Is Your Mathematical Modeling a Good Idea?

The Feeling Good App: Part 2 of 2-- The Surprising Basic Science Findings-- How Does Psychotherapy REALLY Work? And Why Did Everything Change So Fast?  

Feeling Good Podcast Special Edition #2: March 07, 2022

Today’s special podcast features the second part of the recording with David and Jeremy Karmel, David’s founding partner of the Feeling Good App. Jeremy and David discuss the exciting results of the basic science findings most recent beta test, which included 140 participants. David uses an advanced form of statistics, called Structural Equation Modeling (SEM) to identify causal effects and to learn more about how the app actually works. This information has immense practical and theoretical implications.

Here's a portion of what we’ve discovered so far.

1. All seven negative feelings are high correlated because they all share an unknown Common Cause (CC) predicted by David in one of the top psychology research journals in the late 1990’s. Here’s the reference2

Burns, D. D., & Eidelson, R. (1998). Why are measures of depression and anxiety correlated? -- A test of the tripartite theory. Journal of Consulting and Clinical Psychology, 66(3): 461 - 473.

2. The CC accounts for most of the variance in all seven negative feelings, with R-square values ranging from 66% for anger, and 98% for Anxiety. Since there has to be some error variance in the estimates of the negative feelings, there is practically no room left for any significant additional causes.

If you would like to see the standardized output of the SEM model, click here.

3. The CC also has causal effects on Happiness, but these effects are much smaller, with an R-square of only 30%. This proves that Happiness has its own causes that are completely different from the factors that trigger depression. Happiness, in other words, is NOT just the absence of depression.
4. The radical reductions in all seven negative feelings were mediated by the reduction in the user’s belief in their negative thoughts, as predicted by cognitive therapists, like Albert Ellis and Aaron Beck, as well as the Greek Stoic philosopher, Epictetus, nearly 2,000 years ago. This is the first proof of that theory!
5. At least three components of the app have been isolated which appear to have substantial causal effects in the Common Cause, which in turn triggers simultaneous changes all negative feelings as well as happiness. Those three components include:
   1. A cognitive variable: the user’s belief in his or her negative thoughts.
   2. A motivational variable: measured with extremely precise and sensitive instruments.
   3. the user’s liking of the app.

The magnitude of all three causal effects was large. However, the motivational variables and user’s liking did not have direct effects on changes in depression and other negative feelings. The changes were ALL mediated via reductions in the user’s belief in his or her negative thoughts. This finding is consistent with the hypothesis that it is impossible to reduce negative feelings without change the belief in the negative thoughts that trigger those feelings.

6. The SEM models were replicated in two independent groups, including 60 participants with moderate to extremely severe depression at the start of the day, and 73 participants with no or only mild feelings of depression. The fit of the model was outstanding in both groups, and there were few or no significant differences in the parameter estimates. This indicates that the findings are valid and do not represent capitalization on chance.
7. David has reported extremely rapid changes in all negative feelings in his single-session treatment of individuals using TEAM-CBT. Some people have suggested that this is because he often treats mental health professionals as well as individuals who are very acquainted with his work.

CLICK HERE FOR THE FULLL REPORT

However, data from the beta test indicates this is not likely to be true. Mental health professionals did not respond any differently from non-professionals. In addition, the Familiarity with David or with TEAM variables did have modest effects on the degree of liking of the app, but no direct causal effects on changes in depression or the Common Cause.

The basic research is just beginning and ongoing. David believes that the research potential of the Feeling Good App may be as significant as the healing effects documented in the outcome findings with the app in the previous podcast.

If you are interested in participating in our upcoming beta test, you can sign up at www.feelinggood.com/app. We will be testing a radically revised version of the basic training module, plus some powerful new modules, and we will also be looking at relapse and relapse prevention techniques for the first time to find out if the improvements last.

Research on more than 10,000 sessions by human therapists using TEAM indicates that a portion of the gains patients make during individual sessions dissipates between sessions, but the “staying power” of the gains is facilitated by the patient’s homework between sessions. As a result, patient gains tend to reach a steady state after four or five sessions.

We anticipate that something similar may be documented in longitudinal studies with the app, and are eager to see what we can learn in the next study which will extend beyond one day.

So, hopefully, the new study will be pretty cool, too! And who knows what we’ll discover, with your help! Make sure you sign up if you’re interested in being one of our beta testers!

David and Jeremy

Rhonda, Jeremy, and David