A key consideration for developing neural prostheses is understanding of the electrochemistry of a stimulating electrode. In general, one device will out-perform another if it is more selective in targeting excitation to specific populations, avoiding damage to the tissues in the vicinity of the electrode, and minimizes degradation of the stimulating electrode, so having a longer lifetime. The next generation of devices will have more current injection sites. The injection sites (electrode contacts) will be smaller, both positive and negative currents will be injected and will try to extend the reach of the injected currents to compensate for imprecise electrode placement. In all probability we will move charge carrying capacity of electrodes to new levels.
The purpose of neural prostheses is to create propagated action potentials. Action potentials can be initiated on nerves by injecting current into the tissue medium that creates a potential difference along the axon that is sufficient to cause enough voltage-gated sodium ion channels to open and result in a propagating action potential. (Although the release of the transmitter is the usual intended result of a propagating action potential, blocking the release could also be a possible intended goal. Blocking techniques usually require more charge than is required for action potential initiation.)
Current is carried in a metal wire by electron migration, while in the tissue medium current is carried by ion migration. At the electrode/electrolyte interface a conversion between charge carriers occurs. The electrochemistry at the interface can result in the creation of chemical species that do not previously exist. These new species can in turn react with cells to alter their function or even injure them or result in the formation of soluble metal oxides that results in loss of electrode material, corrosion, and failure of the device as a whole.
The mechanisms of deleterious effects on tissues from electrical stimulation are not well understood. Two possible candidates for these effects are over-activation of target neurons and the formation of reactive species at the electrode/electrolyte interface during stimulation. These are not mutually exclusive. Understanding the electrochemistry of charge injection for various materials will facilitate the design of experiments that would elucidate the mechanism of stimulation induced tissue injury and reduce our dependency on expensive animal studies.

A key element for increasing electrode selectivity is minimizing the distance between the stimulating electrode and the excitable target tissue. Unfortunately, this objective may not be easily met, which means that larger current will have to be injected in order to cause the desired excitation.
Using a point source approximation for a stimulating electrode and calculating the resulting potential, enables us to illustrate the importance of electrode-target separation. The potential in the region of the target tissue is proportional to the current injection at the site of the electrode and inversely proportional to the separation distance between the electrode and the target.
This means that the closer the electrode is to the target tissues the lower is the stimulus amplitude required to activate the target population. Conversely, the further the electrode is from the target tissue the greater the stimulus required to activate the target population.
To accommodate for the inability to precisely positioning an electrode on its target, it is desirable to be able to inject larger currents without damage to the tissue and the electrode and to reshape the induced potential field by current injection from adjacent electrodes, “field steering”. The techniques to accomplish this will most likely require smaller electrodes, more injection sites and higher current densities.