A survey paper in Psychological Review collecting studies in which people were asked, immediately afterwards, why they had just done what they did.

The cleanest of them was run in a commercial setting, presented to passers-by as a consumer survey. Four pairs of nylon stockings were laid out in a row. Fifty-two people were asked which was the best quality, and then why.

The stockings were identical. The rightmost was chosen over the leftmost by close to four to one, and not one subject spontaneously mentioned where in the row anything had been lying.

When asked directly about a possible effect of the position of the article, virtually all subjects denied it, usually with a worried glance at the interviewer suggesting that they felt either that they had misunderstood the question or were dealing with a madman.

Nisbett and Wilson, Psychological Review, 1977

What they gave instead were the knit, the sheerness, the weave — properties of a stocking, described with care, by people looking at four objects that did not differ in any of them.

The paper’s conclusion is narrow and is worse than it sounds. The verbal report is not a reading taken off the process. It is produced afterwards, from whatever seems a plausible reason for a choice of that kind, by the same person, in good faith, at conversational speed, and it arrives with no mark on it saying which of the two it is.

The reason was assembled after the choice and fitted to it. Nobody was lying. Nobody could tell.