Alvin Plantinga argues that if evolution and metaphysical naturalism are both true, then we have no reason to trust our judgments when it comes to metaphysics.
Actually, whether or not metaphysical naturalism is true, we shouldn’t trust most of the conclusions we reach in metaphysics (or a priori methods; more on that below). One only need to look at the actual world. In the actual world, if you put three philosophers in a room, they won’t agree about anything, including whether there are three people in the room.
Plantinga might very well be right that evolution wasn’t concerned with us having correct metaphysical beliefs, such as the belief in metaphysical naturalism. Nevertheless, we can be pretty confident in certain metaphysical conclusions like the question of whether God exists. After all, most philosophers and scientists actually agree that God doesn’t exist! In other words, even if we are aware that most of our metaphysical beliefs are false, it doesn’t follow that we can’t be rational (and very confident) in believing that certain metaphysical claims are true. Unlike, for example, ‘Platonism’, theism generates predictions that we can go out in the world and test . And, one can/should be very confident that ghosts and fairies do not exist.
Epistemological and methodological naturalism fit well within the worldview of metaphysical naturalism, and if the history of metaphysics has shown us anything, then the conclusion is that we should derive our metaphysical beliefs from the empirical data. In other words, (pure reason) a priori speculation won’t tell you the way the world is or what exists; if it did, then the ontological argument for God could be a sound argument. In the words of Sean Carroll, “Our metaphysics should follow our physics. That’s what the word ‘metaphysics’ means”.
Someone might object to all this by saying, “That’s scientism! And scientism is self-refuting, because claiming that all knowledge is derived from science or the senses is a claim that isn’t itself scientific or derived from the senses”.
Actually, from what has been said already, it should be apparent what is wrong with this objection. We have tons of evidence from history that ‘just thinking’ really hard about the deep questions of life is not a reliable way to get at truth. Therefore, my claim is empirically based. In addition, for instance, I never said that statements that are true by definition can’t be known a priori.
The upshot is that one should examine a hypothesis and asks what predictions that hypothesis generates. After that, one should go out into the world and see the results. If a hypothesis doesn’t generate predictions, then (right out of the gate) that’s a huge strike against it.
 I actually find mathematical Platonism to be somewhat silly, and it surprises me how many skeptics are Platonists. I don’t know what it means for the number ‘2’ to just exist outside of space and time. I also don’t see any need to invoke Platonism when explaining what “numbers” are.