In this paper, I consider V. S. Ramachandran's in-principle agnosticism concerning whether neurological studies of religious experience can be taken as support for the claim that God really does communicate with people during religious experiences. Contra Ramachandran, I argue that it is by no means obvious that agnosticism is the proper scientific attitude to adopt in relation to this claim. I go on to show how the questions of whether it is (1) a scientifically testable claim and (2) a plausible hypothesis, serve to open up some important philosophical issues concerning interpretive backgrounds that are presupposed in the assessment of scientific hypotheses. More specifically, I argue that naturalism or scientific objectivism in its various forms is not simply a neutral or default methodological backdrop for empirical inquiry but involves acceptance of a specific ontology, which functions as an implicit and unargued constitutive commitment. Hence, these neurological studies can be employed as a lever with which to disclose something of the ways in which different frameworks of interpretation, both theistic and atheistic, serve differently to structure and give meaning to empirical findings.