Monday, October 22, 2012

Science, Pseudo-science and Statistical Noise


A scientist wrote:

<   [The] Index of /Misc._Physics/SB [reports that the San Diego] courthouse tower clock records 5.2 quake

<   http://www.cleyet.org/Misc._Physics/SB%20courthouse%20tower%20clock%20records%205.2%20quake/

<   [A list member who was there] reports he didn't feel the quake.

<   [The writer]  wonders, are after-shocks of statistical interest?  >

My reply:

At first reading, that seems like such a strange question for a scientist to ask, but then maybe it's not.  You're simply saying after-shocks are just statistical noise, signifying nothing, right?  A 5.2 quake wouldn't be just random noise here in Massachusetts, but I can imagine it well might be in Santa Barbara.   

I'm sure you would agree that people discovering and noticing the background noise from any phenomenon of the natural world is a wonderful thing, and an opening to connect and engage them in enjoyable and even useful conversations.  

People, scientists included, evidently have a built-in need to concentrate on the noise and tease a meaningful signal out of it.  We scientists have elevated that need to a new level, inventing and using powerful techniques and tools for verifying that a signal is real, devising explanations of its meaning, verifying and coming to agreement that an explanation makes sense.  (Not always in that order.)  Scientists and lay people alike, we all seem to have built into us - perhaps even built into our wiring - a need to find that signal in the noise, and tease out its meaning.  Many of you will know of experiments that demonstrate this.

Perhaps this apparently innate urge derives from the instincts of the hunter, sitting silently and still in the forest, listening to the chorus of sounds, watching the field of movements, sensing the medley of scents, feeling every vibration of the earth, waiting for a signal or combination of signals that might indicate food or danger, tuning out the rest.  

That we all share this innate capacity is part of what makes it possible to engage with regular people, to share with them our ways of thinking, reasoning and understanding, and to share with them tools for thinking more powerfully about the natural world and about policy questions that concern it.

This innate capacity is also preyed upon by moneyed interests, who use it to rob people of their earnings and shape their political choices.  

I had a friend, "Joe", who was a compulsive gambler.  I used to count cards for him at the Blackjack Tables, and he would always leave them a bit ahead.  But then he would feel the need to prove to me that he could beat the other games as well.  Sometimes he would win at those for a while, but in the end he would always leave the casino broke.  At one table, Joe and each other player had their own private system for finding the pattern in the way the cards were dealt.  As far as I could determine, the cards were dealt as close to entirely randomly as human ingenuity could devise, but because this random pattern would repeat every 400 cards or so a computer could beat this game.  That pattern however would be far beyond the capacity of the human mind to see and remember, so they were all trying to find a simpler one.  The house even provided them with charts, tables and pencils to help them with their search for the winning formula.  

Perhaps an opposite case is that of the deniers of natural selection and global warming and their following.  The leading "deniers" include theocrats, charlatans and pseudo-scientists, many in the pay of powerful monied interests.  A crucial piece of their strategy is persuading people that if they can't sense the signal in the noise.  They then argue on religious authority that there should be no signal.  

Once persuaded, people will simply not see the signal even when it's as plain as day and carefully explained to them.  

I'm sure we all know of examples of this kind of error among scientists as well. 

No comments: