Here is a paragraph from one of his recent columns that illustrates what he has to say:
"... dependency on belief and its host of psychological biases is why, in science, we have built-in self-correcting machinery. Strict double-blind controls are required, in which neither the subjects nor the experimenters know the conditions during data collection. Collaboration with colleagues is vital. Results are vetted at conferences and in peer-reviewed journals. Research is replicated in other laboratories. Disconfirming evidence and contradictory interpretations of data are included in the analysis. If you don’t seek data and arguments against your theory, someone else will, usually with great glee and in a public forum. This is why skepticism is a sine qua non of science, the only escape we have from the belief-dependent realism trap created by our believing brains."