BCIwiki (bciwiki.com) — A real-time spatial filtering pipeline that combines spatio-spectral decomposition (SSD) with beamforming held motor-imagery BCI decoding accuracy at 76 ± 9% while frequency-tuned amplitude-modulated transcranial alternating current stimulation (AM-tACS) was running, compared with a drop to near chance (58 ± 9%) under a standard Laplacian filter. The work by Vermehren M, Peekhaus N, Colucci A, Wiskow M, Haslacher D and Curio G was reported in IEEE Transactions on Neural Systems and Rehabilitation Engineering on September 10, 2026.
Stimulation and decoding have been hard to combine because artifacts generated by the stimulation current overlap in frequency with endogenous brain rhythms, destabilizing real-time decoding. The researchers compared the two spatial filters in 14 healthy participants, assessing BCI control both with and without stimulation. Without stimulation, both pipelines performed comparably: 73 ± 9% for SSD-beamforming and 72 ± 8% for the Laplacian filter (p = .849). With AM-tACS on, Laplacian filtering declined significantly (p < .001), while SSD-beamforming preserved control at 76 ± 9%.
The authors write that the results show robust BCI control during frequency-tuned AM-tACS is achievable, and that enabling simultaneous stimulation and decoding establishes a paradigm for testing the causal contributions of brain rhythms during ongoing BCI control and for advancing stimulation-informed, bi-directional BCI interventions. They also state that future work will determine how AM-tACS can be leveraged to enhance BCI performance and to promote neuroplasticity during restorative BCI applications.