Egocentric bias
We anchor on our own perspective and over-weight our own contribution.
What it means
A family of distortions in which people rely too heavily on their own point of view when judging the world, recalling events, or apportioning credit. Because our own thoughts, sensations, and contributions are vividly and continuously accessible, they crowd out the less available perspectives and inputs of others. It shows up as overclaiming responsibility for joint outcomes, assuming others share our knowledge and feelings, and remembering ourselves as more central to past events than we were. The bias is not mere selfishness — it follows from the asymmetry of information we have about ourselves versus others. It underlies disputes over fairness, the difficulty of genuine perspective-taking, and many communication failures.
Why it happens
Your own thoughts, effort, and sensations stream through consciousness continuously, while other people's are inferred, sampled, and quickly forgotten. This availability gap does most of the work: when you total up who did what, your own acts come to mind first and easiest, so they weigh more. A second route is anchoring. To judge another's viewpoint we start from our own and adjust, but the adjustment is effortful and usually stops short, leaving the estimate tethered to the self. Because the distortion is informational rather than motivational, it moves with attention. Make a coworker's contributions salient, by listing them or having people recall them first, and the over-claiming shrinks or even reverses. The self looms large not because we crave the credit but because we cannot help noticing ourselves.
What the evidence shows
Ross and Sicoly's original studies spanned married couples, basketball players, and discussion groups; across all of them people recalled their own contributions more readily and claimed more responsibility than partners granted them. The signature result is arithmetic: ask each member of a pair or team what share they contributed, and the figures sum past 100 percent. The pattern has held up and sharpened. Schroeder, Caruso and Epley (2016) found that over-claiming grows with group size, on multi-author papers and larger MBA study teams individual claims rose as more people were added, precisely because each extra collaborator's work is harder to keep in view. Savitsky and colleagues showed that unpacking a partner's role into itemized subtasks raises the credit people grant them, confirming that availability, not motive, drives the effect.
Where it shows up
The bias quietly fuels fairness disputes. Two partners, two coworkers, or two co-founders can each feel they carry the heavier load and each feel underappreciated, with no one lying. In communication it becomes the curse of knowledge: writers, teachers, and engineers assume readers share context that is vivid only to them, so instructions skip steps and briefings omit the obvious. It shapes negotiation, where each side anchors on its own concessions, and performance reviews, where self-ratings run high. It also distorts forecasting: predicting how a colleague or customer will react, we over-project our own preferences and knowledge. Wherever credit, blame, or another mind must be estimated from a vantage point saturated with the self, the same tilt appears.
Limits and debiasing
Egocentric bias is not simple self-flattery. Ross and Sicoly found the tilt did not vanish when the joint product was judged a failure; people also over-recall their role in bad outcomes, because availability rather than vanity is doing the work. Debiasing is possible but carries a sting. Caruso, Epley and Bazerman (2006) showed that prompting members to consider others' contributions reduces over-claiming, yet it also leaves high contributors less satisfied and less keen to collaborate again, having now noticed how much they carried. The practical move is to unpack and name each person's specific work before allocating credit, and, when gauging another's view, to ask rather than imagine, since getting a perspective beats guessing at it, while accepting that fairer accounting can feel worse.
Examples
Each spouse, asked what share of the housework they do, gives a figure that when summed across the couple exceeds 100%.
Tapping out a familiar tune, you are sure your friend can hear it — the melody blaring in your head makes the bare knocks seem obvious to everyone but them.
A developer writes setup instructions that skip the three steps she does automatically, because her own knowledge is far more available to her than a newcomer's ignorance.
Two co-founders splitting equity each tally the late nights and risks they personally took; both private ledgers land above half, and the gap only surfaces when a term sheet forces one number.
A product manager, certain the redesigned dashboard is self-explanatory, forecasts that users will need no guide; support tickets then pile up from people who never shared the context that made every button obvious to her.
First described in Ross & Sicoly (1979).
Key references
- Schroeder, J., Caruso, E. M., & Epley, N. (2016). Many hands make overlooked work: Over-claiming of responsibility increases with group size. Journal of Experimental Psychology: Applied, 22(2), 238-246. doi.org/10.1037/xap0000080
- Caruso, E. M., Epley, N., & Bazerman, M. H. (2006). The costs and benefits of undoing egocentric responsibility assessments in groups. Journal of Personality and Social Psychology, 91(5), 857-871. doi.org/10.1037/0022-3514.91.5.857
- Savitsky, K., Van Boven, L., Epley, N., & Wight, W. M. (2005). The unpacking effect in allocations of responsibility for group tasks. Journal of Experimental Social Psychology, 41(5), 447-457. doi.org/10.1016/j.jesp.2004.08.008
- 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. doi.org/10.1037/0022-3514.87.3.327
- Ross, M., & Sicoly, F. (1979). Egocentric biases in availability and attribution. Journal of Personality and Social Psychology, 37(3), 322-336. doi.org/10.1037/0022-3514.37.3.322