This book, volume 6 in a series on advances in brain-computer interface (BCI) research, includes contributions from nominees for a yearly BCI research award sponsored by G.TEC, a company offering systems for BCI research. This particular volume consists of nine such chapters, plus an introduction, an overview of trends, and recent advances in BCI.
Not being deeply involved myself in BCI research, I am nevertheless interested in advances in the field, in particular the use of lower-cost BCI devices in the context of human-computer interaction and the use of artificial intelligence methods for the analysis of BCI signals. All of the contributions in the book describe approaches that require significantly more advanced system configurations, but several of them certainly provided valuable insights for me. The authors are also encouraged to present their work in a way that is accessible to a broader audience.
The contribution by the eventual winners of the award, Gaurav Sharma et al., describes a successful attempt to return some capabilities to people with paralyzed limbs. By connecting a cortical-implant microelectrode array to a neuromuscular electrical stimulation cuff, they enabled a patient with a spinal cord injury resulting from a diving accident to regain some of the lost hand and finger movement capabilities. The study involved a single patient and a complex setup in terms of hardware, software, and participant training. Thus, it can’t be replicated outside of a research setting, but the authors are working toward adaptations of the approach for home use.
The runners-up for the award, Sharlene Flesher et al., describe the use of intracortical BCIs for the control of prosthetic devices. One limitation of this approach is the limited feedback, typically restricted to visual observation of the intended movement. The authors discuss the use of intracortical microstimulation to provide somatosensory feedback, relaying tactile information from sensors in the robotic limb. In experiments, this allowed a user to receive intuitive feedback for individual finger movements, and to improve the controlled use of force when manipulating objects.
Multiple other contributions also focused on improvements to the use of cortically implanted sensors and actuators.
Two contributions examine ways to improve the processing of signals. J. Thielen et al. use reconvolution to learn responses from stimuli. By a careful consideration of the underlying dynamics, the induced model can generalize to novel stimuli and improve overall system performance. In addition, it can be used without prior information (“zero-training”), leading to potential plug-and-play solutions. M. Schultze-Kraft et al. describe experiments to detect intentions to perform movements in real time during an ongoing electroencephalogram (EEG) session. The EEG signals used, readiness potential and lateralized readiness potential, can predict a motor intention several hundred milliseconds before the respective actual movement. While the practical applications of this approach are somewhat limited, it enables future experiments to investigate novel questions from cognitive neuroscience.
In the end, I did not find immediately applicable information for my context, although the “visual cue-guided rat cyborg” described by Y. Wang et al. could open up interesting avenues to explore human-rat robot interaction. Nevertheless, I enjoyed the opportunity to learn about recent advances in brain-computer interaction, and I can recommend this volume as a vehicle to do so.