EyeBreathe: A Mixed Reality Game With Eye Tracking and Breathing Biofeedback for Adolescent Basketball Players - A Randomised Controlled Trial

The study will use a two-arm group design. Participants will be randomly assigned to either the Mixed Reality (MR) intervention group (Breathing MR) or to a control group. Randomisation will be performed at the individual level. The intervention period is planned for six weeks, with two sessions per week for the experimental group. Control participants will continue their regular basketball training without any additional intervention.

This is a follow-up, two-arm parallel-group randomised controlled trial of Breathing MR versus usual basketball training. The study is designed to recruit an anticipated 40-50 participants. Subject to confirmation of comparability of eligibility criteria, intervention delivery, outcome procedures, and data quality, data from this cohort may be combined with the Breathing MR and usual-training control groups from the completed feasibility RCT (Organisation's Unique Protocol ID: 2-PĒK-4/1205/2025-Feasibility) in prespecified exploratory pooled analyses. Models will include cohort as a fixed effect and assess the intervention-by-cohort interaction. Cohort-specific and pooled estimates with 95% confidence intervals will be reported.

In-person study activities will be conducted at participants' usual basketball training venues in Salaspils and Rīga, Latvia. Home-based HRV recordings will be completed by participants using loaned Polar H10 chest straps and the Kubios HRV application.

Interventions will be delivered using the HTC VIVE Focus Vision standalone head-mounted display, requiring no external equipment or PC connection.

Assessments will be conducted at two time points: at baseline (before the intervention period) and post-intervention (within one or two weeks of the final session). The assessment battery includes resting heart rate and heart rate variability (Polar H10 chest strap with KubiosHRV), CO₂ tolerance (CO2TT), breathing pattern (Hi-Lo test), reactive stress tolerance (Vienna Test System - Determination Test), and pre-competition anxiety (CSAI-2R-LV).

Research staff will supervise the pre-intervention assessment phase and the first intervention sessions in person. From week two onward, participants will access sessions independently using individual password-protected profiles on the headset. A midpoint visit will be conducted at week three.

To evaluate the physiological and cognitive efficacy of the standalone MR breathing biofeedback intervention across varying exposure levels, analyses will be structured around pre-specified exposure tiers and continuous dose-response modeling:

1. Primary Per-Protocol (PP) Population (at least 4 sessions) - to evaluate efficacy among athletes receiving at least a minimum training exposure (at least 33% of scheduled volume) and maintain direct comparability with the filtering logic of the 2025 feasibility cohort (2-PĒK-4/1205/2025-Feasibility), primary efficacy models will analyze participants completing at least 4 intervention sessions.
2. High-Adherence Sensitivity Analysis (at least 8 sessions) - a secondary high-adherence per-protocol analysis will evaluate outcomes among participants completing at least 8 sessions (at least 67% of scheduled volume) to test for enhanced physiological adaptation under higher intervention dosage.
3. Continuous Dose-Response Evaluation - to directly investigate the relationship between intervention dose and outcome change without relying on binary attendance cutoffs, linear regression models will evaluate total completed sessions (0 to 12) as a continuous predictor of change in primary and secondary outcomes.

The study was approved by the Riga Stradiņš University Ethics Committee (original approval: 2-PĒK-4/610/2024; additional approvals covering extended details: 2-PĒK-4/814/2025 and 2-PĒK-4/1205/2025). Individual participant data, analytical code, and study documentation will be archived in the repository upon study completion.