in Meditation

Progress and Challenge in Contemplative Studies

As most everyone is aware, meditation is increasingly popular in the West. There has also been a marked increase in clinical, neuroscience, psychological and social science investigations of meditation, focusing on long-term practitioners and those participating in short-term mindfulness and related training programs. Comparing the first to the second five years of this century, there has been a more than 300 percent increase in the publication of basic and applied studies in this area that is now known as “contemplative science.”

Article originally appeared in the Huffington Post

Research evaluating the effectiveness of meditation-based interventions in medical, mental health, educational and business settings is difficult. The gold standard for evaluating interventions in biomedical settings is the randomized, controlled, double-blind experiment. In clinical drug research, participants are randomly assigned to receive either the experimental medication or a so-called placebo which looks and tastes like the experimental drug but is pharmacologically inert. In this way, both the participant and researcher are “blind” to which condition each participant has been assigned, thus controlling for several potential sources of bias. In applied meditation research, participants are often randomized to either a meditation training condition or a “wait-list” control condition in which there is no training but comparable testing before and after the period of time taken for the meditation training condition. A major problem for the interpretation of results from such research designs is that participants can never be truly blind to the condition in which they have been randomized. It is obvious to the participant whether they are or are not receiving meditation training. Why is this a problem?

In addition to whatever may be the impact of the meditation practice per se, those in the training condition are likely to be biased toward different expectations for benefit and such motivations as wanting to please the researcher, compared to those in the wait-list control condition. Thus, it is not possible to disentangle these sources of bias from the specific effects of the meditation practice itself. Because many of the available experimental studies of meditation in applied settings have employed this kind of research design, we are actually less confident about the effectiveness of meditation training than much of the popular press would suggest. Fortunately, alternative designs are possible, in which participants are randomly assigned to a meditation training condition versus another training condition, such as physical exercise, didactic instruction, or simple muscle relaxation training. When the meditation and other active training condition are carefully equated for such things as amount of training, amount of home practice, credibility and enthusiasm of instructors, then investigators can be more confident that any result that differs between the meditation and control groups reflects something specific to the meditation practice itself.

Read full article via Progress and Challenge in Contemplative Studies | Al Kaszniak.

Leave a Reply

  1. My response to the original article:

    I believe that you are referring to the kind of design described in this paper:
    The Pervasive Problem With Placebos in Psychology
    Why Active Control Groups Are Not Sufficient to Rule Out Placebo Effects

    In my opinion, the authors don’t quite pick out out all the relevant issues as well as they might, because researchers who have an expectation about one practice being better than another might subtly influence the expectations and enthusiasm of the instructors of the “placebo” group.
    What is needed as a TEAM of researchers, each with a competing agenda that favors one of the practices over all the others, to be the sole point of contact for the instructors of the practice that they advocate.

    Thus far, there has been only one study on meditation that goes that far, and it was published nearly 25 years ago:
    Transcendental meditation, mindfulness, and longevity: an experimental study with the elderly.
    Each of the three competing meditation practices had an advocate on the research team. Only that advocate interacted with the instructor for the practice they were advocate for. The researchers ensured that all instructors of all practices met the standards of professionalism exhibited by the TM teachers (TM was one of the techniques studied), so that they were highly familiar and confident in their material, wore business attire when teaching and lecturing, etc.
    Researchers were blind to subject participation, and all data was collected by graduate students from Harvard University who were blind to both group participation AND to research protocol.

    Two interesting points:
    1) the “mindfulness” practice studied was NOT the MBSR and related practices studied today, but something developed by Professor Langer several decades ago.
    2) the “low mindfulness relaxation” practice was actually meant to be a variant of Benson’s Relaxation Response, but the practice did so poorly (no significant measure between that and the control group on any measure whatsoever) that the practice was renamed in the abstract to avoid Herbert Benson being embarrassed. Even so, Benson issued a press release denouncing the study, claiming it was NOT using the RR.
    Unless and until meditation studies use the design adopted in teh above study, they will continue to be plagued by placebo-related issues.

    Consider this proposed study for how it is being done in the modern world:
    Design and rationale of a comparative effectiveness trial evaluating transcendental meditation against established therapies for PTSD.
    Note that the prediction is that the two active PTSD therapy groups will be “non-inferior” to each other. That is the only way that such a team of researchers can legitimately be recruited: there is NO preferred outcome from the study, so no researcher is left with the impression that he/she is part of the token placebo group and no such attitude will be passed on to instructors for that group.

    It is not enough to “ensure” that placebo group instructors won’t have a bad attitude. You must only allow researchers who believe that the placebo group is the real group and the other group is the placebo group to interact with such instructors. And the only way to do THAT is to make all active groups equal at all stages of the design by only allowing the rhetoric “non-inferior” to be used at any stage.