Why we think we read others well
The confidence with which we guess what someone is thinking has almost no relation to our accuracy. The paradigm that showed it dates from the 1990s.
A feeling that is never checked
We constantly form hypotheses about what others are feeling, and we almost never check them. A conversation ends, each leaves with their reading, and nothing in the ordinary situation says which one was right.
This absence of feedback explains a good part of our assurance. A capacity never put to the test does not get corrected, and accumulated experience reinforces confidence without improving precision.
The empathic accuracy protocol
William Ickes developed in the 1990s a method that gets around the problem. A conversation between two people is filmed. Each then watches the recording alone and notes what they were thinking at each moment. Then the other is asked to guess those thoughts at the same moments.
The comparison provides an accuracy score independent of any self-assessment. The central result is consistent: the correlation between believing oneself a good reader and guessing correctly is weak.
What really improves the reading
The factors that increase accuracy are less flattering than intuition. Familiarity with the person helps, but less than one thinks, and it can get in the way when it installs fixed expectations. Motivation to understand, for its part, produces a measurable effect: when the situation gives an interest in seeing correctly, accuracy rises.
The most powerful factor remains the information available. Seeing the context, hearing the tone, knowing what happened before improves the reading far more reliably than any personal disposition.
The case of highly perceptive people
Some people do pick up more signals than average: a change of rhythm in the voice, a tension in a room, a discreet withdrawal. This capacity is real and shows itself very early in an interaction.
It does not come with better interpretation. Perceiving a signal and identifying its cause are two distinct operations, and the second rests on information that perception does not supply. A very sensitive antenna badly aimed simply picks up more noise.
One robust rule
The whole of this literature leads to a modest practical conclusion: checking brings in more than guessing. A closed question asked at the end of an exchange corrects more errors than a quarter of an hour of inference.
This holds particularly for those who know themselves to be good readers, because their assurance makes checking seem superfluous to them. That is precisely where the cost of error is highest, since nobody expects it.
Key points
The absence of feedback on our readings of others sustains our confidence without improving our accuracy.
The empathic accuracy protocol shows a weak correlation between confidence and correctness.
The contextual information available improves the reading far more reliably than a personal disposition.
This article describes documented mechanisms and disagreements. It recommends no position, no party and no vote, and commits to no reading other than that of the work it cites.
And you, where do you stand?
The Civix test situates your convictions on six dimensions and matches you with one of the twelve profiles, without ever telling you what to think.
Start my test