Saturday, January 3, 2009

Praying and poker

Do religious people stand to be better poker players than non-religious?

I've been thinking of late on the question of how science can uphold standards in presenting the results of scientific work in the larger public domain. Here's something I picked up on from the New York Times a day or so ago. Michael McCullough and a fellow psychologist at the University of Miami have in press a paper in an upcoming issue of the Psychological Bulletin, entitled "Religion, Self-Regulation, and Self-Control: Associations, Explanations and Implications". The authors ask whether the available data permit conclusions about whether religion's influences on health, well-being and social behavior can be ascribed to its influences on self-control and self-regulation. This is a big paper, reviewing a large body of data from various sources. The Times piece about the paper appears in John Tierney's column. Tierney offers readers a chance to weigh in on the issues in TierneyLab.

I can't point to egregious faults in the paper itself or Tierney's report on it, but there is a looseness of language throughout that bothers me. It is not easy to establish scientifically valid conclusions on questions that revolve around human behavior and thought. Much of the evidence is in the form of data from surveys. But it can be difficult to trace through seemingly sound survey data to something that can reliably be proposed as a cause and effect relationship. There simply doesn't seem to be a way to test the question of whether religion, or religious observation, is causally related to particular attitudes and behaviors. To cite just one bit of non-survey evidence adduced, there is a bit in the paper about whether religious or spiritual behaviors promote self-regulation. One of the studies cited had "experienced meditators or "pray-ers" engage in meditation during functional imaging brain scans." It is a bit unclear from the paper just what is being compared with what, but in any event, I wonder what the results have to do with religion. I think of myself as moderately meditative, and I think I'm also highly self-regulated and self-controlled. But I don't think of my meditative activity as religious in nature.

The complexity of human social interactions and situations makes for tough going in moving from correlations to causal implications. Many people are religious because they were brought up that way, and they continue in those traditions. Many others may have been brought up in a religious tradition, but have rejected it in whole or part in adulthood. If these latter turn out to be well-balanced people with high levels of self-control, empathy, generosity and many other good characteristics, are all those virtues ascribable to their religious upbringing? When we look at people generally how do we factor in the powerful roles of socio-economic environment on parenting and other cultural forces that make for whether one is exposed to a religious influence, or to examples of altruism, kindness, generosity in other contexts? And what about the fact that religion seems to come in a wide range of colors and flavors? In short, is this paper really testing scientifically interesting questions? Tierney doesn’t allude to these kinds of difficulties in his column.

Scientists are by no means of one mind on the question of what constitutes valid scientific evidence. Science prides itself on having internal mechanisms of quality control, such as peer review. Even that mechanism, however, is only as good as the intellectual standards that prevail within each particular science community. Thus, “scientific opinions” on topics of societal concern (climate change, stem cell research, religion and emotional health) are not of uniform quality. Physical and natural scientists tend to dismiss much social science work as being “soft”: not based on replicable experimentation, often presenting conclusions that seem to rest on rather sketchy distinctions or slim statistical margins. But it is all too easy for physical and natural scientists who normally deal with highly controlled laboratory situations, to find fault with empirical social science research. At the same time, because so much work in the human sciences finds a receptive audience with non-scientists, sometimes too much attention is paid to work that is not really very good, or is over-interpreted by the media. Here is yet another challenge for science in its relations with the larger society: How to ensure that the science that makes it into the public sphere meets high standards of significance and reliability.

No comments:

Post a Comment