The Mindfulness Mystery Tour!

And Two Mind-Boggling Discoveries about Meditation!

Featuring Jason Meno

Today, Jason Meno, our beloved AI guy on the Feeling Great App team, shares some incredible and innovative research he recently did on the effect of meditation on how we think and feel. As you know, basic research is a high priority of our app team, and our major focus is to make basic discoveries in how people change, and especially on what triggers rapid and dramatic change. We use that information to develop and refine the app on an ongoing basis, and also to contribute to basic science.

Jason recently created a “New Cool Tools Club” which has 160 members who Jason can notify whenever he has a cool new app tool that he wants to test. If you are interested in joining, you can find his contact information at the end of the show notes. There is no charge if you’d like to join this group!

Jason had a strong background in Buddhism and has been working with our company for several years, focusing in the last year on the AI chat bot portion of the Feeling Great App. He has meditated for many years, and uses TEAM-CBT as well to deal with his personal moments of stress and unhappiness, something that most if not all of us experience at times!

Introduction

Jason was interested in evaluating the short-term impact of meditating, and did a literature review but found that most or all of the published studies had a focus on the effects of daily meditation over longer periods of time, like two months for example.

He was also interested in how long and how often people should meditate, and what types of meditations, if any, were the most effective.

So, he decided to test a one-hour meditation experience consisting of five ten-minute recorded meditations, including

1. A body scan meditation, systematically relaxing various parts of your body, beginning with your feet and toes.
2. A breathing and counting meditation, where you focus on your breathing and count the breaths going in and out.
3. A loving kindness meditation, starting with sending feelings of love, happiness, and health first to someone you love, then to yourself, then to someone you aren’t especially close to, or don’t particularly like, and on and on until you are projecting love and kindness to the entire universe.
4. A mindfulness exercise where you notice if you are thinking, hearing, watching, remembering, and so forth as various thoughts pass through your mind.
5. A “Do Nothing” meditation where you are instructed to simply “do nothing” for ten minutes.

Because previous research on meditation did not use scales that assessed specific kinds of negative feelings in the here-and-now, he decided to use the highly accurate 7-item negative feelings sliders as well as the 7-item positive feelings sliders prior to the start of the medicine, after each meditation, and at the end of the app.

He also asked many questions about motivation and expectations prior to the start of the meditation experiences, all answered from 0 (not at all) to 100 (completely), including

1. How familiar are you with David’s work?
2. How familiar are you with meditation?
3. How strongly do you believe that meditation will make you feel better?
4. How strongly do you believe that meditation will be rewarding?
5. How strongly do you believe that meditation will only have a small effect?
6. How strongly do you believe that meditation will be a waste of time?
7. How strongly do you believe that meditation will make you feel worse?
8. How strongly do you believe that it will be painful or difficult?

You can find these data at this link.

He also asked every participant to generate an upsetting negative thought, like “I’m a loser,” and use 0 to 1000 sliders to indicate how strongly they believed that thought, and how upsetting it was.

60 individuals started the experiment, and 35 completed it, with 25 dropping out prematurely before they completed some of the meditations.

He presented the data as a two-group analysis, those who completed and those who failed to complete the hour of meditation. Here, are just a few of the preliminary findings, and more refined analyses are planned so we can look at causal effects.

1. Both groups were moderately to very familiar with David’s work and with meditation.
2. The completers had higher scores on the questions about positive expectations than the dropouts, although the differences were not great.
3. The dropouts had substantially higher scores on four questions about negative expectations for the experience, like “it will be a waste of time” or “it will be painful or difficult.”
4. The initial scores on the belief in the negative thought were similar in the two groups (76% and 74%, respectively), but the Upsettingness of the thought was a bit higher in the completers (83% and 79%.
5. The mean of the initial scores on the 7 negative feelings sliders was significantly higher in the dropouts (37% and 46%, respectively), while the initial scores on the 7 positive feelings sliders was somewhat lower in the dropouts (49% and 45%, respectively).
6. Both groups expected a modest reduction in negative feelings and a modest boost in positive feelings during the hour of meditation.

Results on the 35 completers

1. After the first ten-minute meditation, there were significant reductions in the negative feeling sliders (from 37% before to 25% after) and increases in the positive feeling sliders (from 45% before to 55% after).
2. There did not appear to be any additional improvements in negative or positive feelings in the subsequent four meditations.
3. There was a significant reduction in the belief in the negative thought after the first meditation, and the reduction continued throughout the next four meditations. (76% to 54%), for a reduction of 29%.
4. There was a significant reduction in the upsetness caused by the negative thought after the first meditation, and the reduction continued throughout the next four meditations (79% to 47%) for a reduction of 40.5%.

You can find the remarkable results if you click here!

There are many fascinating results, but one of the most amazing--which we've replicated almost exactly in independent beta tests--is the remarkable similarity between the changes in negative and positive feelings the participants predicted, and the actual results. They are so close it looks like somebody faked the data, but that's not the case at all.

We will have to do more analyses to figure out what this means, but in simple terms, this seems to be iron clad proof that our expectations of the mood changing results of any intervention can be tremendously powerful. In fact, you could argue--and it would need further statistical analyses to test--that the causal impact of the expectations eclipsed the causal impact of the actual intervention, which in this case was meditation.

One of the cool things about quantitative research is that it nearly always shoots down our favorite hypotheses, and also gives us new and totally unexpected gifts to stimulate our thinking! In this instance, there were at least two mind-boggling and toally unexpected results:

1. When people mediate, the improvement in negative feelings is accompanied by parallel reductions in participants belief in their negative thoughts.
2. Participants predictions of the changes in seven negative and seven positive feelings by the end of the hour of meditation were spot on, and seemed almost impossibly accurate!

Discussion

The findings are exciting and specific, and suggest that the reduction in negative feelings during meditation may be, and is, mediated by the reduction in the users’ belief in their negative thoughts. We will attempt to look into this more deeply using non-recursive analytic methods with SEM (structural equation modeling).

All samples are biased, and it can sometimes be extremely helpful to understand the bias in your sample when interpreting the results. The sample in this case included users favorably disposed to meditation, and responding to an email inviting them to participate in a meditation experiment. Only those who persisted the full hour were analyzed in the final outcome data, which could be another source of bias in the data. How much improvement would we have documented if we were analyzing completers (45) AND dropouts (35)?

Actually, this type of analysis is possible using Direct FIML (Full-Information Maximum Likelihood) with SEM techniques. I will, in fact, do these analyses as soon as I get the data set from Jason. This will allow me to estimate the scores at the end for all participants, including those who dropped out. It seems mathematically impossible, but it actually can be done.

If those who dropped out are systematically different from those who continued, it will “know” and correct for this. For example, if those who dropped out were, on average, doing more poorly, then the estimates based on those who persisted will be biased, and the degree of bias could potentially be infinite. The SEM analyses will also tell us if there are no significant differences in those who  persisted and those who dropped out.

Finally, the data LOOKS like the meditation “caused” some fairly significant improvements, although the results were in some