In yesterday's NY Times Harvard psychologist and "happiness researcher" (sic!) Daniel Gilbert has an essay titled "What you don't know makes you nervous". The text is available at: http://happydays.blogs.nytimes.com/2009/05/20/what-you-dont-know-makes-you-nervous/
Quoting one of the main points made by Gilbert: |That's because people feel worse when something bad might |occur than when something bad will occur. Most of us aren't |losing sleep and sucking down Marlboros because the Dow is |going to fall another thousand points, but because we don't |know whether it will fall or not - and human beings find |uncertainty more painful than the things they're uncertain about. A couple of points: (1) Gilbert describes experiments (no references; how do I get a gig where I can superficially describe research and not provide references?) that show that uncertainty about future aversive events made people either more afraid or "unhappy" (not clear how happiness was measured; perhaps neuroimaging was involved to "sex up" the research). Now, I am all for doing experimental research. I'm a "internal validity" kind of guy (i.e., let's establish a causal relationship and then see how it behaves in different contexts). Even so, I realize that external validity, the ability to generalize beyond the original context of the experimental study is a tricky business because most experiments are conducted in "closed systems" (i.e., "third variables" that can affect the independent and dependent variables are controlled for; mediation, moderation, and other forms of interactions can not be observed) while real life occurs in "open systems" (where the influence of an unknown number of variables can operate and perhaps even change the nature of the causal relationship shown in the original research). Making generalizations are always troublesome because the person making the generalization needs to understand that an experimental result may not generalize to ordinary life -- we need to establish this empirically. Failure to anticipate interactions with other variables may give an inaccurate representation of the casual relationship that one may think is universal (something I call the "main effect fallacy"). (2) There are clear exceptions to the "wisdom" that Gilbert provides and it is useful to look are reader's comments on his essay. I quote from one of them: |I can't be the only person who read this article with tremendous |frustration. There are a lot of things I am certain of: I'm closer to |60 than 50, my house sale in a bear market barely covered my |debt, I'm living with relatives in a situation which is satisfactory to |no one, most of my possessions are in storage, I'm out of work |and there are no jobs right now: not no jobs in my field, no jobs, |full stop. I've got plenty of certainty, and it brings me no comfort |at all. Neither do Pollyanna articles like this one, which IMHO are |one step up from telling me to eat cake. I would just point out that there are many comments, some confirming what Gilbert wrote and some providing counter examples invaliding his advice/insight. See the following (more comments provided under the heading "Let the Bad Times Roll"): http://happydays.blogs.nytimes.com/2009/05/21/let-the-bad-times-roll/?th&emc=th So, I am left with a question: is this kind of essay/epistle the sort of thing that psychologists really should be doing? Is it "wisdom" or "conceit" that we are communicating? -Mike Palij New York University [email protected] --- To make changes to your subscription contact: Bill Southerly ([email protected])
