Why you need to know this: the previous article in this series described someone who cannot feel their own mistakes, and many readers wrote in to say the person they had in mind was the exact opposite. This one is about them.
A version of this trait looks nothing like the one everybody has heard about, and the working assumption almost everybody brings to it is wrong in a way that costs the people around them.
The first surprise is where they sit. When 344 managers were compared with people from the general population on both forms of the trait, the managers came out higher on the loud, self-promoting form and lower on the quiet, thin-skinned one (Dåderman & Kajonius, 2024). So, on balance, the quiet one keeping score is not your boss. They are far more likely to be sitting beside you.
The second surprise is what it predicts. In that same group of managers, the loud form predicted none of the three kinds of work performance the researchers measured. The quiet form predicted more of the behaviour that damages an organisation and less of the work that was actually asked for.
The third surprise is the one worth carrying around. Across three studies of 938 supervisors in Germany and the United Kingdom, the leaders who reported treating their people worst were the ones higher on the quiet form, and the pattern did not repeat when the researchers ran the same tests on the loud form (Braun, Schyns, Zheng & Lord, 2025).
The route it takes is the strange part. These leaders do not deflect the blame for a failure. They absorb it, decide the failure means something about them personally, feel the shame that follows, and then aim it at whoever is nearest and least able to object. In the UK study, when the design fixed who was at fault in advance, the link between the trait and hostile treatment disappeared entirely.
Which flips the practical advice. The colleague who says, "This is all my fault,” is not asking to be contradicted, and reassuring them more doesn’t touch the mechanism. What appears to matter is whether the failure gets separated from the person, out loud, early, and in front of the people who witnessed it.
Three further findings sit alongside those. A pooled analysis of 437 studies and 123,043 people found the aggression link holds for the quiet form as well as the loud one, and roughly doubles once the person feels provoked (Kjærvik & Bushman, 2021). Roughly two-thirds of what the standard questionnaire measures turns out to be ordinary emotional instability (Miller et al., 2018). And the most-used ten-question measure splits into two barely related halves, only one of which behaves like a dark trait at all (Starlinger, Voracek & Tran, 2022).
That last one is the reason a chapter here is given over to the people you will wrongly accuse. In this territory, the most common error is missing it. It’s finding it in somebody anxious.