I'm completely fascinated by the concept that being under the pressure of a stereotype can cause people to subconsciously underperform, and ever since first learning about this when Claude Steele came to the Williams campus when I was doing a lot of work in multiculturalism and diversity training, it's been near the top of my mind when I think about group differences in performance and expectations that we have for people.
This isn't just about gender and race, though those are obviously two major categories of group identity where people are frequently told that their membership in that group is likely to define their ability. This can happen in any grouping where individuals invest identification and may, as a result, be the target of stereotypes about intellectual performance. I haven't read it, but I found the abstract for a study on stereotype threat and MDMA users very interesting.
If the effects of stereotype threat are entirely subconscious and measurable, what does that tell us about the reality of measured differences between groups? This is directly relevant to a lot of the work I'm doing professionally, but it's also directly relevant to how I think about expectations and stereotypes. In general, we like to think that people live up to expectations, and it turns out they can live down to them, too, even if they're trying hard not to.