In her latest post, Sarah Constantin writes:

Whatever ability IQ tests and math tests measure, I believe that lacking that ability doesn’t have any effect on one’s ability to make a good social impression or even to “seem smart” in conversation.

I remember that I made the opposite prediction a few months ago, while I was studying psychometrics more actively. In particular, I had the sense that because of the positive manifold (i.e. almost all positive mental attributes quite strongly correlate), it seems like judging someone's IQ should be relatively easy, and also quite valuable in assessing a large number of other positive qualities, which should make it quite evolutionarily advantageous to be able to assess it.

Two concrete experiments I've thought about:

1. Take a group of people whose IQ you now, then just take pictures of their faces (or short videos of them saying something), and then have a group of participants rate them on their intelligence (probably in order). See how strong they correlate.

2. Throw a bunch of participants into a group, have them talk to each other, then afterwards ask them about the relative judgement of the intelligence of the other members of the group, see how predictive they are.

These seem like very straightforward experiments, that I can imagine someone having already run, and where I would be pretty interested in the results. If only so that I can be better calibrated on how much to trust my gut judgement of someone's intelligence.

Does anyone know of any similar experiments that have been run?

Crell

Crell

Crell

Crell

Crell

Crell