I’m not sure where this puts me on the rationalist spectrum, but I think it can be rational to believe something for the sake of the practical benefits. Suppose you got cancer and learned that believing in the effectiveness of chemotherapy increases the likelihood that your cancer will go into remission. In that case, you would have prudential (non-epistemic) justification for believing that it would in fact work.
Pascal’s Wager says that we have most prudential reason to adopt Christianity, seeing as a) the price of being a Christian who winds up wrong is minimal, b) the benefits of being right are enormous, c) the benefits of turning out to be right as an atheist are minimal, and d) the price of being an atheist who turns out to be wrong is enormous—infinite negative utility in hell, and missing out on Heaven!
The commonest reply to the Wager is known as the “many gods” objection: there are an infinite number of systems of supernatural claims that pose the same dilemma, and nothing to distinguish them in terms of likelihood. The problem with this response is that if there is at least one religion that has a greater probability of being true than another, there will be reason to prefer religious belief to non-belief.
To get the expected value of being a Christian, you have to multiply both the probability of the truth of Christianity by the size of the benefits if it turns out to be right, which are infinite. And there is reason to think Christianity has more (very weak) evidence in its favor than other religions, namely the fact that it is the most ubiquitous religion practiced by humanity. On a Bayesian view of probability, anything that is expected on the hypothesis is evidence in its favor, and there’s reason to expect that Christianity would be the most influential religion in the world if it was true that God was interested in the affairs of humans and intervening on behalf of Christianity.
Another objection comes from the Doxastic involuntarists (DIs). I often encounter the weakest version of this claim from people who say that it is impossible to make oneself believe something without sufficient evidence. This has the hilarious implication that Christianity must be true, because Christians believe it. If they can’t believe it without good evidence, and they believe it, it follows that they must believe it on the basis of sufficient evidence.
But that’s unfair to the less naive version of DI, which says we have no voluntary control over our beliefs, although they can be influenced by factors other than the available evidence. But in that case, we will have to consider what those factors are, and once we specify them we’ll realize they’re exactly the sort of thing we can control. Namely: group affiliation, prestige-seeking, selective attention, the relative desirability of different beliefs are among the non-epistemic factors which hold influence over our beliefs.
Of course, one can simply take advantage of the above and inculcate a commitment to Christianity: affiliate with a Christian community, surround yourself with people who are Christians and who reward you for your devotion, selectively expose yourself to Christian information sources and ignore non-Christian sources, and meditate on the desirability of Christianity a la Pascal’s Wager.