Scientists are in the business of wanting proof for everything, and I find myself caught somewhere in the metaphorical middle on this issue. As a musician, I’m reminded on a daily basis of the utterly ineffable, indescribable powers of music. I’ve also witnessed the healing power of music firsthand. In old people’s homes and convalescent hospitals, when people have lost their memory due to Alzheimer’s disease, stroke, or other degenerative brain trauma, music is one of the last things to go. Old people who are otherwise unable to remember the names of their spouse or children, or even what year it is, can be brought arrestingly back to focus by hearing the music of their youth—songs that they sing along with, tap their feet to, and can remember all the notes and lyrics of. I’ve seen patients who could barely move, people with Parkinson’s who couldn’t walk, who can suddenly walk, trot, dance, and skip as soon as I start playing Glenn Miller or Artie Shaw on the rest-home CD player. There are reported cases of children with Down syndrome who can’t tie their shoes unless the sequence is set to music.
This ineffable power of music shows up not just in listeners but also in creators of music. The great songwriters and improvisers talk about not so much
creating
music, but having it written
through
them, as though the music comes from outside their bodies and their heads, and they are merely the conduit for it. Many great musicians, particularly in Third World cultures, reach a state of total ecstasy, a trance state, while playing music, in which their minds and bodies seem to be possessed by otherwordly forces. I’ve also felt this, whether improvising onstage at the Santa Monica Civic Auditorium with Mel Tormé, or writing incidental music for the film
Repo Man.
In describing the writing of one of my favorite songs of hers (1990’s “What We Really Want”), Rosanne Cash told me “It felt like I had just stuck up my hand and caught the song like you’d catch a ball in a catcher’s mitt—like it was out there all along waiting for me to grab it.” Our scientific theories have to be able to reconcile this common experience and the strong intuition that music is—dare I say it?—
magical.
On the research front, many of the studies on the effectiveness of music therapy were not performed according to rigorous scientific standards, and so their claims remain unproven. This situation parallels the unfortunate history of psychic research. One of the most crucial features of a rigorous experiment is the use of the comparison or control condition. In essence, we need to ask the following question: If there were
no
effect at all from the thing I’m studying, would this outcome have happened anyway? Too many music therapy experiments had inadequate controls, meaning that we aren’t shown what might have happened to the people in the experiment
without
music therapy.
Consider, for example, that out of twenty people who complain of headaches, a certain number are going to get better
anyway
if you just wait a few hours. If we play soothing classical music to twenty people with tension headaches and six of them say their headaches went away, we don’t know if some or all of those six headaches would have just gone away on their own. A control group in such an experiment should be similar in all respects to the people we’re studying, and get all the same treatment except for the one thing we’re interested in. If we give ten headache sufferers classical music and have them sit in a comfortable, sunlit room, and we give another ten headache sufferers
no
classical music but have them sit in an uncomfortable, darkened room, we have made the mistake of varying three parameters at once: We can’t determine which of those parameters had the effect.
In one published study on music therapy, a group of Korean researchers took stroke survivors and gave them an eight-week program of physical therapy that involved synchronized movements to music. The patients recovered a wider range of motion and flexibility compared to a control group. So far so good. But the control group had
no
therapy—no personal contact, no movement (with or without music), no one rooting for them or telling them that they would get better. We don’t know now whether the benefits to the first group came from the music, the movement, or simply the good feeling that came from knowing that a medical professional was looking out for and following their progress. Health improvements have been observed with far less.
I mentioned psychic research, and it is an irresistible subject. Some of the most interesting experiences I’ve had in my entire life were serving on review panels for scientists who had applied for funding to undertake research on psychic phenomena. I was asked to review their pilot data, findings from preliminary experiments that they felt showed evidence of psychic phenomena. In every single case, a lack of careful scientific controls rendered the data uninterpretable. In one study I reviewed, the person who was “reading minds” could only answer questions correctly if the experimenter already knew the answer and was allowed to interact with the “mind reader.” If the experimenter was silenced, the effect went completely away. I don’t think that the pair were trying to hoodwink anyone, but a parsimonious explanation—and a review of the experimental transcripts—suggests strongly that the experimenter was providing subtle, if unconscious clues to the “mind reader.”
What I found so interesting was the tenacity with which people, even trained scientists, held onto their beliefs about the supernatural when confronted with evidence that the experiments were flawed. First, here’s an example of how probability theory pertains to psychic claims. Suppose you have a standard deck of fifty-two playing cards. A friend of yours tries to guess the suit (hearts, clubs, diamonds, or spades) of each card—you can either look at the card (and try to psychically transmit the information) or you can keep it turned down until after your friend guesses. Now without working through a formal mathematical/probabilistic treatment of the problem, it should be clear that if your friend only guesses and has no psychic ability at all, she will guess a few of the cards right. In fact, in the long run, she will tend to get 25 percent of them right. It is the function of probability and statistics to help specify just how many she would have to get right for us to know, with reasonable certainty, that she wasn’t guessing.
While reviewing one such experiment in Silicon Valley, California, a very complicated experiment with many different facets, I pointed out to the lead research scientist (who held a Ph.D. in physics) that the chances of guessing a right answer in his psychic experiment were one our of four and he agreed. I pointed out that his best subject, after testing twenty people, had only gotten one out of four correct. He agreed to that. I suggested that she might have been only guessing.
“No!” he insisted. “She
told
me that she was really concentrating.”
I asked what his explanation was that she got a meager 25 percent correct, the same number that would have been guessed by a machine generating random numbers.
“She showed her psychic powers on 25 percent of the trials—what more do you want?” he demanded. He was getting agitated now. He started to speak very slowly. “Psychic powers can come and go like anything else. Even Artur Rubinstein doesn’t play Beethoven perfectly every time he sits down at the piano.” He knew my weakness.
“She had her powers on those 25 percent of the trials. And on the other 75 percent of the trials, the ones she got wrong,
those
are the ones where she was guessing!” I held my ground. If she had been truly guessing on those 75 percent of the trials, she would have gotten 25 percent of
them
right. He would have none of that. He had now stood up from the table and was red in the face with fists clenched, and his knuckles were turning a kind of ghostly whitish yellow. He seemed to be trying to stare me down.
“I have an idea,” I said at last. “Why don’t you have your subjects
tell
you which trials they’re guessing on, and which they
really, really know.
If your subject can get 25 percent of the suits right and can say ahead of time, before she gets any feedback, that those and only those trials are the ones where she is using psychic power, then I think we might have something.”
“We’ve done hundreds of experiments already using our existing method. We have all the data. Why should we go back to the experiments again just to satisfy one $@%* like you? I
know
that she has psychic powers,
she
knows it. Why can’t
you
admit it, Dan? Why do you have to be so
negative
all the time!”
The professional magician and skeptic James Randi has offered a one-million-dollar prize to anyone who can prove the existence of psychic phenomena, anytime and anywhere, anyone who can read minds, predict the future, influence the toss of a coin, or divine what playing card is about to be turned up, without using magic. No one has even come forward to try to claim the prize, but the money is in a certified escrow account, there for the taking. A researcher must simply follow the protocols designed to distinguish flimflam from fact.
Which brings me to the healing power of music. There are mountains of data on the effectiveness of music on illness, but not all reliable or reputable. Trying to separate the good from the bad would be enough work to earn some enterprising young investigator a Ph.D. thesis. If I sound skeptical or negative, I do not mean to denigrate the many fine music therapists who
are
helping people. Indeed, the American Music Therapy Association is just as interested as I am in weeding out those who are fakers, exploiters, and just plain incompetent. By the association’s own definition, music therapy is the “
evidence-based
use of music interventions to accomplish individualized goals within a therapeutic relationship by a credentialed professional . . .” [emphasis mine]. Certified music therapy is used for pain and stress reduction, motivation, anger management, as an adjunct to physical therapy in the case of motor difficulties, and for a variety of other purposes.
In just the past three or four years, however, an emerging body of evidence is pointing scientists in new directions. There have only been a dozen or so careful, rigorous studies and so I don’t want to overstate the case, but they seem to point to what the ancient shamans already knew: music—and particularly joyful music—affects our health in fundamental ways. Listening to, and even more so singing or playing, music can alter brain chemistry associated with well-being, stress reduction, and immune system fortitude. In one study, people were simply given singing lessons and their blood chemistry was measured immediately afterward. Serum concentrations of oxytocin increased significantly. Oxytocin is the hormone released during orgasm that causes us to feel good. When people have orgasms together and oxytocin is released in both, it causes them to feel strong bonds toward one another. “I feel good/I knew that I would/I got you.” You can see how
this
would be an evolutionary adaptation. Because the act of lovemaking (at least in the pre-birth control world) often led to pregnancy, it would be adaptive for the man and woman to feel a sense of connection, because that would increase the chances that the man would help raise the child, in turn significantly increasing the child’s chances of survival. Significantly, also, oxytocin has just been found to increase trust between people. Why oxytocin is released when people sing together is probably related evolutionarily to the social bonding function of music we saw in the previous chapter.
Looking beyond mental health to physical health, immunoglobulin A (IgA) is an important antibody that is needed for fighting colds, flus, and other infections of the mucous system. Several recent studies show that IgA levels increased following various forms of music therapy. In another study, levels of melatonin, norepinephrine, and epinephrine increased during a four-week course of music therapy, and then returned to pretherapy levels after the music therapy ended. Melatonin (a naturally occuring hormone in the brain) helps to regulate the body’s natural sleep/waking cycle and has been shown effective in treating seasonal affective disorder, a type of depression. It is also putatively linked to the body’s immune system because some researchers believe that it increases cytokine production, which in turn signals T-cells to travel to the site of an infection. Both norepinephrine and epinephrine affect alertness and arousal, and activate reward centers in the brain. All this from a song.
Music listening also directly affects serotonin, the well-known neurotransmitter that is very closely associated with the regulation of mood. (Prozac and a number of other recent antidepressants act on the serotonin system and belong to the class of pharmaceuticals called SSRIs,
selective serotonin reuptake inhibitors.
) Seratonin levels were shown to increase in real time during listening to pleasant, but not unpleasant music. And different genres of music caused different neurochemical activity! Techno music increased levels of plasma norepinephrine (NE), growth hormone (GH), adrenocorticotropic hormone (ACTH), and β-endorphin (β-EP) concentrations, all chemicals closely associated with improvements in human immune function. Techno was also shown to increase cortisol levels (not good for the immune system, but outweighed perhaps by the other increases), while meditative music decreased cortisol and noradrenaline. In the same study, rock music was shown to cause decreases in prolactin (at least in this group of techno-loving listeners), a hormone associated with feeling good.
We all suffer from stresses today that are very different from the stressors experienced by our ancestors, those very ancestors whose lifestyles caused the changes in DNA that we call evolution. When changes in lifestyle or environmental conditions created a subset of people who were better adapted to those early conditions, natural selection teaches us that those people were the ones who survived to pass on their DNA. This whole process can take a lot of time, thousands or tens of thousands of years. In other words, many parts of our DNA were selected for by evolution to cope with the world the way it was five thousand or even fifty thousand years ago. As biologist Robert Sapolesky points out, we are living in bodies and thinking with brains that were designed to solve problems that almost none of us has today.