What electrical stimuli do brain implants use?

Question

I was reading about artificial eyes and came to think about how the brain works. More specifically, what "signals" it uses in the case of cortical visual prosthetics in blind people? Cortical prosthetics apply current stimulations through electrodes placed on the surface of the visual cortex.

Now suppose I would want to let a blind person wearing a cortical prosthesis to see the color red, what signal would I send through the electrodes? Would a sine wave of 200 Hz do the job?

Answer 1

Short answer
The track record of cortical visual prosthetics is limited. However, a lot more is known on auditory prostheses (mainly cochlear implants) and retinal prostheses. In these implants, biphasic, charge balanced pulse trains are generally applied, mainly for safety reasons and to reduce current spread through the neural tissue. Sine waves have been tested in cochlear implants, but have long since been abandoned as useful stimulus. Color perception in visual cortical prosthetics is uncovered ground so far, but in retinal implants some limited data is available that color perception can be manipulated by altering the shape and frequency of the electrical pulse trains.

Background
First off, neural prosthetics (including auditory and visual prosthetics, as well as pace makers) in general operate through biphasic pulses instead of sine waves.

Biphasic pulses have the advantage that they can be very short (in the order of tens of microseconds in the case of cochlear implants, and hundreds of microseconds in retinal implants). This is advantageous, because the typical biphasic pulse has two identical phases but of opposite polarity. This means that the injected current is quickly neutralized within microseconds. Note that direct current is damaging to the delicate neural tissues. That's why charge-balanced pulses are used in modern cochlear implants to avoid direct current (DC) stimulation that may damage neural tissues (Bahmer & Baumann, 2013).

Sine waves have been used in cochlear implants (Clark, 2006), as most of the acoustic speech information is conveyed in frequencies between 500 and 400 Hz. Indeed, the auditory nerve does show phase locking when the electrical (or acoustic) stimulus is about 100 Hz or lower. However, biphasic pulse train are the norm nowadays, because it is safer, more power efficient and more effective, as pulses on adjacent electrodes can be alternated to prevent electrical current to summate on closely spaced electrodes. This is called interleaving and has been used widely since the 1990's as it came known as the continuous-interleaved sampling (CIS) strategy (Wilson et al., 1993). A CIS-like strategy has been adopted in retinal implants too, for example the Argus II prosthesis.

Visual prosthetics in general deliver visual perceptions on a gray scale. Cortical prosthetics have not been investigated much. However, retinal implants are currently commercially available from at least two companies. In retinal implants, phosphenes appear mostly as white spots of light, but yellow has been reported too (Stronks & Dagnelie, 2014b) as well as red to orange phosphenes (Humayun et al., 2003). Interestingly, in retinal prostheses it has indeed been shown that by carefully adjusting the pulse rate and pulse shape some crude form of color perception could be induced (Stronks & Dagnelie, 2014a). However, the only consistent result seem to have been that high-pulse rates resulted in blue phosphenes when the stimulation was stopped (OFF response) (Humayun et al., 2003). The thought is that different stimulus characteristics stimulate a different set of fibers in the retina but this is very preliminary at this stage. In cortical implants research has not even touched upon color sensations as yet, as far as I am aware.

<sub>References
- Bahmer & Baumann, Hear Res, 306: 123-30
- Clark, Philos Trans R Soc Lond B Biol Sci (2006); 361(1469): 791–810
- Humayun et al., Vis Res (2003); 43(24): 2573-81
- Stronks & Dagnelie, Exp Rev Med Dev (2014a); 11(1): 23-30
- Stronks & Dagnelie, Encyclopedia of Computational Neuroscience (2014b): 1-4
- Wilson et al., J Rehabil Res Dev (1993); 30(1): 110-6</sub>

Answer 2

This question is frankly a bit vague, but generally speaking (i.e. answering the question from the title) the brain uses electrochemical signaling. The trouble is, of course, we are far from fully understanding what goes on in there, or how those signals achieve all that we can do. Presently we can simulate tiny fragments of a rat's brain, as reported in Nature news, 2015.

To answer the question form the body, "if I hooked up some electrode[s] to a blind persons visual cortex and wanted them to see the color red, what would i send through the el[e]ctrodes?" The answer is we don't know. That's because, as Conway et al. 2007, state:

On a gross level, it remains disputed whether color is localized to a particular brain region; on a microscopic level, it is uncertain what contribution single cells make to the perception of specific hues. While some brain-imaging studies have suggested that color processing may be localized within the extrastriate brain [...], single-cell electrophysiological studies, which have higher spatial and temporal resolution than imaging, have produced conflicting results [...] and cast doubt on the notion of a specialized color center [...].

Now if you're asking about artificial eyes... there are many kinds of proposed prosthesis, but if we talk about[bypassing the optical nerve completely such as bionic eye brain implants as reported by MIT Technology Review, 2017, then currently that only works for black and white... and even then

The patient was able to see spots of light with no significant adverse side effects. [...]

“If we could figure out how to process and filter visual information to correctly stimulate the electrodes we could eventually improve the type of image that person will be able to perceive,”

So it seems to be an open problem, even without color.

_{Conway, B. R., Moeller, S., & Tsao, D. Y. (2007). Specialized color modules in macaque extrastriate cortex. Neuron, 56(3), 560-573.}