Due to curiosity alone, I'm very interested in this topic, and a lot of the papers were very revealing. However, I see a lot more bad science than good papers.
Spontaneous Focusing on Numerosity Mathematical Skills of Young Children
This paper notes that no significant difference was found between male and female children. However, because it wasn't trying to detect this difference in the first place, and because it doesn't claim this result as one of its conclusions, the result holds no weight. If I watched a solar eclipse and noted that I couldn't see any shift in the positions of the surrounding stars, that can't be used as evidence against GR.
It annoys the hell out of me when people say that women are less smart than men. This is of course not true at all, as I've experienced many times personally.
Nobody is claiming that no woman can be smart, or that no man can be dumb. We're arguing about statistics, and your experiences don't count as a proper statistical study.
Math and Gender
http://spectrum.ieee.org/at-work/education/math-and-gender
"One strand of evidence comes from a study at the University of Wisconsin, published in the Proceedings of the National Academy of Sciences, which finds that the overall progress of women in society is a surprisingly good indicator of their performance at the highest math levels."
Well, no kidding.
My overall progress in society is also a good indicator of my performance at the highest math levels. It's obvious that those at the top of society have more opportunities, more access to educational materials, and more free time to pursue interests. I don't think this sheds any light on innate ability. No matter how good or bad girls might inherently be at math, one would expect the number of girls who go into math to increase as gender equality increases.
Also, take a look at the graph at the bottom, and note the complete lack of correlation. I'd be willing to bet money that even randomly-produced data would tend to have higher correlation than this data.
Isn't it ironic that:
1) estrogen is associated with improved mental faculties:
http://www.sciencedaily.com/releases...1117161246.htm
And
2) testosterone kills neurons:
http://www.jbc.org/content/early/200...93200.full.pdf
It would be, except masculization of the brain is done by estrogen, not testosterone. Estrogen, a metabolite of testosterone, can't easily cross the blood-brain barrier while testosterone can. In males, testosterone crosses the blood-brain barrier and is metabolized into estrogen, specifically estradiol, which then carries out the chemical reactions involved in masculization.
As for the evolution argument, I think this is a joke.
If this refers to me, I'm not using evolution as an argument. It's perfectly OK to speculate about the evolutionary origins of gender differences, but no reputable scientist would tell you that evolution can predict gender differences not yet observed.