The Emily Rosa Therapeutic Touch Test

Therapeutic touch, developed within nursing from the 1970s and taught in nursing schools for decades afterwards, holds that a practitioner can perceive a field extending beyond the body and can smooth or rebalance it with the hands, generally without contact. The claim contains a testable component that had apparently gone untested: if the field can be perceived, a practitioner should be able to say which of their own hands another person's hand is hovering over. The apparatus was a cardboard screen with two holes at the bottom, through which the practitioner put their hands palm up, and a towel laid over the forearms so the experimenter's movements could not be seen.

The experimenter flipped a coin, held a hand a few inches above the practitioner's left or right palm as the coin dictated, and asked which. Twenty-one practitioners, recruited from those willing to be tested, completed 280 such trials between them. They were correct on 44 percent, where chance alone predicts 50. The test cost nothing and required no expertise beyond the discipline to run it exactly as specified. Fourteen of the twenty-one were tested twice, in sessions several months apart, which allowed the investigators to check whether any individual performed consistently rather than only whether the group did. None did.

The paper was reported everywhere as the story of a child who debunked a therapy, and Emily Rosa was indeed nine when she ran the trials for a school science fair and eleven when JAMA published. That framing did the result a disservice in both directions. It let supporters treat the study as a stunt rather than an experiment, and it let critics of the practice enjoy the humiliation instead of the finding. The substantive point is duller and more damaging: the practice had been taught in nursing schools, offered in hospitals and defended in its own professional literature for roughly twenty-five years, and the simplest possible test of its central perceptual claim had not been published by anyone in that time.

A field that has not asked its own easiest question has a problem that no single result can fix. The response from practitioners was to argue that the test was artificial, that perceiving a field in a laboratory is not the same as working with a patient, and that the therapeutic benefit does not depend on the perception. Those replies may be true. They also concede the specific claim the test examined, and no comparable test has since produced a positive result. The test is now a standard teaching example of how cheap a decisive experiment can be.