← The Casebook

A confession in two voices

Egocentric Bias

Egocentric Bias

What you confess

The tendency to overweight one's own perspective, memory, and contribution when judging shared events, so that people systematically over-credit themselves and discount collaborators.

How it sounds when you do it

Watch for moments when the people in a shared project, when asked separately, give self-contribution estimates that sum to well over 100% — each person can vividly recall their own effort but not their collaborators'. It is especially likely in larger teams and when judgments are made fast or under pressure, and it can run in either direction (over-claiming abundant contributions, under-claiming scarce ones).

A confession on record

Banting vs Macleod — 1923 Nobel Prize in Medicine (insulin) · 1923

Frederick Banting, sharing the 1923 Nobel Prize with John Macleod for the discovery of insulin, was so incensed at Macleod's share of the credit that he boycotted the Stockholm ceremony and split half his prize money with lab coworker Charles Best; Macleod reciprocally omitted Banting from speeches. Caruso, Epley & Bazerman (2006) open their paper with this as a real instance of egocentric credit-claiming and resulting conflict.

Who first named this

Michael Ross & Fiore Sicoly, 1979Egocentric Biases in Availability and Attribution

The counsel

Two complementary mechanisms are proposed. (1) Differential availability (Ross & Sicoly, 1979): one's own contributions, thoughts, and efforts are encoded and recalled more easily than others', so when reconstructing a shared event the self looms larger. (2) Egocentric anchoring-and-adjustment (Epley et al., 2004): when judging others, people anchor on their own perspective and adjust insufficiently, which is why the bias grows under time pressure and shrinks with accuracy incentives. Group-size moderation (Schroeder et al., 2016) follows directly: larger groups supply more others' contributions to overlook, so summed over-claiming increases.

Some treat egocentric over-claiming as partly motivational (self-enhancement / self-serving) rather than purely cognitive-availability driven; Ross & Sicoly themselves noted self-presentation cannot fully explain results because the bias persisted for negative outcomes (attenuated, not reversed). Whether the core driver is availability, anchoring, or motivated reasoning remains debated, and the accounts are not mutually exclusive.

The evidence answers

  • Each spouse claimed more responsibility than the partner attributed to them; combined self-claimed responsibility exceeded the logical 100% ceiling. The bias appeared for both positive and (attenuated) negative activities.

    Michael Ross, Fiore Sicoly, 1979 · Combined self-attributed responsibility M ~= 103.1% (figure as reported in the 1979 paper and cited by Schroeder, Caruso & Epley, 2016) · Married couples (Experiment 1 of five experiments; exact n not confirmed from retrieved text)

  • Replicated over-claiming; prompting people to consider others' contributions reduced self-allocations but, in cooperative groups, left high contributors less satisfied and less interested in future collaboration — a 'cost' of debiasing.

    Eugene M. Caruso, Nicholas Epley, Max H. Bazerman, 2006 · not reported (specific d/eta-squared not confirmed from retrieved text) · Co-authors of published OB articles (Study 1) plus three lab experiments

  • Self-claimed responsibility rose with group size (more collaborators to overlook); explicitly considering others' contributions reduced over-claiming. A manipulation that focused attention on self produced summed responsibility M = 122.8% (SD = 29.5%) versus lower in other-focused conditions.

    Juliana Schroeder, Eugene M. Caruso, Nicholas Epley, 2016 · Group-size effect beta = 0.19, p < .001; self-focus vs combined other conditions t(74) = 3.21, p < .01, d = 0.75 · Field study 2: 699 MBA students; plus co-authorship field data and two lab experiments

  • Estimates of others' perceptions started from the self and adjusted serially; egocentric bias increased under time pressure and decreased with accuracy incentives, supporting an anchoring-and-adjustment mechanism.

    Nicholas Epley, Boaz Keysar, Leaf Van Boven, Thomas Gilovich, 2004 · not reported (specific effect sizes not confirmed from retrieved text) · Five studies (sample sizes not confirmed from retrieved text)

The record

Egocentric bias is the documented tendency to anchor judgments of shared events on one's own perspective and more-available memories, leading people to claim more responsibility for joint outcomes than is logically possible. Michael Ross and Fiore Sicoly established it experimentally in 1979 across couples, sports teams, and lab groups; the signature finding is that summed self-attributed responsibility exceeds 100% (about 103% for married couples). Later work (Caruso/Epley/Bazerman 2006; Schroeder/Caruso/Epley 2016) replicated the effect and showed over-claiming grows with group size, while the mechanism is best understood as egocentric anchoring-and-adjustment plus differential availability of self-relevant information.

Absolution — replication

The charge stands.

One of the better-replicated effects in the field — you do not get to plead it away.

The core over-claiming phenomenon (summed self-attributed responsibility exceeding 100%) has been reproduced across decades, populations, and domains: couples (Ross & Sicoly 1979, M~103%), co-authors and MBA study groups (Caruso et al. 2006; Schroeder et al. 2016, M up to ~123% under self-focus, group-size beta=0.19 p<.001), athletics (Brawley 1984; Forsyth & Schlenker 1977), and fundraising (Zander 1971). What remains contested is the exact mechanism (availability vs anchoring vs motivation) and the magnitude/moderators rather than the existence of the effect.

Your penance

Before estimating your own share, explicitly list each collaborator's specific contributions first (a 'unpacking others' / consider-the-opposite prompt). Caruso et al. and Schroeder et al. show this reliably shrinks over-claiming — though be aware it can dampen high contributors' satisfaction, so pair it with genuine recognition.

In Schroeder et al. (2016), prompting people to consider others' contributions reduced over-claiming; a self-focus condition produced summed responsibility M=122.8% vs lower other-focused conditions (t(74)=3.21, p<.01, d=0.75). Epley et al. (2004) show accuracy incentives also reduce egocentric adjustment errors.

Grounding confidence: strong. The seminal effect, its direction, and its replication across domains are well grounded in retrieved primary sources. Some precise figures could not be fully verified against the original text: Ross & Sicoly's exact per-experiment sample sizes were not extractable (the 1979 PDF returned 403), and the M~103.1% couples figure is sourced from Schroeder et al. (2016) citing Ross & Sicoly rather than read directly from the 1979 paper. Specific effect sizes for Caruso et al. (2006) and Epley et al. (2004) were not located in retrieved text and are marked 'not reported'. The mechanism (availability vs anchoring vs motivation) is genuinely debated. The Banting/Macleod case is a real historical event used illustratively by the authors, not itself an experimental measurement of bias.

Cross-referenced confessions

File your own case

Open the same case on your own draft.

Paste a memo, a research draft, or a strategy argument. It is scored against all 175 cards, and the strongest two or three risks come back with the evidence quoted and one practical next check.

Open a case on your draft →