This week I was taken to task a bit by a friend who claimed that decision theory was ultimately atheistic and opposed to higher values. Lest anyone else think such a thing, I need to respond. That simply is not true. The charge of atheism implies an active opposition to theism (the belief in an interactive deity). Decision theory makes no such claim. After all, Rev. Bayes, whose theorem is the basic one that started it all, was a Presbyterian minister and an ardent theist.

A better description is that decision theory is used to make predictions of probability of occurrence of events in the physical world based on past observations and analysis: it is agnostic with respect to anything that cannot be observed.

Concepts such as “good” or “evil” can only be tested subject to operational definitions.

Implicit in decision theory are several philosophical assumptions. One important assumption is that the world is rational. That is, even if our senses are bombarded with a cacophony of seemingly unrelated data, a sufficiently well-developed analysis can make sense of it.

Humanity’s ability to perform the proper analysis started from a simple animal response to stimulus and has grown over history. That growth is one reason some people think decision theory reflects simple materialistic values. In the good old days, like 3 – 4 thousand years ago, our ability to analyze data was minimal. But we needed tools to carry out our daily lives. Myths, legends, and various religions supplied the necessary paradigms. As time went by, physical observations were better explained by scientific analysis. The myths fell as physical truths, but often transformed into philosophical truths and moral lessons. We learned that the Earth goes around the sun, not vice versa, and that mental illness is not the result of demonic possession. Entrenched powers fought the obvious loss of prestige and power that advancing knowledge took from them. Galileo was placed under house arrest and others killed in ways that are not polite to discuss. Even today some sects seek to dilute the knowledge of evolution with myths.

All religions assumed the roles of ethics and values advisor along with the role of science advisor. Given the state of humanity at the time, one can argue this was a useful tool for most people. Sadly, even today we read about occasional mishaps when uneducated people attempt to perform exorcisms on children when proper medical treatment would be preferable. Exorcism is misuse of an outdated tool. Parents should know proper medical care is preferable.

My friend said that decision theory is “soulless.” In a literal sense, that is correct. A soul, as normally defined, has no interaction with the physical world and so cannot be observed and quantified. That does not mean that decision theory denies the existence of souls. It just means that given the available data and definitions, the probability of the existence of immortal souls is negligibly small. This observation does not necessarily discourage investigators from trying to get better observations. At least one formal study attempted (in 1907), to determine the weight of a soul by putting dying patients on a precision scale and weighing them just before and after death when their souls would have presumably left the physical body. The initial results showed that a soul weighed about the same as a piece of toast (no butter). Later studies showed similar results with animals, which are usually not suspected of having souls. The loss of weight at death was later attributed to purely physical effects such as loss of water. Still, it was a good try.

In response to the interest my original tutorial generated, I have completely rewritten and expanded it. Check out the tutorial availability through Lockergnome. The new version is over 100 pages long with chapters that alternate between discussion of the theoretical aspects and puzzles just for the fun of it. Puzzle lovers will be glad to know that I included an answers section that includes discussions as to why the answer is correct and how it was obtained. Most of the material has appeared in these columns, but some is new. Most of the discussions are expanded compared to what they were in the original column format.

[tags]decision theory, statistics[/tags]