There have been a lot more tests than those described in Stella's link though. Your (i) and (ii) tests are described in studies in my old favourite Delusions of Gender by Cordelia Fine.kabbes - i'm not sure if your description of those tests is accurate - from a brief scan of that paper the difference in results wasn't between
i) women doing a test and having to tick a box saying they are a woman, and
ii) women doing a test and not having to answer any gender questions
it was more a case of if they were told that the test they were doing had shown gender differences in the past in terms of outcome then those differing male/female outcomes were more likely to arise again than if they weren't told the results of previous tests in relation to gender differences
clearly this is still priming, but it's a bit different to what you describe - it's not some internal thing that says purely because i've reminded myself that i am a woman i will end up doing worse in this test than men of equivalent intelligence, but a response to a very specific of priming - which i would imagine may well arise if say two sets of men (or women) with different attributes done a test and they were primed with information that said one set of them usually did better or worse
It's a well-trodden path, basically, psychological experiment-wise.