be overcome when there is sufficient justification to ignore the outcome. In one study, a majority of their 393 participants (73%) felt that the outcome information enabled them to make more accurate judgements about the responses of others in the study. Such a result is consistent with participants believing that their foreseeability judgments were more accurate or realistic in hindsight than foresight. Given the time constraints in the present research, participants might choose to ignore the constrain to ignore the answers in making their judgments.

Implications and Practical Considerations

The present findings in a non-evaluative context lack questioning decision-making or the need to evaluate others’ capabilities. Unlike the former, the latter fuels an incentive for change and the use of intention and purposeful thinking that trumps one’s own perspective-taking. This difference also points to further investigating judgments for oneself and others in an evaluative context based on Blank’s et al. (Blank et al., 2008) component model of hindsight bias. The model can also accommodate positive and negative outcomes and provide an understanding of another’s puzzling behavior, such as President Trump’s comments on the pandemic crisis.

Because it is infeasible to test empirically, the crisis yielded numerous events subjected to speculative reasoning about his decision-making. An event like the pandemic has multiple causes, and it created a great deal of ambiguity for Americans. In their perspective-taking, President Trump and others may resort to mental shortcuts like hindsight bias (“bound to happen”) to reduce uncertainty to make sense of the unfolding events during the pandemic.

President Trump’s decision-making has raised two questions. One was President Trump’s handling of the pandemic crisis. Did he react too late to the magnitude of the pandemic? Two, was the ventilator shortage predictable? The former President Trump initially stated that the number of infections and deaths were quite low and minimized the disaster’s severity.

Perhaps, he chose this approach because he thought Americans would find it comforting. However, as the number of incidents increased exponentially (negative outcome), he seemed to react at times that it was unforeseeable (defensively) or inevitable (retroactive pessimism). In either instance, he denied responsibility and blame. Due to the current electorate’s bipartisan makeup, people may view the former President to be much like or different from themselves. Without the benefit of a foresight perspective, people in hindsight can overestimate the pandemic’s foreseeability while others may see it to have been inevitable. The former belief heightens responsibility and blame accusations, while the latter belief lowers it when directed at the former President.

This same speculative account is also applicable to Dale et al., 2020) reporting on the former President Trump’s statements about the ventilator shortage for the critically ill. According to Dale et al. (2020), he disputed the claim about the shortage by saying that “nobody” could have predicted it and reinforced it later by “It takes a period to build them (ventilators), and again nobody could have known a thing like this could happen.”

Overall, the consideration of prior research utilizing self and peer judgments along with the two examples in evaluating President Trump identifies an important source of further research for perspective-taking effects and egocentrism. In the former, participants are making inferential judgments about Trump from their own perspective without a direct comparison to themselves. Such a comparison reveals whether hindsight bias (unforeseeable/inevitable) varies from themselves relative to Trump. Depending on the outcome, the results provide further nuances about Trump’s actions. The Trump examples without a direct comparison to self-judgments suggest that participants’ own perspective-taking was altered by their purposeful thinking.


Participants faced with adopting the peer’s perspective showed essentially the same hindsight bias as themselves. This result of sharing perspective-taking adds to previous research showing that individuals believe that their hindsight knowledge is commonly present in others. Cognitively, the saliency and accessibility of their own knowledge or what they think they knew, prompted judgments for themselves and a peer with the tendency to view their predictions for the latter to be more accurate. Extending the present research to include cognitive-motivational processes (e.g., blame) provides a richer account of perspective-taking in hindsight bias for problem-solving and other social behavior like stereotyping and prejudice.

Researchers like Dekker (2004) and Hedden (2019), and others have proposed ideas that are challenging the direction of hindsight bias research. Dekker (2004) questions whether hindsight bias is even a bias and Hedden (2019) argues that it is not necessarily an irrational process. Von der Beck et al. (2019) ask whether hindsight bias can even exist without the receipt of definite outcome knowledge by showing hindsight bias in people’s use of conjectures about the outcome. Dietvorst and Simonsohn (2019) question the assumption that people exhibit hindsight because they cannot ignore the outcome information. Oeberst et al. (2018) reveal sources of hindsight bias in the mass media and that even the reading of a biased article generates hindsight bias in a reader. Most recently, Meuer et al. (2021) showed that hindsight bias conditions the way we write. Based on the research of Son et al. (2021) to debias hindsight bias via perspective-taking, and the present investigation, researchers should not overlook perspective-taking effects in hindsight bias as much social interaction embraces reasoning about another person.


  • Arkes, H. R., Faust, D., Guilmette, T. J., & Hart, K. (1988). Eliminating the hindsight bias. Journal of Applied Psychology, 73(2), 305–307. https://doi.org/10.1037/0021-9010.73.2.305

    Article  Google Scholar 

  • Bernstein, D. M., & Harley, E. M. (2007). Fluency misattribution and visual hindsight bias. Memory, 15(5), 548–560. https://doi.org/10.1080/09658210701390701

    Article  PubMed  Google Scholar 

  • Bernstein, D. M., Erdfelder, E., Meltzoff, A. N., Peria, W., & Loftus, G. R. (2011). Hindsight bias from 3 to 95 years of age. Journal of Experimental Psychology: Learning, Memory, and Cognition, 37(2), 378–391. https://doi.org/10.1037/a0021971

    Article  PubMed  Google Scholar 

  • Bernstein, D. M., Wilson, A. M., Pernat, N., & Meilleur, L. (2012). Auditory hindsight bias. Psychonomic Bulletin & Review, 19(4), 588–593. https://doi.org/10.3758/s13423-012-0268-0

    Article  Google Scholar 

  • Bernstein, D. M., Aßfalg, A., Kumar, R., & Ackerman, R. (2015). Looking backward and forward on hindsight bias. In J. Dunlosky & S. K. Tauber (Eds.), The Oxford handbook of metamemory (pp. 289–304). Oxford University Press.

    Google Scholar 

  • Bernstein, D. M., Kumar, R., Masson, M. E., & Levitin, D. J. (2018). Fluency misattribution and auditory hindsight bias. Memory & Cognition, 46(8), 1331–1343. https://doi.org/10.3758/s13421-018-0840-6

    Article  Google Scholar 

  • Birch, S. A. J., Brosseau-Liard, P. E., Haddock, T., & Ghrear, S. E. (2017). A ‘curse of knowledge’ in the absence of knowledge? People misattribute fluency when judging how common knowledge is among their peers. Cognition, 166, 447–458. https://doi.org/10.1016/j.cognition.2017.04.015

    Article  PubMed  Google Scholar 

  • Blank, H., & Peters, J. H. (2010). Controllability and hindsight components: Understanding opposite hindsight biases for self-relevant negative event outcomes. Memory & Cognition, 38(3), 356–365. https://doi.org/10.3758/MC.38.3.356

    Article  Google Scholar 

  • Blank, H., Nestler, S., von Collani, G., & Fischer, V. (2008). How many hindsight biases are there? Cognition, 106(3), 1408–1440. https://doi.org/10.1016/j.cognition.2007.07.007

    Article  PubMed  Google Scholar 

  • Bradfield, A., & Wells, G. L. (2005). Not the same old hindsight bias: Outcome information distorts a broad range of retrospective judgments. Memory & Cognition, 33(1), 120–130. https://doi.org/10.3758/BF03195302

    Article  Google Scholar 

  • Bryant, F. B., & Guilbault, R. L. (2002). "I knew it all along" eventually: The development of hindsight bias in reaction to the Clinton impeachment verdict. Basic and Applied Social Psychology, 24(1), 27–41. https://doi.org/10.1207/153248302753438903

    Article  Google Scholar 

  • Chuan, S.-H., Hoffmann, R., Liu, B., & Tan, M. (2019). Is knowledge cursed when forecasting the forecasts of others? Journal of Behavioral Finance, 20(1), 66–72 https://doi/org/10.1080/15427560.2018.1464454

    Article  Google Scholar 

  • Dale, D., Cohen, M., & Subramaniam, T. (2020). Fact check: Trump again misleads on ventilator shortages and coronavirus timeline at off-topic briefing. https://www.cnn.com/2020/04/01/politics/fact-check-trump-coronavirus-briefing-april-1/index.html

  • Danz, D. (2020). Never underestimate your opponent: Hindsight bias causes overplacement and overentry into competition. Games and Economic Behavior, 124, 588–603. https://doi.org/10.1016/J.GEB.2020.10.001

    Article  Google Scholar 

  • Darcy, O. (2020). Trump viciously attacks NBC News reporter in extended rant after being asked for message to Americans worried about coronavirus. https://edition.cnn.com/2020/03/20/media/trump-rant-at-nbcnews-peter-alexander/index.html

  • Dekker, S. W. (2004). The hindsight bias is not a bias and not about history. Factors and Aerospace Safety, 4(2), 87–99.

    Google Scholar 

  • Dietvorst, B. J., & Simonsohn, U. (2019). Intentionally “biased”: People purposely use to-be ignored information, but can be persuaded not to. Journal of Experimental Psychology: General, 148(7), 1228–1238. https://doi.org/10.1037/xget0000541

    Article  Google Scholar 

  • Epley, N., Keysar, B., van Boven, L., & Gilovich, T. (2004). Perspective taking as egocentric anchoring and adjustment. Journal of Personality and Social Psychology, 87(3), 327–339. https://doi.org/10.1037/0022-3514.87.3.327

    Article  PubMed  Google Scholar 

  • Felson, R. B., & Palmore, C. (2018). Biases in blaming victims of rape and other crime. Psychology of Violence, 8(3), 390–399. https://doi.org/10.1037/vi0000168

    Article  Google Scholar 

  • Fischhoff, B. (1975). Hindsight is not equal to foresight: The effect of outcome knowledge on judgment under uncertainty. Journal of Experimental Psychology: Human Perception and Performance, 1(3), 288–299. https://doi.org/10.1037/0096-1523.1.3.288

    Article  Google Scholar 

  • Fischhoff, B., Gonzalez, R. M., Lerner, J. S., & Small, D. A. (2005). Evolving judgments of terror risks: Foresight, hindsight, and emotion. Journal of Experimental Psychology: Applied, 11(2), 124–139. https://doi.org/10.1037/1076-898X.11.2.124

    Article  PubMed  Google Scholar 

  • Ghrear, S., Chudek, M., Fung, K., Mathew, S., & Birch, S. A. J. (2019). Cultural variations in the curse of knowledge: The curse of knowledge bias in children from a nomadic pastoralist culture in Kenya. Journal of Cognition and Culture, 19(3–4), 366–384. https://doi.org/10.1163/115685373-12340064

    Article  Google Scholar 

  • Giroux, M. E., Coburn, P. I., Harley, E. M., Connolly, D. A., & Bernstein, D. M. (2016). Hindsight bias and law. Zeitschrift fur Psychologie, 224(3), 190–203. https://doi.org/10.1027/2151-2604/a000253

    Article  Google Scholar 

  • Giroux, M. E., Coburn, P. I., Connolly, D. A., & Bernstein, D. M. (2017). Perspective taking abilities across the lifespan: A review of hindsight bias and theory of mind. In M. Toplak & J. Weller (Eds.), Individual differences in judgment and decision making from a developmental context (pp. 147–165). Psychology Press.

    Google Scholar 

  • Groß, J., & Pachur, T. (2019). Age differences in hindsight bias: A meta-analysis. Psychology and Aging, 34(2), 294–310. https://doi.org/10.1037/pag0000329

    Article  PubMed  Google Scholar 

  • Harley, E. M. (2007). Hindsight bias in legal decision making. Social Cognition, 25(1), 48–63. https://doi.org/10.1521/soco.2007.25.1.48

    Article  Google Scholar 

  • Harley, E. M., Carlsen, K. A., & Loftus, G. R. (2004). The "saw-it-all-along" effect: Demonstrations of visual hindsight bias. Journal of Experimental Psychology: Learning, Memory, and Cognition, 30(5), 960–968. https://doi.org/10.1037/0278-7393.30.5.960

    Article  PubMed  Google Scholar 

  • Hedden, B. (2019). Hindsight bias is not a bias. Analysis, 79(1), 43–52. https://doi.org/10.1093/analys/any023

    Article  Google Scholar 

  • Higham, P. A., Neil, G. J., & Bernstein, D. M. (2017). Auditory hindsight bias: Fluency misattribution versus memory reconstruction. Journal of Experimental Psychology: Human Perception and Performance, 43(6), 1143–1159. https://doi.org/10.1037/xhp0000405

    Article  Google Scholar 

  • Hoch, S. J., & Loewenstein, G. F. (1989). Outcome feedback: Hindsight and information. Journal of Experimental Psychology: Learning, Memory, and Cognition, 15(4), 605–619. https://doi.org/10.1037/0278-7393.15.4.605

    Article  Google Scholar 

  • Hom Jr., H. L., & Ciaramitaro, M. (2001). GTIDHNIHS: I knew-it-all-along. Applied Cognitive Psychology, 15(5), 493–597. https://doi.org/10.1002/acp.716

    Article  Google Scholar 

  • Hom Jr., H. L., & Kaiser, D. L. (2016). Role of hindsight bias, ethics, and self-other judgments in students’ evaluation of an animal experiment. Ethics & Behavior, 26(1), 1–13. https://doi.org/10.1080/10508422.2014.963223

    Article  Google Scholar 

  • Krueger, J., & Clement, R. W. (1997). Estimates of social consensus by majorities and minorities: The case for social projection. Personality and Social Psychology Review, 1(4), 299–313. https://doi.org/10.1207/s15327957pspr0104_2

    Article  PubMed  Google Scholar 

  • LaBine, S. J., & LaBine, G. (1996). Determinations of negligence and the hindsight bias. Law and Human Behavior, 20(5), 501–516. https://doi.org/10.1007/BF01499038

    Article  Google Scholar 

  • Louie, T. A., Curren, M. T., & Harich, K. R. (2000). "I knew we would win": Hindsight bias for favorable and unfavorable team decision outcomes. Journal of Applied Psychology, 85(2), 264–272. https://doi.org/10.1037/0021-9010.85.2.264

    Article  Google Scholar 

  • Mark, M. M., Boburka, R. R., Eyssell, K. M., Cohen, L. L., & Mellor, S. (2003). “I couldn’t have seen it coming”: The impact of negative self-relevant outcomes on retrospections about foreseeability. Memory, 11(4–5), 443–454. https://doi.org/10.1080/09658210244000522do

    Article  PubMed  Google Scholar 

  • McDermott, C. M., Miller, M. K., & DeVault, A. (2020). They should have known: Hindsight and outcome biases in child abduction cases. Journal of Applied Social Psychology, 50(7), 430–438. https://doi.org/10.1111/jasp.12671

    Article  Google Scholar 

  • Meuer, M., Nestler, S., & Oeberst, A. (2021). Debiasing media articles-reducing hindsight bias in the production of written work. Journal of Applied Research in Memory and Cognition. https://doi.org/10.1016/j.jarmac.2020.12.006

  • Mullen, B. (1983). Egocentric bias in estimates of consensus. The Journal of Social Psychology, 121(1), 31–38. https://doi.org/10.1080/00224545.1983.9924463

    Article  Google Scholar 

  • Nickerson, R. S. (1999). How we know—And sometimes misjudge—What others know: Imputing one's own knowledge to others. Psychological Bulletin, 125(6), 737–759. https://doi.org/10.1037/0033-2909.125.6.737

    Article  Google Scholar 

  • Oeberst, A., von der Beck, I., Back, M. D., Cress, U., & Nestler, S. (2018). Biases in the production and reception of collective knowledge: The case of hindsight bias in Wikipedia. Psychological Research, 82(5), 1010–1026. https://doi.org/10.1007/s00426-017-0865-7

    Article  PubMed  Google Scholar 

  • Pezzo, M. V. (2011). Hindsight bias: A primer for motivational researchers. Social and Personality Psychology Compass, 5(9), 665–678. https://doi.org/10.1111/j.1751-9004.2011.00381.x

    Article  Google Scholar 

  • Pohl, R. F., Bender, M., & Lachmann, G. (2002). Hindsight bias around the world. Experimental Psychology, 49(4), 270–282. https://doi.org/10.1026/1618-3169.49.4.270

    Article  PubMed  Google Scholar 

  • Pohl, R. F., Schwarz, S., Sczesny, S., & Stahlberg, D. (2003). Hindsight bias in gustatory judgments. Experimental Psychology, 50(2), 107–115. https://doi.org/10.1026/1618-3169.50.2.107

    Article  PubMed  Google Scholar 

  • Roese, N. J., & Maniar, S. D. (1997). Perceptions of purple: Counterfactual and hindsight judgments at northwestern wildcats football games. Personality and Social Psychology Bulletin, 23(12), 1245–1253. https://doi.org/10.1177/01461672972312002

    Article  Google Scholar 

  • Roese, N. J., & Vohs, K. D. (2012). Hindsight bias. Perspectives on Psychological Science, 7(5), 411–426. https://doi.org/10.1177/1745691612454303

    Article  PubMed  Google Scholar 

  • Ross, L., Greene, D., & House, P. (1977). The false consensus effect: An egocentric bias in social perception and attribution processes. Journal of Experimental Social Psychology, 13(3), 279–301. https://doi.org/10.1016/0022-1031(77)90049-X

    Article  Google Scholar 

  • Son, L. K., Hong, S. S., Han, L., Lee, Y., & Kim, T. H. (2021). Taking a naïve other’s perspective to debias the hindsight bias: Did it backfire? New Ideas in Psychology, 62. https://doi.org/10.1016/j.newideapsych.2021.100867

  • Stanley, M., Stone, A. R., & Marsh, E. J. (2021). Cheaters claim they knew the answers all along. Psychonomic Bulletin & Review, 28(1), 341–350. https://doi.org/10.3758/s13423-020-01812-w

    Article  Google Scholar 

  • Strohmaier, N., Pluut, H., van den Bos, K., & Adriaanse, J. (2021). Hindsight bias and outcome bias in judging directors’ liability and the role of free will beliefs. Journal of Applied Social Psychology, 51(3), 141–158. https://doi.org/10.1111/jasp.12722

    Article  Google Scholar 

  • Tullis, J. G. (2018). Predicting others’ knowledge: Knowledge estimation as cue utilization. Memory & Cognition, 46(8), 1360–1375. https://doi.org/10.3758/s13421-018-0842-4

    Article  Google Scholar 

  • Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124–1135.

    Article  Google Scholar 

  • Tykocinski, O. E., Pick, D., & Kedmi, D. (2002). Retroactive pessimism: A different kind of hindsight bias. European Journal of Social Psychology, 32(4), 577–588. https://doi.org/10.1002/ejsp.120

    Article  Google Scholar 

  • von der Beck, I., Cress, U., & Oeberst, A. (2019). Is there hindsight bias without real hindsight bias? Conjectures are sufficient to produce hindsight bias. Journal of Experimental Psychology: Applied, 25(1), 88–99. https://doi.org/10.1037/xap0000185

    Article  PubMed  Google Scholar 

  • Walmsley, S., & Gilbey, A. (2019). Understanding the past: Investigating the role of availability, outcome, and hindsight bias and close calls in visual pilots’ weather-related decision-making. Applied Cognitive Psychology, 33(6), 1124–1136. https://doi.org/10.1002/acp.3557

    Article  Google Scholar 

  • Wilson, C. G., Shipley, T. F., & Davatzes, A. K. (2020). Evidence of vulnerability to decision bias in expert field scientists. Applied Cognitive Psychology, 34(5), 1212–1223. https://doi.org/10.1002/acp.3677

    Article  Google Scholar 

  • Wood, G. (1978). The knew-it-all-along effect. Journal of Experimental Psychology: Human Perception and Performance, 4(2), 345–353. https://doi.org/10.1037/0096-1523.4.2.345

    Article  Google Scholar 

Download references

What is meant by hindsight bias?

Hindsight bias is our tendency to look back at an event that we could not predict at the time and think the outcome was easily predictable. It is also called the 'knew-it-all-along' effect.

What is hindsight bias and how can it make research findings seem like common sense?

Hindsight bias is the tendency to believe, after learning an outcome, that we would have foreseen it. Thus, learning the outcome of a study can make it seem like obvious common sense. Scientific inquiry and critical thinking can help us overcome this tendency to overestimate our unaided intuition.

What is meant by hindsight bias quizlet?

Hindsight bias. the tendency to believe, after learning an outcome, that one would have foreseen it. (Also known as the I-knew-it-all along phenomenon.) Critical thinking. thinking that does not blindly accept arguments and conclusions.

Why is hindsight bias so common?

Hindsight bias might not be tied to only the ineffective processing of information but be rooted in adaptive learning. People are susceptible to hindsight bias because it's comforting to think that the world is predictable and thus somewhat orderly. As a result, we seek to see unpredictable events as predictable.