Brainloop

4. BCI - Brain–computer interfaces
	4.2. Present-day BCIs   4.2.5. Cortical neuronal action potentials

Since the 1960s, metal microelectrodes have been used to record action potentials of single neurons in the cerebral cortices of awake animals during movements. A few studies have explored the capacity of animals to learn to control neuronal firing rates and showed that monkeys could learn to control the discharge of single neurons in motor cortex.
A humans nearly locked-in has learned to control neuronal firing rates and uses this control to move a cursor to select icons or letters on a computer screen.
By using neuronal activity to control one dimension of cursor movement and residual EMG control to control the other dimension and final selection, communication rates up to about 3 letters/min (i.e. about 15 bits/min) have been achieved.

4. BCI - Intro
	4.1. Definition and features of a BCI   4.1.1. Dependent and independent BCIs 4.1.2. BCI use is a skill 4.1.3. The parts of a BCI   4.1.3.1. Signal acquisition 4.1.3.2. Signal processing: feature extraction 4.1.3.3. Signal processing: the translation algorithm 4.1.3.4. The output device 4.1.3.5. The operating protocol 4.2. Present-day BCIs   4.2.1. Visual evoked potentials 4.2.2. Slow cortical potentials 4.2.3. P300 evoked potentials 4.2.4. Mu and beta rhythms   4.2.4.1. The Wadsworth BCI 4.2.4.2. The Graz BCI 4.2.5. Cortical neuronal action potentials 4.3. The future of BCI-based communication

---

Source: Brain–computer interfaces for communication and control, Clinical Neurophysiology 113 (2002) 767–791, Jonathan R. Wolpaw, Niels Birbaumer, Dennis J. McFarland, Gert Pfurtscheller, Theresa M. Vaughan