In an article in Nature (limited access), there is discussion of how people not only have their own opinions, but recognize their own facts–and forget about the science.
People endorse whichever position reinforces their connection to others with whom they share important commitments. As a result, public debate about science is strikingly polarized. The same groups who disagree on ‘cultural issues’ — abortion, same-sex marriage and school prayer — also disagree on whether climate change is real and on whether underground disposal of nuclear waste is safe.
The ability of democratic societies to protect the welfare of their citizens depends on finding a way to counteract this culture war over empirical data. Unfortunately, prevailing theories of science communication do not help much. Many experts attribute political controversy over risk issues to the complexity of the underlying science, or the imperfect dissemination of information. If that were the problem, we would expect beliefs about issues such as environmental risk, public health and crime control to be distributed randomly or according to levels of education, not by moral outlook. Various cognitive biases — excessive attention to vivid dangers, for example, or self-reinforcing patterns of social interaction — distort people’s perception of risk, but they, too, do not explain why people who subscribe to competing moral outlooks react differently to scientific data.
The implications for Humanism, it seems to me, is that people will be resistant to accepting the obvious implications of science in discrediting the supernatural unless secularlism can provide for the emotional needs which are currently being satisfied by a combination of religion and materialism.