Critical views of science in the news

Cybertherapy, placebos and the dodo effect: Why psychotherapies never get better


When the media report on a new diet that supposedly helps people lose weight once and for all, I wonder, "Does anyone still believe these claims, given the dismal track record of diets?" I have the same reaction to new treatments for psychological disorders, such as "cybertherapy."

In a long, lavishly illustrated article in The New York Times, Benedict Carey reported that psychotherapists are harnessing virtual reality for treating social anxiety disorder, alcoholism, agoraphobia, gambling addiction, post-traumatic stress disorder and a host of other mental ailments. Therapists can, in effect, place an alcoholic in a bar, an acrophobe in a rooftop party and someone who fears public speaking in front of a large, restless audience. They then can manipulate the virtual environment to test patients' ability to cope with different situations.

Patients can also discuss their troubles with a virtual therapist, relying on voice recognition and artificial intelligence software to parse remarks and respond appropriately. The therapist can take any form—male, female, black, white, young, old. The virtual therapist is a souped-up version of Eliza, an automated therapist that was created at the Massachusetts Institute of Technology in the 1960s and communicated via typed messages. 

The U.S. Army is spending $4 million a year on research into cybertherapy for traumatized veterans, according to the Times. In a program developed at the University of Southern California, veterans roll through a virtual Iraqi village in a Humvee, which is attacked by bullets and bombs. The Army's interest in cybertherapy evokes a certain cognitive dissonance because the military also employs virtual reality to prepare soldiers—from ground troops to pilots—for combat.

Cybertherapy sounds fascinating, and fun, but does it work? The Times story addresses this crucial question in its 31st paragraph. Researchers at the University of Quebec compared patients who received conventional talk therapy with others who got cybertherapy. "Both groups showed improvement, faring much better than a comparison group put on a waiting list," the Times reported.

So cybertherapy is about as effective—or ineffective—as more conventional talk therapy. This finding confirms the so-called dodo effect, which was originally proposed by the psychologist Saul Rosenzweig in the 1930s, when psychoanalysis was spawning a host of variants. He speculated that all talking cures share certain common factors—such as a caring therapist who establishes a bond with the patient—that make them equally helpful.

"Dodo" refers to an episode in Lewis Carroll's fable Alice's Adventures in Wonderland in which Alice and other characters wash up onto an island. There they encounter a dodo bird who persuades them to race around the island. The dodo finally announces that the race is over and proclaims, "Everyone has won, and all must have prizes!"

Over the last few decades, the psychologist Lester Luborsky of the University of Pennsylvania tested the dodo effect by comparing different psychotherapies, including psychoanalysis, cognitive-behavioral therapy and interpersonal therapy. His research confirmed that all methods are equally helpful to patients. Claims that one therapy is more effective than others, Luborsky showed, can usually be explained by the "allegiance effect," the tendency of researchers to find evidence for the therapy that they practice or favor.

Ironically, Luborsky, who died in 2009, displayed the allegiance effect himself. He strongly defended the value of psychotherapy in general and psychodynamic therapy in particular; psychodynamic therapy is a watered-down form of psychoanalysis that Luborsky favored in his clinical practice.

Other prominent researchers—notably Jerome Frank, a psychiatrist at Johns Hopkins—realized that the dodo effect undermined the validity of all psychotherapies. Frank's own research corroborated the dodo effect. In one study, he and colleagues provided depressed patients with three treatments: weekly individual therapy, weekly group therapy and minimal individual therapy, which consisted of just one half-hour session every two weeks. "To our astonishment and chagrin, patients in all three conditions showed the same average relief of symptoms," Frank wrote in Persuasion and Healing: A Comparative Study of Psychotherapy, first published in 1961 by Johns Hopkins Press and reissued in 1993.

Frank asserted that "relief of anxiety and depression in psychiatric outpatients by psychotherapy closely resembles the placebo response, suggesting that the same factors may be involved." The specific theoretical framework within which therapists work has little or nothing to do with their ability to "heal" patients, Frank contended. The most important factor is the therapist's ability to persuade patients that they will improve.

Frank's view should disturb anyone who thinks psychotherapy has a scientific basis. It doesn't matter whether your therapist is a Jungian, cognitive behaviorist, witch doctor—or a cybertherapist that exists only in a computer. What matters is whether you believe you will get better.

If talk therapy is really just a form of faith healing, should we abandon it and rely only on psychopharmacology for treating disorders such as depression and anxiety? Not necessarily. As Sharon Begley, one of my favorite science writers, pointed out in Newsweek in January, the placebo effect—and the allegiance effect—may also account for the reported benefits of antidepressants. All are losers, and none must have prizes.

Image of Sigmund Freud courtesy of Wiki Commons.

The views expressed are those of the author and are not necessarily those of Scientific American.

Share this Article:


You must sign in or register as a member to submit a comment.

Email this Article