Goldberg · Journal of consulting and clinical psychology 2024 · longitudinal experience sampling analysis within a randomized trial arm · n=243

Does it matter how meditation feels? An experience sampling study.

Cited 5 times in the scientific literature.

Level 3 - non-randomized controlled study

Longitudinal cohort analysis within a single arm of an RCT evaluating observational associations over time

PubMed 39347788 · doi:10.1037/ccp0000857 · record verified 2026-08-26

What was done

Researchers conducted an experience-sampling study within the meditation app arm (Healthy Minds Program) of a randomized controlled trial. Participants included 243 public school employees (80.9% with clinically elevated depression or anxiety) who recorded positive and negative affect during meditation sessions. Two-level multivariate latent growth curve models were used to evaluate trajectories of in-session affect over time and assess whether these changes predicted psychological distress at posttest and 3-month follow-up.

What was found

Negative affect during meditation decreased over time, whereas positive affect showed no significant overall change. Both increased positive affect and decreased negative affect during meditation were associated with reductions in psychological distress at posttest and follow-up. Positive affect change was a stronger predictor of follow-up distress reduction than negative affect change. The abstract reports no exact numerical effect sizes or model estimates.

Why it matters

These findings suggest that the immediate affective valence experienced during meditation correlates with mental health improvements, contrasting with traditional mindfulness frameworks that view in-practice emotional valence as incidental.

Limits

This was an observational, correlational analysis within a single intervention arm without randomized manipulation of affect. The sample was restricted to public school employees, limiting generalizability to broader populations. Numerical effect sizes and confidence intervals were omitted from the abstract.

Cited by