wet · 12 labels
15.3%
15.3% balanced accuracyParticipant-mean balanced accuracy
95% interval 13.7%–17.1%Adaptation budget · common future test blocks
How much calibration?
More labeled trials do not have one universal value. On this 12-target wearable protocol, target-only spectral ridge improves gradually but remains low; single-band ensemble TRCA benefits sharply on wet recordings from 24 to 48 labels and only modestly on dry recordings.
Target-only spectral ridge
Twelve, 24 and 48 labeled trials
Each target-person model uses only that person's labeled calibration recordings. All conditions are evaluated on the same future blocks 5–10, across 102 people and 7,344 test trials.
wet · 24 labels
18.2%
18.2% balanced accuracyParticipant-mean balanced accuracy
95% interval 15.9%–20.7%wet · 48 labels
21.3%
21.3% balanced accuracyParticipant-mean balanced accuracy
95% interval 18.3%–24.5%dry · 12 labels
15.0%
15.0% balanced accuracyParticipant-mean balanced accuracy
95% interval 13.6%–16.5%dry · 24 labels
18.3%
18.3% balanced accuracyParticipant-mean balanced accuracy
95% interval 16.3%–20.7%dry · 48 labels
21.8%
21.8% balanced accuracyParticipant-mean balanced accuracy
95% interval 19.3%–24.5%Trial counts are not minutes. The export records 12, 24 or 48 labeled trials per person. It does not convert those trials into a setup-time claim.
A separate arm
Zero target labels is source-only transfer
The zero-budget spectral-ridge rows use a model trained on other participants from one source sensor. They are not the starting point of an incremental target-person update curve.
| Training source | Test on wet | Test on dry |
|---|---|---|
| wet sourceother participants only | 39.2%35.4%–43.0% | 35.4%32.0%–39.0% |
| dry sourceother participants only | 39.0%35.2%–42.9% | 37.2%33.6%–40.9% |
Method matters
Single-band eTRCA responds differently
The target-only eTRCA arm uses 24 or 48 calibration trials and the same future test blocks. CCA at zero labels is an analytic reference, not an unfitted eTRCA point.
+24.2 ppWet eTRCA: 48 labels minus 24 labels
A larger budget helped wet recordings much more
The paired participant interval is +19.7 to +28.8 pp. Dry eTRCA improved by +2.7 pp, interval +1.3 to +4.2 pp. Neither calibrated point exceeds the analytic CCA reference on the same test blocks.
| Method and labeled target trials | Wet | Dry |
|---|---|---|
| CCA · 0 labelsanalytic target-recording reference | 81.1%77.2%–84.8% | 65.1%60.0%–70.0% |
| Single-band eTRCA · 24 labelstarget-only adaptation | 39.8%34.1%–45.8% | 12.4%10.4%–15.0% |
| Single-band eTRCA · 48 labelstarget-only adaptation | 64.0%57.1%–70.7% | 15.1%12.6%–18.2% |
Methods & limits
What the budget curves do and do not mean
All results are tied to a frozen protocol. They describe labeled-trial budgets for these methods, sensors and fixed test blocks.
Different zero-label mechanisms
Spectral ridge at budget zero is source-only supervised transfer. CCA at budget zero is an analytic target-recording reference. Neither should be connected to target-only points as one model's learning curve.
Target-only means target-only
The 12/24/48 ridge and 24/48 eTRCA conditions fit only the target person's calibration recordings, then test on blocks 5–10.
Prespecified single band
eTRCA uses the fixed 6–80 Hz prepared band. It is not the reference repository's three-filter-bank experiment, so low scores do not establish an optimized TRCA ceiling.
Descriptive intervals
Bootstrap intervals summarize participants under this protocol. They do not add retraining uncertainty or establish general calibration requirements for another headset.
An Open Dataset for Wearable SSVEP-Based Brain-Computer Interfaces
Zhu, F., Jiang, L., Dong, G., Gao, X., & Wang, Y. (2021). An Open Dataset for Wearable SSVEP-Based Brain-Computer Interfaces (Version 4) [Data set]. Figshare. https://doi.org/10.6084/m9.figshare.13560281.v4
Zhu, F., Jiang, L., Dong, G., Gao, X., & Wang, Y. (2021). An Open Dataset for Wearable SSVEP-Based Brain-Computer Interfaces. Sensors, 21(4), 1256. https://doi.org/10.3390/s21041256
Official Tsinghua author mirror snapshot acquired 2026-09-20 · CC BY 4.0. Aggregate results only.
Data source: reviewed aggregate JSON · schema bci-report-public-deployment-topics-v1 · generated 2026-09-20.