I just read a wonderfully obscure paper entitled “Factors in the Selection of Surface Disinfectants for Use in a Laboratory Animal Setting,” published in the Journal of the American Association for Laboratory Animal Science in March, by Michael Campagna and colleagues at the University of California, Los Angeles. They studied, well, disinfectants. Various kinds are used in labs, and the scientists examined, among other things, which ones have odors that lab mice find aversive. Why study that? Because it’s a small piece of a big issue related to genes and behavior: the role of the environment.

Suppose scientists want to know what a gene—let’s call it Gene Z—has to do with behavior. Using genetic-engineering wizardry, they generate a line of mice lacking Gene Z (“knockout” mice), plus another line with an extra copy of Gene Z (“transgenic overexpression” mice). Then they see if there’s something different about the behavior of either group when compared with unmanipulated control mice.

So a lab discovers that Gene Z is pertinent to, say, anxiety. Knock out the gene, and mice don’t get anxious; overexpress the gene, and they’re more prone to anxiety. (How can you measure mouse anxiety? Mice, being nocturnal, are afraid of light. Researchers might put food in the middle of a brightly lighted arena and see how long it takes for a hungry mouse to leave a comforting, shaded corner to get the food.)

More on this...

With the results in, other labs get some of the Gene Z mice, eager to study different aspects of anxiety. And surprisingly often, something disquieting occurs. Another group confirms the link between Gene Z and anxiety, but they don’t see as big of an effect. Then another lab reports that the gene has no effect on anxiety. Yet another finds that Gene Z decreases anxiety.

Yikes. Everyone wonders if the scientists don’t know what they’re doing, or if the test is reliable. But research started in the 1990s by neuroscientist John Crabbe of the Oregon Health and Science University suggests a different explanation.

Click for more from the Wall Street Journal.