[strlen]

Would be great to have a link to the actual study, but studies typically control for socio-economic-status, family situation, etc... I share the your general sentiment, but keep in that Mensa's threshold of 130 means that 2% of a population with a 100 median IQ (assuming a 15 SD test) is eligible to join.

While far from everyone eligible to join Mensa does so (and as you've said, there's a very good reason for that), there is likely large enough to be able to form a statistically significant sample that resembles the general population with the same IQ which I suspect is readily available (at least in US SES by IQ can be gauged from GSS -- where the WORDSUM question is a "good enough IQ test", or from college statistics -- where the SAT is also a "good enough" IQ test for the purpose of a general study; not sure about UK.)

This isn't a "correlation vs. causation": I don't believe the study was looking for or claiming a causal relationship; I suspect it could well be the reverse: increased anxiety about predators being a push for better ability to predict where the predator might appear -- i.e., visuo-spatial intelligence (there is also some data that suggest that Homo Sapiens intelligence may have evolved accidentally out of certain neurological disorders).

So if the question is "is this a real correlation", there's good reason to think that the experiment was indeed designed to handle any confounding factors like the one you suggested. This is probably a better experimental sample than "college students who volunteer to take part in a psychology study" and easily allows for the most obvious variables to be controlled.