The most commonly used EEG control signals are visually evoked potentials (VEP), slow cortical potentials (SCP), event related potentials (ERP), and sensorimotor rhythms.
VEPs are modulations in brain activity that occur in response to a visual stimulus. A subcategory known as steady-state visual evoked potentials (SSVEP) is popular in BCI systems. SSVEP-based BCIs allow users to select a target by focusing their gaze. One benefit of this BCI is that very little user training is required. However, a significant drawback is that the user must keep his eyes fixed on the desired target. Many users have cited fatigue as a limiting factor in SSVEP-based BCIs.
SCPs are voltage shifts in EEG activity. Negative SCPs correlate with increased brain activity, while positive SCPs indicate decreased activity. With sufficient training, people can learn to regulate changes in SCP that can be used for tasks such as cursor control. However, information transfer rate is poor, and studies have revealed that not all people are capable of achieving adequate control. Training periods for SCP-based BCIs are often last for weeks or months.
Event related potentials are fluctuations in EEG that result from presented stimuli, which can be visual, auditory or somatosensory. A typical application of ERP-based BCIs is a digital keyboard, which presents a matrix of letters and numbers. Users focus their attention on the desired target; when the target is stimulated, an ERP response occurs. An advantage of ERP-based BCIs is that very little training is required. However, challenges associated with signal-to-noise ratios have typically limited detection accuracy, hampering the performance of ERP-based BCIs.
Sensorimotor rhythms, also known as motor imagery, are oscillations in brain activity found in the mu and beta frequency bands. The amplitude of the waves varies when a person mentally-performs a motor task (such as moving their hands or feet), although no physical movement is required. With training, people can learn to generate these sensorimotor rhythm modulations to control connected devices, ranging from computer cursors to prosthetic limbs. Accordingly, sensorimotor rhythms are a major subject of BCI research.