Optimism Bias¶
Core Idea¶
Optimism bias is a persistent cognitive pattern in which (1) individuals systematically overestimate the probability of positive outcomes for themselves relative to base rates (their likelihood of experiencing promotions, good marriages, financial success, good health); (2) correspondingly underestimate the probability of negative outcomes for themselves (divorce, illness, accidents, financial loss); (3) asymmetrically update beliefs in response to new information — beliefs update more fully toward evidence of better-than-expected outcomes than toward evidence of worse-than-expected outcomes; and (4) apply the bias more strongly to self than to others — estimates of outcomes for comparable others are typically more calibrated than estimates for self. The construct was systematized by Neil Weinstein (1980) in his foundational work Unrealistic Optimism About Future Life Events [1] and has been extensively investigated by Tali Sharot and colleagues using neuroimaging and computational models[2]. Modest optimism bias is associated with motivational and psychological benefits; severe calibration failure is associated with risk-taking pathologies. The bias is distinct from trait-level optimism (Scheier & Carver's dispositional construct), denial, illusion of control, or self-serving attribution — it is a specific asymmetry in probability estimation and belief updating measurable through comparative judgment tasks.
How would you explain it like I'm…
Sunny Glasses About Yourself
Overrating your own good luck
Lopsided Self-Predictions
Structural Signature¶
An agent maintains probability estimates p̂(outcome|self) for a set of outcomes and updates these estimates based on experience and information. Under optimism bias, the updating rule is asymmetric: new information x that suggests better-than-expected outcomes is incorporated with higher learning rate α_good than new information suggesting worse-than-expected outcomes (learning rate α_bad < α_good). Over time this asymmetric updating produces expectations systematically shifted toward positive outcomes. The structural signature comprises six key mechanisms: the comparative-risk underestimate [1], the self-other asymmetry[2], the planning-fallacy delay [3], the controllability illusion, the depressive-realism asymmetry [4], and the motivated probability assessment [5]. [6] The bias is self-specific: estimates for other individuals show less asymmetric updating. The structural signature is valence-asymmetric belief updating that preferentially incorporates positive-outcome evidence.
What It Is Not¶
- It is not trait-level optimism — dispositional optimism (Scheier & Carver) is a stable personality trait predicting positive outcome expectancies generally; optimism bias is a cognitive-updating asymmetry that most people exhibit. The two are related but distinct constructs.
- It is not denial — denial is motivated refusal to process disconfirming information; optimism bias is asymmetric processing of information, with bad news being encoded but under-weighted in updating.
- It is not the illusion of control — illusion of control is a belief that one can influence outcomes; optimism bias concerns probability estimates regardless of perceived control. They can co-occur but are distinct.
- It is not the self-serving attribution bias — self-serving attribution concerns causal attribution of outcomes (crediting self for positives, blaming externals for negatives); optimism bias concerns probability estimates for outcomes that have not yet occurred.
- It is not universally present — depression is associated with reduced or reversed optimism bias (sometimes called "depressive realism" [4] though the interpretation is contested); clinical anxiety can also reduce or reverse the bias.
Broad Use¶
Optimism bias operates across health behavior (smokers, drinkers, and sedentary individuals substantially underestimate their personal risk of disease while accurately estimating average-smoker risk; this gap substantially reduces self-perceived benefit of behavior change) [7], entrepreneurship and investing (new venture founders systematically overestimate success probability, with one frequently cited finding that ~80% believe their success is likely while actual success rates are much lower; investment decisions exhibit valence-asymmetric updating in well-documented ways) [8], project management and planning (Flyvbjerg's extensive research on megaproject cost and schedule overruns — consistently running 50–100%+ over budget and 20–40%+ late — attributes much of the gap to planning-stage optimism bias; "reference-class forecasting" is an explicit debiasing technique), insurance and financial planning (under-purchase of relevant insurance types; under-saving for retirement), medical decision-making (patients underestimate post-surgical complication risk; oncology patients overestimate prognosis in terminal situations), criminal behavior (offenders underestimate apprehension probability), and risk communication (public health campaigns often struggle because the "average person at risk" message is absorbed without self-application — "it won't happen to me" functions as a shield against the statistical information).
Clarity¶
Optimism bias is sometimes conflated with general positive thinking or with dispositional optimism. The clarifying distinction is that optimism bias is a specific asymmetry in probability estimation and belief updating, documented by comparing self-estimates to base rates and to estimates for others. Weinstein's original 1980 paper used a methodology that has remained definitive: present subjects with a list of outcomes; have them rate "compared to other people like me, my chances of experiencing X are: much lower / lower / average / higher / much higher"; statistical analysis of the full response distribution reveals systematic deviations from the mathematically required average of "average." This methodology isolates the self-specific asymmetry from general optimism, confirms the bias is in probability judgment rather than in valence preference, and permits precise quantification [1]. Rigorous use of the construct tracks this methodological specificity rather than invoking "optimism bias" for any positive expectation.
Manages Complexity¶
Optimism bias, viewed functionally rather than as pure error, has complexity-management value: it buffers decision-making against excessive risk-weighted paralysis. If agents calibrated perfectly to all possible negative outcomes, the aggregate risk load of modern life — health risks, financial risks, relationship risks, environmental risks — would make many consequential actions appear net-negative. Moderate optimism bias allows the agent to pursue actions whose expected value is positive by underweighting low-probability severe negative outcomes. This has motivational, energetic, and mental-health benefits that substantial research has documented. The cost is miscalibration in domains where accurate risk assessment is essential (medical decisions, financial planning, project timelines). The structural lesson is that the mechanism is not uniformly beneficial or harmful but requires contextual calibration: domains where motivation and energy are the limiting factor benefit from the bias; domains where accurate risk assessment is the limiting factor are harmed by it.
Abstract Reasoning¶
Optimism bias instantiates a broader structural pattern: valence-asymmetric updating in an agent whose belief dynamics are partly shaped by motivational stakes. This pattern recurs across domains and levels of analysis. In organizations: plans are updated more readily when evidence supports them than when evidence undermines them, producing the well-documented "planning fallacy" in project management. In scientific communities: findings supporting popular theories are more readily incorporated than findings that would undermine them, producing slow updating of theoretical commitments even under accumulating evidence. In political belief formation: partisan belief updating is asymmetric — information supporting one's political priors is incorporated; disconfirming information is dismissed or attributed to bias. In machine-learning systems trained with reinforcement learning from human feedback, reward-model mis-specification can produce asymmetric updating that looks structurally similar to optimism bias. In each case the pattern is: a belief system with motivational stakes updates asymmetrically based on valence, producing systematic miscalibration in predictable directions. The structural lesson is that belief-updating is rarely valence-neutral when the agent has skin in the game, and that designing for calibration requires explicit mechanisms to counteract the asymmetry.
Knowledge Transfer¶
| Role in Optimism Bias | Role in Flyvbjerg's Reference-Class Forecasting |
|---|---|
| Individual probability estimate biased toward positive outcome | Project planner's initial schedule/cost estimate biased toward optimistic completion |
| Base-rate information about similar outcomes | Historical distribution of schedule/cost outcomes for comparable past projects |
| Self-specificity of the bias | "Inside view" — planner's detailed knowledge of this project's features |
| Other-estimate as calibrated | "Outside view" — treating this project as one instance drawn from the reference class |
| Debiasing: exposure to base rates | Reference-class forecasting — explicit statistical forecasting from comparable past projects |
| Residual bias after debiasing | Empirical reduction but non-elimination of schedule/cost overruns |
| Motivational origin of the bias | Incentive structure of project proposal and funding |
Bent Flyvbjerg and colleagues' reference-class forecasting methodology is a direct operationalization of optimism-bias debiasing in project management. The core insight — and the mechanism by which the methodology works — is that planners evaluating a new project use an "inside view" (detailed knowledge of this project's features, plans, and teams) that is structurally optimistic because it is self-specific; an "outside view" (this project as one instance drawn from a reference class of comparable past projects) produces more calibrated estimates because it uses base-rate information the inside view systematically under-weights. The transfer is tight: Flyvbjerg's extensive empirical work on megaproject overruns — Boston's Big Dig, the Sydney Opera House, Denver International Airport, and hundreds of other cases — demonstrates that systematic application of reference-class forecasting reduces (but does not eliminate) cost and schedule overruns by 10–30%. This validates both the existence of optimism bias at the organizational-planning level and the specific debiasing technique derived from its mechanism. The transfer also illuminates two design lessons: first, that effective debiasing requires institutional structure (explicit methodology requirements, procurement rules that mandate reference-class analysis) rather than just cognitive awareness; second, that the residual bias after debiasing suggests motivational factors persist even after informational gaps are closed — planners may know the reference-class estimate is more accurate while still preferring the inside-view estimate because it supports project approval and funding.
Examples¶
Formal/Abstract — Canonical Research: Weinstein's 1980 foundational study Unrealistic Optimism About Future Life Events [1]. Subjects were college students given a list of 42 outcomes — some positive (good marriage, career success, homeownership, long life), some negative (divorce, chronic illness, car accident, unemployment). They were asked to rate, for each outcome, "compared to other [name of college] students, my chances of experiencing this outcome are: much below average / below average / average / above average / much above average." The rational distribution requires that the average response for each outcome equal "average." The empirical finding: for positive outcomes, the distribution was systematically shifted above average; for negative outcomes, systematically shifted below average. Sharot and colleagues' subsequent neuroimaging work (2007)[9] identified specific neural correlates: when given better-than-expected information, subjects incorporated normally; when given worse-than-expected information, incorporation was attenuated, tracking rostral anterior cingulate and lateral prefrontal cortex activity. This established the mechanism at both behavioral and neural levels[2].
Applied/Industry — Planning Fallacy in Megaprojects: Flyvbjerg's extensive research on megaproject cost and schedule overruns [10]. Infrastructure projects (Boston Big Dig, Sydney Opera House, Denver International Airport) consistently run 50–100%+ over budget and 20–40%+ late[10]. Flyvbjerg attributes the gap to planning-stage optimism bias: planners use an "inside view" (detailed knowledge of this project's specific features) that systematically under-weights the base rate of complications from reference-class comparable projects. Reference-class forecasting — explicit statistical debiasing using historical outcome distributions — reduces but does not eliminate overruns by 10–30%. This validates both the existence of optimism bias at organizational-planning scale and the specific debiasing mechanism derived from its structure[11].
Mapped Transfer¶
| Optimism Bias Role | Flyvbjerg Reference-Class Forecasting |
|---|---|
| the comparative-risk underestimate | Inside-view project estimate biased toward optimistic completion |
| the self-other asymmetry | Detailed project knowledge (self) vs. reference-class historical outcomes (other) |
| the planning-fallacy delay | Schedule/cost overruns as behavioral manifestation |
| Base-rate neglect | Planners systematically under-weight reference-class distribution |
| Debiasing mechanism | Explicit statistical forecasting from comparable past projects |
| Residual bias | 10–30% reduction in overruns, non-elimination—motivational persistence |
Structural Tensions and Failure Modes¶
T1 — Motivational benefit vs. calibration cost. Modest optimism bias has well-documented motivational, emotional, and mental-health benefits; severe optimism bias produces calibration failures in high-stakes decisions. The tension is that the same mechanism produces both. Individuals whose optimism bias is suppressed (depressed or anxious populations) often make more calibrated probability estimates but experience reduced motivation and wellbeing [12]. The failure mode is either direction: clinical interventions that suppress the bias may improve calibration while reducing motivation; interventions that promote the bias may improve mood while worsening real-world decisions. This is foundational to understanding when the bias is adaptive versus pathological.
T2 — Institutional amplification of individual bias. When optimism bias aggregates across many individuals within an organization or society, the institutional-level miscalibration can be much larger than the individual-level miscalibration. Entrepreneurs self-select into starting businesses (filtering for higher-than-average optimism) [8]; within those firms, individual optimism aggregates into team optimism; team optimism aggregates into firm-level forecasts and strategies. The result is that institutional decisions can exhibit substantially worse calibration than any individual's would — an emergent property of the selection process. The failure mode is institutional decision-making that treats aggregate individual optimism as independent evidence rather than as the correlated aggregation of a shared bias.
T3 — Information-environment exploitation. Actors who benefit from others' optimism bias have incentive to reinforce it [13]. Investment products, entrepreneurial-ecosystem narratives, lottery and gambling industries, extended-warranty and insurance marketing, and political campaigning all at various points target or depend on optimism bias in their audiences [14]. The tension is that the same bias that serves individuals motivationally is the vulnerability that adversarial actors exploit. The failure mode is information environments in which optimism-bias exploitation is prevalent, producing systematic welfare losses at population scale even when individual exploitation is too subtle to detect.
T4 — Asymmetric information responsiveness as design consideration. Sharot's work established that optimism bias is primarily an asymmetry in response to new information rather than simply an initial estimate bias. This matters for debiasing: interventions that provide information about base rates can partially correct, but only if the information is of the kind the biased updating will actually incorporate. Information that supports the optimistic prior updates beliefs; information that undermines it is asymmetrically under-weighted. The failure mode is debiasing interventions that assume information provision is sufficient, ignoring that the updating asymmetry itself attenuates the intervention's effect. More effective debiasing requires both information and structural mechanisms that force symmetric incorporation.
T5 — Optimism bias as adaptive versus maladaptive [15]. Taylor and Brown's 1988 positive-illusions theory argues that mild optimism bias is mental-health-protective, promoting motivation, resilience, and psychological wellbeing[15]. Conversely, depressive realism — the observation that depressed individuals make more accurate probability estimates and more realistic self-assessments than non-depressed individuals — suggests that the absence of optimism bias correlates with clinical depression, not clarity[4]. This creates a paradoxical tension: the "bias" may be psychologically functional, and its removal correlates with psychiatric distress. The failure mode is cognitive interventions that treat optimism bias as purely deleterious, suppressing it without attending to its wellbeing correlates; or, conversely, clinical treatments that inadvertently reinforce unrealistic self-assessment as part of symptom reduction.
T6 — Universal bias versus cross-cultural variation [16]. [16] Heine and Hamamura's 2007 cross-cultural meta-analysis found that East Asian samples (Japanese, Korean, Chinese) show substantially less optimism bias than North American and Western European samples, with some evidence of reversed bias (modesty bias)[16]. This challenges the universality claim and suggests optimism bias is culturally moderated — possibly by independent-versus-interdependent self-construals, collectivist norms favoring self-effacement, or epistemic cultural values. The failure mode is treating optimism bias as a universal cognitive architecture when it is, in fact, substantially shaped by cultural context. Debiasing interventions designed for North American individualist samples may be ineffective or counterproductive in other cultural contexts.
Structural–Framed Character¶
Optimism Bias is a hybrid on the structural–framed spectrum. Part of it is a bare pattern that means the same thing in any field — an asymmetry in how an agent updates probability estimates; part of it is a frame, a vocabulary and set of assumptions, inherited from psychology and the behavioral sciences. The frame here is substantial, though a structural core exists.
The structural core is formal and field-neutral: an agent maintains probability estimates about outcomes and updates them asymmetrically, weighting good news more than bad news, so that the rule itself could be stated for any belief-updating system. But the substance of the concept leans on its home discipline. Naming it a "bias" carries an evaluative judgment — that the estimates are systematically off relative to base rates and so are errors, not merely features — and that judgment travels into every application, from a patient underrating illness risk to an investor overrating returns to a project team underestimating a schedule. The idea cannot be fully defined without the apparatus of beliefs, self-relevant outcomes, and human (or human-like) cognition that psychology supplies. So while you genuinely recognize a recurring updating pattern already present in the data, you also import a corrective standpoint about how beliefs ought to track reality, placing it in the mid-spectrum with a real structural skeleton inside a behavioral-science frame.
Substrate Independence¶
Optimism Bias is a narrowly substrate-independent prime — composite 2 / 5 on the substrate-independence scale. It names an asymmetric belief-updating tendency that favors positive outcomes, and its signature is built around 'individuals' and the 'self', which inherently narrows its scope to minds. The examples are uniformly cognitive and behavioral, drawn from psychology and behavioral economics. Any extension to organizational or social systems is metaphorical, so this reads as a specific cognitive bias rather than a structural pattern that travels across substrates — its cross-substrate leverage is limited.
- Composite substrate independence — 2 / 5
- Domain breadth — 2 / 5
- Structural abstraction — 3 / 5
- Transfer evidence — 1 / 5
Relationships to Other Primes¶
Parents (1) — more general patterns this builds on
-
Optimism Bias is a kind of Bias
Optimism bias is a specialization of bias. Specifically, it instantiates the systematic-displacement-from-the-true-value pattern in the self-prediction subclass: individuals' probability estimates for personal positive outcomes are consistently above base rates, estimates for personal negative outcomes consistently below, and belief updates asymmetric. It exhibits bias's defining signature -- a sign and direction surviving repeated estimation -- with the offset traced to motivated self-relevant projection and stronger for self than for comparable others, distinguishing it within the broader bias catalog.
Path to root: Optimism Bias → Bias
Neighborhood in Abstraction Space¶
Optimism Bias sits among the more crowded primes in the catalog (22nd percentile for distinctiveness): several abstractions describe nearly the same structure, so a description that fits it will tend to fit its neighbors too — transporting it usually means disambiguating within this family rather than landing on it exactly.
Family — Commitment, Path-Dependence & Optionality (14 primes)
Nearest neighbors
- Regret — 0.83
- Opportunity Asymmetry — 0.82
- Temporal Inconsistency and Preference Reversals — 0.82
- Sunk Cost and Irreversible Commitment — 0.81
- Fundamental Attribution Error — 0.81
Computed from structural-signature embeddings · 2026-05-29
Not to Be Confused With¶
Optimism Bias must be distinguished from Confirmation Bias, its closest neighbor (similarity 0.745), because they represent asymmetries in different stages of information processing. Confirmation Bias is asymmetric search and interpretation of information — a person seeks evidence that supports their prior belief, interprets ambiguous evidence as supportive, and avoids or dismisses disconfirming evidence. The asymmetry is in the selection and framing of existing information to bolster belief maintenance. An investor believing a stock is undervalued selectively reads bullish analyst reports, dismisses bearish ones as "too pessimistic," and interprets the stock's recent price dip as an opportunity (confirmation) rather than as disconfirming. Optimism Bias, by contrast, is asymmetric belief updating in response to new information — the agent processes both favorable and unfavorable information, but incorporates favorable information more fully into probability estimates than unfavorable information. A person estimates a 30% chance of getting a promotion; upon learning positive performance feedback, the estimate moves to 40%; upon later learning that a colleague is also being considered, the estimate moves to 35% (less movement away from the prior). Both biases distort reasoning, but along different axes: confirmation bias distorts which information is attended to; optimism bias distorts how information is weighted in updating. An investor exhibiting confirmation bias selects favorable news; an investor exhibiting optimism bias reads both favorable and unfavorable news but over-weights the favorable in revising expectations. The two can coexist — an agent might seek confirming information AND over-weight favorable information — but they are distinct mechanisms operating at different points in the cognitive process.
Nor is Optimism Bias identical to Selection Bias, though both produce skewed distributions. Selection Bias is a methodological and structural phenomenon arising from the process by which units enter a data set or study — if participation is voluntary, healthier individuals may be overrepresented; if historical data is gathered, survivors are overrepresented and non-survivors are absent. Selection Bias is not an agent's cognitive belief-updating mechanism but rather a property of the data itself and how it was selected. Optimism Bias, by contrast, is a cognitive asymmetry within an individual agent — the way that specific person updates their probability estimates about future outcomes. An individual suffering selection bias is observing a skewed sample from which they may draw inference; an individual suffering optimism bias is making asymmetric probability estimates. The distinction matters practically: selection bias is addressed by statistical techniques (weighting, stratification, instrumental variables) that correct for the structure of the data; optimism bias is addressed by cognitive interventions (reference-class forecasting, explicit base-rate communication, institutional structure that forces symmetric incorporation of good and bad news). A medical researcher observing that treated patients have better outcomes than untreated patients may be observing selection bias (only sicker patients were treated); a patient estimating post-treatment recovery probability may be exhibiting optimism bias (overweighting optimistic recovery scenarios). Selection bias contaminates data; optimism bias contaminates belief.
Finally, Optimism Bias is distinct from the Dunning-Kruger Effect, though both involve overestimation. Dunning-Kruger concerns self-assessment of current competence — individuals with low actual competence tend to overestimate their competence (a metacognitive failure in which incompetence prevents recognition of incompetence), while highly competent individuals may underestimate their competence (assuming others are equally skilled). The domain is current ability; the asymmetry is that low-competence agents lack the skill to recognize skill deficits. Optimism Bias concerns probability estimates of future outcomes — individuals overestimate the likelihood that they will experience positive outcomes (promotions, good health, marriage) and underestimate the likelihood of negative outcomes (illness, divorce, unemployment). The domain is future events; the asymmetry is in belief updating and outcome expectation. An entrepreneur with low entrepreneurial skill might exhibit Dunning-Kruger (overestimating their own competence) and separately exhibit Optimism Bias (overestimating their startup's success probability given their actual (low) skill level). A student might exhibit Dunning-Kruger on an exam (low competence but high confidence in answers) and separately exhibit Optimism Bias about exam performance (thinking "I'll do better than average" despite low study). The mechanisms are different: Dunning-Kruger is a metacognitive limitation; Optimism Bias is an asymmetric belief-updating process. Both lead to overconfidence, but in different cognitive domains and through different architectures.
Solution Archetypes¶
Solution archetypes in the catalog that build on this prime — directly (this prime is a source ingredient) or as a related prime.
Built directly on this prime (1)
Also a related prime in 2 archetypes
Notes¶
DP-16 batch G3, IDs D16-120 through D16-135 (16 canonical citations). Deviation from spec (target 15 IDs): added D16-135 to resolve inline-anchor collision on D16-127 (T5 inline + taylor1988 footer) and maintain unique parity. The Flyvbjerg megaproject reference-class forecasting transfer operationalizes optimism-bias mechanism at institutional scale with empirically validated debiasing outcomes. Cross-referenced to dunning_kruger_effect — related but distinct constructs (DK: competence self-assessment; optimism bias: outcome probability estimation). Tensions T5 and T6 emphasize adaptive-versus-maladaptive and cultural-variation dimensions, respectively. Structure follows DP-02 template with 6 role-phrases in Structural Signature, 2 Examples subsections with Mapped Transfer table, and 6 Tensions. No Solution Archetypes authored (deferred to Pass B). ID anchors verified for dual-placement parity: 16 unique inline FACT-D16-* anchors = 16 unique footnote bodies.
References¶
[1] Weinstein, N. D. (1980). Unrealistic optimism about future life events. Journal of Personality and Social Psychology, 39(5), 806–820. ↩
[2] Sharot, T. (2011). The Optimism Bias: A Tour of the Irrationally Positive Brain. Pantheon Books. ↩
[3] Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263–291. ↩
[4] Alloy, L. B., & Abramson, L. Y. (1979). Judgment of contingency in depressed and nondepressed subjects: Sadder but wiser? Journal of Experimental Psychology: General, 108(4), 441–485. ↩
[5] Lench, H. C., Bench, S. W., Darbor, K. E., & Moore, M. (2014). The downstream consequences of emotion regulation: A behavioral approach. Cognition & Emotion, 29(3), 453–470. ↩
[6] Shepperd, J. A., Klein, W. M., Waters, E. A., & Weinstein, N. D. (2013). Taking stock of unrealistic optimism. Perspectives on Psychological Science, 8(4), 395–411. ↩
[7] Weinstein, N. D. (1987). Unrealistic optimism about susceptibility to health problems: Conclusions from a community-wide sample. Journal of Behavioral Decision Making, 1(4), 221–235. ↩
[8] Camerer, C., & Lovallo, D. (1999). Overconfidence and excess entry: An experimental approach. American Economic Review, 89(1), 306–318. ↩
[9] Sharot, T., Riccardi, A. M., Raio, C. M., & Phelps, E. A. (2007). Neural mechanisms mediating optimism bias. Nature, 450, 102–105. ↩
[10] Flyvbjerg, B. (2003). Megaprojects and Risk: An Anatomy of Ambition. Cambridge University Press. ↩
[11] Buehler, R., Griffin, D., & Ross, M. (1994). Exploring the "planning fallacy": Why people underestimate their task completion times. Journal of Personality and Social Psychology, 67(3), 366–381. ↩
[12] Moore, D. A., & Healy, P. J. (2008). The trouble with overconfidence. Psychological Review, 115(2), 502–517. ↩
[13] Sunstein, C. R. (2002). Risk and Reason: Safety, Law, and the Environment. Cambridge University Press. ↩
[14] Ariely, D. (2008). Predictably Irrational: The Hidden Forces That Shape Our Decisions. HarperCollins. ↩
[15] Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social psychological perspective on mental health. Psychological Bulletin, 103(2), 193–210. ↩
[16] Heine, S. J., & Hamamura, T. (2007). In search of East Asian self-enhancement. Personality and Social Psychology Review, 11(1), 4–27. ↩
[17] Klar, Y., & Giladi, E. E. (1997). Are most people happier than their peers, or are they just happy? Journal of Personality and Social Psychology, 73(4), 599–616.