The pattern in one claim
Ask people whether they are biased. Many will admit bias exists as a human problem and still place themselves on the less-biased side of whatever comparison group you name. That asymmetry is the blind spot: bias is visible as a trait of others and underweighted as a trait of the self.
Pronin, Lin and Ross (2002), published in Personality and Social Psychology Bulletin, measured exactly that self-versus-other gap across several biases and several comparison targets. The effect is not a claim that everyone is equally wrong. It is a claim about asymmetric perception.
Follow-up papers kept the same core asymmetry while changing the evidence people were allowed to use for self versus other, which is why the timeline on this page is part of the delivery rather than a bibliography dump.
Research timeline (dated)
Year | Entry | What it added
2002 | Pronin, Lin and Ross, Personality and Social Psychology Bulletin | Named and measured the bias blind spot: self-ratings of susceptibility below ratings of average Americans, classmates, and airport travelers; learning a bias description did not erase the gap
2004 | Pronin, Gilovich and Ross, Psychological Review | Linked divergent perceptions of bias in self versus others to objectivity illusions and naive realism in conflict
2005 | Ehrlinger, Gilovich and Ross, Personality and Social Psychology Bulletin | Extended assessments of bias in self versus others, including how people use different evidence for self and other judgments
2007 | Pronin and Kugler, Journal of Experimental Social Psychology | Tested the introspection illusion as a source of the blind spot: valuing private thoughts for the self while reading others through behavior
2007 | Pronin, Trends in Cognitive Sciences | Review tying misperception of bias to introspection habits and naive realism for a wider scientific audience
2015 | Scopelliti and colleagues, Management Science (individual-difference measure line) | Validated a reliable bias-blind-spot scale and linked higher scores to self-enhancement and resistance to advice or debiasing
What the 2002 studies asked
Participants read short descriptions of common biases, then rated how much they personally showed each bias compared with others. The consistent result was self-ratings lower than ratings of the average person, peers, or travelers in an airport sample. Learning about a bias did not automatically erase the gap; people could understand the concept and still exempt themselves.
That matters for media arguments. Being able to name "confirmation bias" or "motivated reasoning" in an opponent is not evidence you have escaped the same machinery. The 2002 paper also discussed how availability of one's own good intentions and self-enhancement motives can feed the asymmetry.
In practice that means a commenter can correctly spot motivated sampling on the other side and still run a quieter version of the same habit: reading only the replies that flatter their frame, then treating the resulting calm as proof of objectivity. The blind spot is compatible with partial accuracy about others.
Two mechanisms Pronin highlights
Introspection illusion: for yourself, you look inward at thoughts and feelings that often contain no trace of the bias, then treat that clean introspective report as proof you are unaffected. For others, you look at behaviour and outcomes, which are easier to read as biased. Pronin and Kugler (2007) treat that asymmetry in evidence as a driver of the blind spot.
Naive realism: your perception feels like a direct window onto the world. When someone disagrees, the default story is that they are uninformed, irrational or biased, not that you both have partial views. Pronin's 2007 Trends in Cognitive Sciences review and the 2004 Psychological Review paper with Gilovich and Ross place this mechanism next to conflict and misunderstanding.
Taken together, the 2002 effect, the 2004 objectivity paper, the 2005 cue-split findings, and the 2007 introspection tests form a single story: people are not merely boastful. They are using different evidence streams for self and other, then treating the mismatch as moral clarity.
How it shows up in comment threads
A familiar script: "They are trapped in their bubble; I just follow the evidence." The blind spot does not prove the speaker is wrong on the facts. It predicts that both sides will tell the same story about the other side.
Measurement work in the mid-2010s (Scopelliti and colleagues) treats the blind spot as an individual difference that can predict who shrugs off advice or debiasing prompts. That is useful for media literacy without turning every disagreement into a diagnosis.
Ehrlinger, Gilovich and Ross (2005) sharpen a related point: people often lean on different cues when judging bias in themselves than when judging it in others. Self-judgments lean on private intentions; other-judgments lean on observed outcomes. That split helps explain why "I meant well" feels like a complete defense to the speaker and like evasion to the audience.
Pair this with how platforms reward rage bait when the disagreement is being amplified for engagement, and with belief perseverance when a first take refuses to update after corrections.
A check that does not require false humility
You do not need to declare yourself "very biased" to reduce the blind spot. A narrower habit works better: for any bias you diagnose in an opponent, write one concrete way the same bias could distort your own selection of sources, frames or counterexamples. If you cannot name one, you are probably still exempting yourself.
This page is a definition, mechanism map, and dated research line, not a personality test. It will not tell you who is right in a live dispute.
It will tell you when a debate has switched from evidence about a claim to evidence about who is allowed to see bias at all.