Acute Neurophysiological and Autonomic Responses to a Respiratory-Coupled Sound-Based Digital Therapeutic (BELL-001) Versus Sham, Followed by 7 Days of Home Use, in Adults With Insomnia Symptoms: A Randomized, Sham-Controlled Study
Sponsor study identifier
Secondary identifiers
OTHER · Korea Institute of Science and Technology Institutional Review Board
OTHER_GRANT · Korea Technology and Information Promotion Agency for SMEs (TIPA), Ministry of SMEs and Startups, Republic of Korea
OTHER_GRANT · National Research Foundation of Korea (NRF), Ministry of Science and ICT, Republic of Korea
interventional
72
1 country
1
Study Start
November 18, 2025
Actual (reported)Primary Completion
February 13, 2026
Actual (reported)Study Completion
February 20, 2026
Actual (reported)The purpose of this study was to describe how brain activity and heart rhythm respond to BELL-001, a smartphone application that plays sounds linked to the user's own breathing, compared with a sham application that looks the same but does not link its sounds to breathing. Adults with self-reported insomnia symptoms were randomly assigned to BELL-001 or to the sham application. During one laboratory visit, brain activity was recorded with electroencephalography (EEG), heart activity with electrocardiography (ECG), and breathing with a chest belt while participants rested with their eyes closed, then used the assigned application actively for 10 minutes, and then listened passively for 50 minutes. After the visit, all participants, regardless of laboratory group, used the BELL-001 application at home for 7 nights in an open-label period while wearing a smartwatch.
Registry-reported information. Elapsed time and enrollment totals alone do not measure study progress or recruitment pace.
February 20, 2026
Actual (reported)Click on a node to explore related trials.
Some registry-reported names shown below are not yet linked in this graph.
Dates and qualifiers reported by the registry. Estimated dates do not confirm that a milestone has occurred.
November 18, 2025
Actual (reported)February 13, 2026
Actual (reported)February 20, 2026
Actual (reported)October 7, 2026
Actual (reported)October 13, 2026
Actual (reported)Change in Sensor-Level Electroencephalography (EEG) Power in the Delta Band (0.1 to 4 Hz) From Rest to the Active and Passive Stages
EEG power (microvolts squared) in the delta band (0.1 to 4 Hz) at each of 56 scalp electrodes, calculated as the active-stage value minus the resting-stage value and as the passive-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10), active stage (minutes 10 to 20) and passive stage (minutes 20 to 70) of the 70-minute recording in the single laboratory visit
Change in Sensor-Level Electroencephalography (EEG) Power in the Theta Band (4 to 8 Hz) From Rest to the Active and Passive Stages
EEG power (microvolts squared) in the theta band (4 to 8 Hz) at each of 56 scalp electrodes, calculated as the active-stage value minus the resting-stage value and as the passive-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10), active stage (minutes 10 to 20) and passive stage (minutes 20 to 70) of the 70-minute recording in the single laboratory visit
Change in Sensor-Level Electroencephalography (EEG) Power in the Alpha Band (8 to 12 Hz) From Rest to the Active and Passive Stages
EEG power (microvolts squared) in the alpha band (8 to 12 Hz) at each of 56 scalp electrodes, calculated as the active-stage value minus the resting-stage value and as the passive-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Change in Median of Relative RR Intervals From Rest to the Active and Passive Stages
Resting stage (minutes 0 to 10), active stage (minutes 10 to 20) and passive stage (minutes 20 to 70) of the 70-minute recording in the single laboratory visit
Change in Interquartile Range of Relative RR Intervals From Rest to the Active and Passive Stages
Resting stage (minutes 0 to 10), active stage (minutes 10 to 20) and passive stage (minutes 20 to 70) of the 70-minute recording in the single laboratory visit
Change in Mean RR Interval From Rest to the Active and Passive Stages
Resting stage (minutes 0 to 10), active stage (minutes 10 to 20) and passive stage (minutes 20 to 70) of the 70-minute recording in the single laboratory visit
Change in Heart Rate From Rest to the Active and Passive Stages
Resting stage (minutes 0 to 10), active stage (minutes 10 to 20) and passive stage (minutes 20 to 70) of the 70-minute recording in the single laboratory visit
Change in Standard Deviation of NN Intervals (SDNN) From Rest to the Active and Passive Stages
Resting stage (minutes 0 to 10), active stage (minutes 10 to 20) and passive stage (minutes 20 to 70) of the 70-minute recording in the single laboratory visit
Participants used the BELL-001 smartphone application during one laboratory visit, with sound delivered through insert earphones. After a short practice, a brief auditory evoked potential recording and 10 minutes of eyes-closed rest, participants completed a 10-minute active stage in which they held the touchscreen pressed during each inhalation and released it during each exhalation while the application played sounds that followed this press-and-release signal, and then a 50-minute passive listening stage in which the pace of the sound slowed gradually from the breathing rate established in the active stage. Eyes remained closed throughout the recording. After the visit, participants used BELL-001 at home for 7 nights in an open-label period, as did participants in the sham group.
Participants used a sham application with the same appearance and screen flow during one laboratory visit, with the same practice, auditory evoked potential recording and 10 minutes of eyes-closed rest. During the 10-minute active stage, they tapped the touchscreen each time they heard a brief tone presented at random intervals of 2 to 12 seconds. During the 50-minute passive stage, a single fixed pre-recorded sound that was not linked to breathing was played. Eyes remained closed throughout the recording. After the visit, participants in this group also used the BELL-001 application (not the sham) at home for 7 nights in an open-label period while wearing a wrist-worn smartwatch.
Smartphone-based digital therapeutic that plays sound linked to the user's breathing phase through insert earphones. In this study, the participant's own touchscreen press (inhalation) and release (exhalation) served as the breathing-phase signal during the active stage, followed by a sound sequence that slowed from the breathing rate established in the active stage. A respiration belt was used for recording only and was not an input to the application.
Korea Institute of Science and Technology
Seoul, 02792, South Korea
Yoonseob Lim, PhD
Korea Institute of Science and Technology
72 participants (actual enrollment reported by the registry).
First Submitted
October 7, 2026
First Posted
October 13, 2026
Resting stage (minutes 0 to 10), active stage (minutes 10 to 20) and passive stage (minutes 20 to 70) of the 70-minute recording in the single laboratory visit
Change in Sensor-Level Electroencephalography (EEG) Power in the Low Alpha Band (8 to 10 Hz) From Rest to the Active and Passive Stages
EEG power (microvolts squared) in the low alpha band (8 to 10 Hz) at each of 56 scalp electrodes, calculated as the active-stage value minus the resting-stage value and as the passive-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10), active stage (minutes 10 to 20) and passive stage (minutes 20 to 70) of the 70-minute recording in the single laboratory visit
Change in Sensor-Level Electroencephalography (EEG) Power in the High Alpha Band (10 to 12 Hz) From Rest to the Active and Passive Stages
EEG power (microvolts squared) in the high alpha band (10 to 12 Hz) at each of 56 scalp electrodes, calculated as the active-stage value minus the resting-stage value and as the passive-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10), active stage (minutes 10 to 20) and passive stage (minutes 20 to 70) of the 70-minute recording in the single laboratory visit
Change in Sensor-Level Electroencephalography (EEG) Power in the Beta Band (12 to 30 Hz) From Rest to the Active and Passive Stages
EEG power (microvolts squared) in the beta band (12 to 30 Hz) at each of 56 scalp electrodes, calculated as the active-stage value minus the resting-stage value and as the passive-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10), active stage (minutes 10 to 20) and passive stage (minutes 20 to 70) of the 70-minute recording in the single laboratory visit
Change in Sensor-Level Electroencephalography (EEG) Power in the Low Beta Band (12 to 20 Hz) From Rest to the Active and Passive Stages
EEG power (microvolts squared) in the low beta band (12 to 20 Hz) at each of 56 scalp electrodes, calculated as the active-stage value minus the resting-stage value and as the passive-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10), active stage (minutes 10 to 20) and passive stage (minutes 20 to 70) of the 70-minute recording in the single laboratory visit
Change in Sensor-Level Electroencephalography (EEG) Power in the High Beta Band (20 to 30 Hz) From Rest to the Active and Passive Stages
EEG power (microvolts squared) in the high beta band (20 to 30 Hz) at each of 56 scalp electrodes, calculated as the active-stage value minus the resting-stage value and as the passive-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10), active stage (minutes 10 to 20) and passive stage (minutes 20 to 70) of the 70-minute recording in the single laboratory visit
Change in Sensor-Level Electroencephalography (EEG) Power in the Gamma Band (30 to 70 Hz) From Rest to the Active and Passive Stages
EEG power (microvolts squared) in the gamma band (30 to 70 Hz) at each of 56 scalp electrodes, calculated as the active-stage value minus the resting-stage value and as the passive-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10), active stage (minutes 10 to 20) and passive stage (minutes 20 to 70) of the 70-minute recording in the single laboratory visit
Change in Sensor-Level Electroencephalography (EEG) Power in the Total Band (0.1 to 70 Hz) From Rest to the Active and Passive Stages
EEG power (microvolts squared) in the total band (0.1 to 70 Hz) at each of 56 scalp electrodes, calculated as the active-stage value minus the resting-stage value and as the passive-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10), active stage (minutes 10 to 20) and passive stage (minutes 20 to 70) of the 70-minute recording in the single laboratory visit
Change in Source-Level EEG Functional Connectivity (Weighted Phase Lag Index) in the Delta Band (0.1 to 4 Hz) From Rest to the Active Stage
Weighted phase lag index (wPLI; unitless, range 0 to 1) in the delta band (0.1 to 4 Hz) for each of 1,035 pairs among 46 cortical regions of the Desikan-Killiany atlas, estimated from source-reconstructed EEG, calculated as the active-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10) and active stage (minutes 10 to 20) of the 70-minute recording in the single laboratory visit
Change in Source-Level EEG Functional Connectivity (Weighted Phase Lag Index) in the Theta Band (4 to 8 Hz) From Rest to the Active Stage
Weighted phase lag index (wPLI; unitless, range 0 to 1) in the theta band (4 to 8 Hz) for each of 1,035 pairs among 46 cortical regions of the Desikan-Killiany atlas, estimated from source-reconstructed EEG, calculated as the active-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10) and active stage (minutes 10 to 20) of the 70-minute recording in the single laboratory visit
Change in Source-Level EEG Functional Connectivity (Weighted Phase Lag Index) in the Alpha Band (8 to 12 Hz) From Rest to the Active Stage
Weighted phase lag index (wPLI; unitless, range 0 to 1) in the alpha band (8 to 12 Hz) for each of 1,035 pairs among 46 cortical regions of the Desikan-Killiany atlas, estimated from source-reconstructed EEG, calculated as the active-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10) and active stage (minutes 10 to 20) of the 70-minute recording in the single laboratory visit
Change in Source-Level EEG Functional Connectivity (Weighted Phase Lag Index) in the Low Alpha Band (8 to 10 Hz) From Rest to the Active Stage
Weighted phase lag index (wPLI; unitless, range 0 to 1) in the low alpha band (8 to 10 Hz) for each of 1,035 pairs among 46 cortical regions of the Desikan-Killiany atlas, estimated from source-reconstructed EEG, calculated as the active-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10) and active stage (minutes 10 to 20) of the 70-minute recording in the single laboratory visit
Change in Source-Level EEG Functional Connectivity (Weighted Phase Lag Index) in the High Alpha Band (10 to 12 Hz) From Rest to the Active Stage
Weighted phase lag index (wPLI; unitless, range 0 to 1) in the high alpha band (10 to 12 Hz) for each of 1,035 pairs among 46 cortical regions of the Desikan-Killiany atlas, estimated from source-reconstructed EEG, calculated as the active-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10) and active stage (minutes 10 to 20) of the 70-minute recording in the single laboratory visit
Change in Source-Level EEG Functional Connectivity (Weighted Phase Lag Index) in the Beta Band (12 to 30 Hz) From Rest to the Active Stage
Weighted phase lag index (wPLI; unitless, range 0 to 1) in the beta band (12 to 30 Hz) for each of 1,035 pairs among 46 cortical regions of the Desikan-Killiany atlas, estimated from source-reconstructed EEG, calculated as the active-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10) and active stage (minutes 10 to 20) of the 70-minute recording in the single laboratory visit
Change in Source-Level EEG Functional Connectivity (Weighted Phase Lag Index) in the Low Beta Band (12 to 20 Hz) From Rest to the Active Stage
Weighted phase lag index (wPLI; unitless, range 0 to 1) in the low beta band (12 to 20 Hz) for each of 1,035 pairs among 46 cortical regions of the Desikan-Killiany atlas, estimated from source-reconstructed EEG, calculated as the active-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10) and active stage (minutes 10 to 20) of the 70-minute recording in the single laboratory visit
Change in Source-Level EEG Functional Connectivity (Weighted Phase Lag Index) in the High Beta Band (20 to 30 Hz) From Rest to the Active Stage
Weighted phase lag index (wPLI; unitless, range 0 to 1) in the high beta band (20 to 30 Hz) for each of 1,035 pairs among 46 cortical regions of the Desikan-Killiany atlas, estimated from source-reconstructed EEG, calculated as the active-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10) and active stage (minutes 10 to 20) of the 70-minute recording in the single laboratory visit
Change in Source-Level EEG Functional Connectivity (Weighted Phase Lag Index) in the Gamma Band (30 to 70 Hz) From Rest to the Active Stage
Weighted phase lag index (wPLI; unitless, range 0 to 1) in the gamma band (30 to 70 Hz) for each of 1,035 pairs among 46 cortical regions of the Desikan-Killiany atlas, estimated from source-reconstructed EEG, calculated as the active-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10) and active stage (minutes 10 to 20) of the 70-minute recording in the single laboratory visit
Change in Source-Level EEG Functional Connectivity (Weighted Phase Lag Index) in the Total Band (0.1 to 70 Hz) From Rest to the Active Stage
Weighted phase lag index (wPLI; unitless, range 0 to 1) in the total band (0.1 to 70 Hz) for each of 1,035 pairs among 46 cortical regions of the Desikan-Killiany atlas, estimated from source-reconstructed EEG, calculated as the active-stage value minus the resting-stage value, and compared between the BELL-001 and sham groups.
Resting stage (minutes 0 to 10) and active stage (minutes 10 to 20) of the 70-minute recording in the single laboratory visit
Application with the same appearance and screen flow as BELL-001 that plays a single fixed pre-recorded sound not linked to breathing, with brief tones at random intervals during the active stage to match the touch task.
Registry-reported names are shown in full. Names without a link have not yet been matched to a Syfrah profile.
Contact the study team to confirm eligibility.
Study Start
November 18, 2025
Actual (reported)Primary Completion
February 13, 2026
Actual (reported)Study Completion
February 20, 2026
Actual (reported)Last Updated
October 22, 2026
Record last verified: 2026-10
De-identified participant-level values of the derived measures used in the published analyses (electroencephalography (EEG) spectral power, EEG functional connectivity and heart rate variability features) will be shared on the Open Science Framework upon acceptance of the resulting manuscript. Raw EEG, electrocardiography (ECG) and respiration recordings may be shared with qualified researchers under a data-use agreement, where permitted by participant consent and the data-sharing policies of the Korea Institute of Science and Technology and Hanyang University.
Download the complete ClinicalTrials.gov record imported by Syfrah, including fields not shown in this page. The registry may contain a newer revision.
Source: ClinicalTrials.gov.
JSON downloads include source attribution and retrieval details alongside the unchanged original record or the separately normalized values.
Source retrieved: