Out of Control? Make up stuff! October 17, 2008Posted by gordonwatts in science.
I listened to a fascinating podcast from NPR Science Friday on the way into work today. The segment, titled “When not in control, people imagine order”, describes how people in (stressful) situations out of their control they will come up with reasons for why things are happening around them – even if they aren’t correct. One example used in the broadcast was stamping your feet before going into a meeting. If you do that, the people in the meeting will be more likely to accept and agree with your ideas. Clearly, there can be no external cause-and-effect, but you could see… we all have these little superstitions. How about a base-ball batter. 🙂 They noted that military personal in battle environments tend to do the same thing. Fun fact: sales of astrology books increased during the great depression (does anyone know how to see sales of astrology books as a function of time on Amazon?).
As you might imagine, recognizing patterns in what is going on around us is part of our survival programming. If we can make order out of what is going on, we can predict it, and thus we can take advantage of it. Leave meat out? Bear comes around. Hey – lets set a trap! Except, it can go too far – stopping your feet can’t possibly have any effect on the people in the meeting; the best it can do is boost your confidence. But if you knew that perhaps you could take a more direct path to boosting your confidence (shot of tequila!?).
What I really liked about this was that this is what my job is supposed to be. I’m supposed to look at data and come up with patterns to describe the data I see. There are lots of effects that might or might not be relevant. All of us in this field think up patterns where they don’t actually exist. Indeed, one is encouraged to come up with patterns no matter how crazy. The key is that the patterns have to be tested. The stomping feet thing might be based on a one-time accident. But here we would have to invent a double-blind study to test that before declaring the correlation was causal.
The program is worth a listen – it is short, about 20 minutes. And you get to hear someone call in and note that this study can’t really be correct because it implies that everyone who thinks the Kennedy assassination was a conspiracy must be making up a pattern that doesn’t exist because they are under stress. And we all know that conspiracy theory is correct.