Only theism requires the ability to believe.
Time was we could only argue about this from the point of view of linguistics, philosophy, etc. That is no longer true, as even acknowledging the many limitations to the various studies on this issue, there remain certain things we
can say:
"Despite these uncertainties (or perhaps because of them), the one apparent consensus among the commentators is that atheists are best accounted for by Hypothesis 2 (Natural Variation):
atheists are simply one end of a continuum of belief. On the face of it, this is unsurprising, even an anti-climax. Like numerous other traits in nature, beliefs vary - so what? However, if this is true, then there are in fact several striking implications. First, it implies that the mean of the distribution is some positive level of religious belief (that is, there is a consensus that
natural selection has favored cognitive mechanisms underlying belief,
and/or religion itself). Second, it implies that atheism is (or was) a suboptimal strategy for human beings. Third, it implies that atheists - given their status at the tail end of the distribution - are (or were) selected against."
Johnson, D. (2012). What are atheists for? Hypotheses on the functions of non-belief in the evolution of religion.
Religion, Brain & Behavior,
2(1), 48-70.
And although explanations for religious belief have long been a subject of (usualloy bad) study regarding evolutionary origins, it is in fact the increasing study of disbelief, atheism, agnosticism, whatever, that has improved our understanding of belief in general and in particular the evolutionary causes that motivate both belief and disbelief.
"atheism is
more prevalent and enduring than would be expected if it was solely driven by effortful rejection of intuitive theism, that disbelief does not always require hard or explicit cognitive effort, and that rational deliberation is only one of several routes to disbelief."
Norenzayan, A., & Gervais, W. M. (2012). The origins of religious disbelief.
Trends in cognitive sciences.
Not only is atheism not characterized by simply a lack of belief, but self-reports atheists give about their non-commitment are often belied by biological indicators to the contrary:
"We asked atheists (Studies 1 and 2) and religious individuals (Study 1) to verbally dare God to cause unpleasant events, like murders and illnesses to happen to themselves and their intimates. Atheists did not regard the statements as unpleasant as the religious participants did in their explicit self-report. The impact of conviction was strong as it explained 38% of the variance in the unpleasantness ratings. However, when the participants' emotional arousal was analyzed by their skin conductance level during their verbal dares, a different picture emerged."
Lindeman, M., Heywood, B., Riekki, T., & Makkonen, T. (2013). Atheists become emotionally aroused when daring God to do terrible things.
International Journal for the Psychology of Religion, (just-accepted).
Then there is the ways that concepts are represented in general.
You can state that atheism is a default position as much as you wish, but you cannot change the fact that this self-designation is a belief around which and by which you organize a conceptual network and quite literally its physical representation.
Of course, for some atheists (and agnostics) this doesn't mean much. Certainly, theists and the religious will, at least with respect all things religious, show a tendency towards greater diversity in classification. Nothing about any of the above is surprising in the least, actually, unless of course one has an ideological commitment to the idea that atheism represents some sort of base state.