The starting point is the experience sampling method, or ESM, in which participants repeatedly report on their momentary mood and symptoms. Prior studies that used the exponentially weighted moving average, or EWMA, procedure to monitor for imminent depressive episodes relied on daily sample sizes of five or ten questionnaires sustained over extended periods. The EWMA procedure is a statistical process control technique for monitoring ESM data in real time.

The catch is burden. Asking participants to complete multiple questionnaires per day over long stretches of time is demanding, and the authors identify this participant burden as a key limitation of the existing fixed-sample approach.

To address this, the research team, which includes Fien De Pauw, Evelien Schat, Francis Tuerlinckx and Eva Ceulemans, proposed a variable sample size EWMA, or VSS EWMA. In this variant, the daily sample size depends on the extremity of participants’ ESM scores on previous days. When an individual is doing well, the daily sample size, and therefore the burden, stays low.

The researchers evaluated the performance of VSS EWMA through a simulation study and through an application to empirical ESM data. Their findings indicate that VSS EWMA improves the balance between detection speed and participant burden, the two competing goals at the heart of real-time mood monitoring.

Beyond the headline result, the paper offers practical guidance. The researchers provide recommendations on how to tune VSS EWMA to different priorities, allowing the scheme to be adjusted depending on how detection speed and participant load are weighted in a given setting.

The authors also outline directions for future research. The findings come from simulation work and an application to empirical ESM data rather than from a clinical trial, so the approach remains a statistical proposal whose value in applied monitoring contexts will depend on further evaluation.

The study, published as an original manuscript in Volume 58 of Behavior Research Methods, suggests that adaptive questionnaire scheduling may ease the load on participants without sacrificing the ability to flag warning signs of an approaching depressive episode, offering a potential path toward more sustainable real-time monitoring schemes.

Subject of Research: Adaptive sampling to reduce participant burden in ESM-based monitoring of mood disorder onset

Article Title: Reducing participant burden of ESM-based monitoring of mood disorder onset through variable sample size EWMA

Article References: De Pauw, F., Schat, E., Tuerlinckx, F., & Ceulemans, E. (2026). Reducing participant burden of ESM-based monitoring of mood disorder onset through variable sample size EWMA. Behavior Research Methods, 58(12), Article 332. https://doi.org/10.3758/s13428-026-03183-x

Image Credits: AI Generated

DOI: 10.3758/s13428-026-03183-x

Keywords: experience sampling method, EWMA, statistical process control, mood disorders, depressive episodes, participant burden, adaptive sampling, Behavior Research Methods, simulation study, real-time monitoring, Reducing, participant