In particular, the extent to which a person can predict the consequences of his/her actions is probably highly dependent on intelligence.
Religion can be viewed as taking some of this unfairness away.
For example, only 7 percent of the members in the U.S. National Academy of Sciences believe in a personal god [1]. I forget the numbers for the general public of the US (I don't have the book with me), but it was the vast majority who believed in a personal god.
Of course: grains of salt, correlation vs. causation, etc.
I don't think being an atheist requires a certain level of intelligence, but I do think that people of a certain level of education have been taught to think critically about everything they read/learn/work with, which probably bleeds over into their personal life.
[1] Victor J. Stenger. Has Science Found God? p.78
That is not to say that there is no correlation. In general educated people are more likely to be exposed to thoughtful commentary on atheism, and are therefore more likely to consider the idea. But the correlation seems to be with exposure, and not propensity to be convinced by it.
A modicum of research reveals that 'The authors reported a correlation of 0.60 between atheism rates and level of intelligence, which is “highly statistically significant.”'
Religion could reduce this and as a result be potentially beneficial to society.