Dunning-Kruger and the curse of knowledge

The Dunning-Kruger effect is a cognitive bias where a relatively unskilled individual overestimates his level of skill within a particular area. It is thought to arise because the individual lacks the knowledge needed for accurate calibration. If you don’t know what an expert performance looks like then you won’t realise that you are lacking that expertise.

The curse of knowledge is a different and yet related cognitive bias and is effectively a failure of empathy. If you know something then this thing that you know seems transparent. The ease with which you can retrieve this knowledge leads you to think that everyone must know it; that it’s obvious.

What happens when teachers suffering from the curse of knowledge teach students suffering from the Dunning-Kruger effect? I suspect you get an unwitting conspiracy where everyone thinks the situation is fine: the teacher thinks her students understand and the students think they understand. Yet we should see a major impact on the performance of the lower-achieving, Dunning-Kruger affected individuals.

What’s the solution? Teachers need to be regularly confronted with what the students don’t know. This needs to be systematised so that it is not up to the teacher to decide when and if to check. The curse of knowledge is powerful: In my own teaching, I make assumptions about what students have learnt that are overly optimistic and I repeatedly make these assumptions even though I am aware of the cognitive bias.

Students also need this information. We must not protect them from the truth about their performance because this only feeds the delusion.