I'm currently re-reading Stephen Maitzen's excellent paper, "Divine Hiddenness and the Demographics of Theism" (Religious Studies 42 (2006), pp. 177-191). As the title suggests, Maitzen appeals to the data of the demographics of theism to defend the argument from divine hiddenness. However, in the same paper, he also offers a powerful argument against the existence of a sensus divinitatis (i.e., a natural capacity to form properly basic belief in God in a wide variety of circumstances or "triggering-conditions", such as (e.g.) looking at the starry heavens, reading the Bible, etc.) postulated by reformed epistemologists of the likes of Alvin Plantinga and others. Below is a summary of that portion of Maitzen's article.
The argument summarized: The sensus divinitatis is supposed to be an innate capacity or faculty in all humans. Now innate human faculties tend to be evenly distributed among humans (e.g., hearing, sight, etc.). But the kind of belief that’s supposed to issue from the sensus divinitatis is very unevenly distributed among humans (e.g., Saudia Arabia contains 26 million people, and 95 percent of them are theists. By contrast, Thailand contains 65 million people, and only 5 percent of them are theists). We thus wouldn’t expect this data of “geographic patchiness” with respect to theistic belief if humans were endowed with a sensus divinitatis. However, such data is readily and naturally explained purely in terms of anthropological, sociological, political and economic factors. Therefore, the data of the demographics of theism provides good evidence against the existence of a sensus divinitatis within humans.
Objection: Christian theism entails that the sensus divinitatis was damaged – and perhaps even rendered inoperable – by human sin, whether the sin in question is the original sin from the Fall or the particular sins of humans in their earthly lives. Furthermore, Christian theism entails that it requires the work of the Holy Spirit (e.g., Plantinga’s notion of the internal instigation of the Holy Spirit) to repair and restore the sensus divinitatis in a given individual. Therefore, the fact that some people -- or even very many -- lack belief in the theistic god doesn’t provide good evidence against the existence of a sensus divinitatis within humans.
Reply: This objection misses the point. The problem isn’t the non-universality of theistic belief; the problem is the uneven distribution of theistic belief. As Maitzen puts it, “why has God bestowed this restorative grace so unevenly, contributing to a pattern that, coincidentally, social scientists say they can explain entirely in terms of culture?” Thus, appeal to human sin and divine grace fails to adequately mitigate the criticism.
 To see Maitzen's excellent criticism of a molinist response to problems related to the demographics of theism, see the exchange between Maitzen and Jason Marsh in the following two papers: (i) Marsh, Jason. "Do the Demographics of Theistic Belief Disconfirm Theism? A Reply to Maitzen", Religious Studies 44 (2008), pp. 465-471; (ii) Maitzen, Stephen. "Does Molinism Explain the Demographics of Theism?", Religious Studies 44 (2008), pp. 473-477.
 Maitzen, "Divine Hiddenness and the Demographics of Theism", 187.