Social isolation, not loneliness

I keep thinking back to this experiment we did in our Psychopharmacology Lab course as undergraduates. The goal of the experiment was to test the effects of a drug that is supposed to reduce anxiety, using a mouse model system. "Mouse model system" is a fancy way of describing a biologically-relevant procedure in mice that is supposed to mimic anxiety in humans. In this case, it was a fairly simple process: take a mother mouse, with a bunch of pups. Remove one of those pups, and put it on a cold surface. The isolation from its mother, coupled with the cooler temperature, makes baby mice produce ultrasonic vocalizations. Basically, they'll start crying for their mothers. If you have a sensitive microphone, you can record these vocalizations, administer a putative anxiety-reducing drug, and see what changes, if any, it provokes. Better to do this with mice than with human children, and, after all, it's temporary, so the mice won't experience all that much trauma.

Anyway, NPR just blogged about a study that claims that social isolation and NOT loneliness leads to a higher risk of mortality.

I'd have to prod more deeply to see if this study has included measurements of any biological correlates of subjects' conditions.

Also, this whole branch of research gets pretty depressing sometimes.