Drunkard’s Fallacy

Two lads are making their way home, after some jars at the bar: “Seamus, would you give over circlin’ round that lamppost? You’re makin’ me head spin!” “Ahh, but Desmond, I’m tryin’ to find me feckin’ keys.” “Oh, now, Seamus, I t’ink I heard a ‘clink’ when we was coming t’rough de alley, back dere.” “Yeah, me an’ all, Desmond; but de light’s better under dis lamp.”

This shocking stereotype of Irish inebriation and false logic was offered to us in graduate school, in a course on research design, to illustrate the “Drunkard’s Fallacy” [the tendency for researchers to “search where the light is better,” and thereby overlook the “keys in the alley”]. Sir Francis Galton, for instance, believed that intelligence was highly correlated with head circumference [which is easily and cheaply measured]; and he tried to encourage fatheads to marry other fatheads, for the improvement of Mankind. Later, Dr. William Sheldon put forth the theory that one’s body type–fat, muscular, or thin [which is evident, even to the casual observer]–was highly correlated with 3 distinct sets of personality traits. All of which would be highly amusing, except that their “scientific evidence” has been used as the rationale for eugenics–most notoriously, but not exclusively, by The Third Reich.

These days in neuropsychological research, there is often a generous sponsor “paying the light bill,” who then–sometimes blatantly, but other times subtly–sets the agenda for “where to search.” Senator Charles Grassley has done Menschlich work, in my opinion, by doggedly insisting that medical researchers disclose the source of their funding, so that consumers can then take their “findings” with a grain of salt. But what if the funding source is Uncle Sam? Could there still be a tendency to “circle the lamppost,” rather than “go down the dark alley,” in search of scientific “truth”?

In 2005 the journal Nature Neuroscience published the results of an NIMH-funded study, which followed over 200 individuals from birth to 26 years, to assess their risk of becoming “depressed” by Stressful Life Events, and its correlation to the presence or absence of “the Serotonin Transporter Gene (5-HTTLPR)” in each individual’s DNA. Don’t you just know, the researchers found the two factors–“depression” in response to bummer events, and the presence of that specific gene–to be highly correlated. Well, the media was all over it like a cheap suit. Cute articles about “Blue Genes” came out like a rash. And the always-only-sleeping-not-dead eugenics lobby began to bang on about genetic screening for 5-HTTLPR, rationalizing that the opposite of bumming out at bad news was Being Resilient; and who wouldn’t want to breed Resilient kids, in these troubled times?

Also–and here’s the Beauty Part, if you’re a government agency, trying to contain costs for Mental Health treatment–if bumming out is “all in your genes,” no need to wander down that dark [time-consuming] alley of “trying to understand what got up your nose, which made you angry, which then made you depressed.” That’s like fiddling while Rome burns. Like trying to understand why your body has become insulin-resistant, or why your arteries are clogged. What you need is a chemical–not an insight. Faster, cheaper, better for Mankind [and the bottom line].

So, here’s the thing. In this week’s issue of the Journal of the American Medical Association: a meta-analysis of all studies which could possibly have replicated the ballyhoo’d NIMH results found no correlation between the two factors. Bupkes, nowt, Nichts. [Incidentally, their n = 14,250, of whom 1,769 were classified as having “depression.”] Do you know what was significantly correlated with “depression” in all the studies these researchers meta-analyzed? Stressful Life Events.

Well, Seamus, I guess it’s back to the dark alley, if we ever want to find those feckin’ keys…

Leave a comment

Filed under confounds, murky research, pro bono publico

Leave a Reply