Speech Strategies

Cochlear implants are capable of delivering stimulation in several different ways, each of which is called a speech strategy.  Different implants have offer different speech strategies.  The speech strategies offered with each implant are as follows:

Nucleus 22: SPEAK
Nucleus 24: SPEAK, ACE, CIS
Clarion: SAS, MPS, CIS
Clarion CII: High Res Broadband
Med-El: CIS+

I. Explanation of Speech Strategies

A brief description of each speech strategy and how it works is as follows:

CIS (Clarion, Med-El, N22, N24): CIS because its probably the most straightforward. Basically, it uses the same set of 6-8 electrodes (any
additional ones available are simply not used) and fires them in sequence (one after the other) to deliver sound to different parts of the cochlea in a way that reflects the original sound signal.  It's stimulation rate can be as high as 20,000 pulses per second, depending on the device.  Med-El uses what it calls "CIS+", which is effectively is CIS with a slightly enhanced processing algorithm.  CIS stands for Continuous Interleaved Sampler.

SAS (Clarion only): Fires all of the electrodes simultaneously, which is how sound is delivered in the normal ear (to all parts of the cochlea at once). The advantages of this is that this more closely resembles how sound is delivered in the normal ear, and more sound information can be transmitted with each sweep of the cochlea (up to 100,000 updates of sound per second). The disadvantage is that the stimulation from one electrode can interfere with the stimulation from another, in what is known as "channel interaction." The Clarion HiFocus electrode was designed to minimize channel interaction dramatically, but not eliminate it altogether.  SAS uses bipolar simulation (i.e it pairs electrodes for stimulation) and thus uses 16 electrodes to deliver 8 channels.  SAS stands for Simultaneous Analog Stimulation.

MPS (Clarion only): This is a hybrid between CIS and SAS, in an attempt to gain some of the theoretical benefits of SAS without the drawbacks of channel interaction. In MPS, two electrodes or more are stimulated simultaneously, but they are far enough apart so as not to experience channel interaction.  Like SAS, MPS uses bipolar stimulation and so uses 16 electrodes to deliver 8 channels.  MPS is also sometimes known as "PPS".

SPEAK and n-of-m (N22, N24): The theory behind SPEAK (also called n-of-m) speech strategies is to accentuate the strongest sound frequencies (spectral maxima) and repress others, in an attempt to bring out the most pertinent sound clues necessary to understand speech.  It does this by selecting a different set of up to 8 electrodes with each sweep of the electrode array. This may be particularly helpful in places where there is a lot of background noise, but the drawback of this is that it is confined to a relatively slow stimulation rate and, by accentuating certain frequencies over others, it is a deliberate distortion of the original sound signal.  An experimental form of n-of-m has been tried with Clarion by independent researchers, who reported promising results.  SPEAK is short for for "Spectral PEAK extraction".

ACE (N24 only): ACE is a hybrid between CIS and SPEAK, much like MPS is a hybrid between CIS and SAS.  ACE combines the higher stimulation rate of CIS with the way SPEAK highlights certain frequencies.  Typically, with each sweep of the electrode array, ACE activates between 6-12 electrodes out of 22 available ones.  ACE stands for Advanced Combination Encoders.

High Resolution (Clarion CII only): This is an experimental platform that uses higher stimulation rates than have ever been possible with any implant, and more channels than the previous generation Clarion device.  In theory, the high stimulation rate offers the advantages of SAS without the disadvantage of channel interaction.  Researchers are experimenting with the best way to deliver all this additional sound information, so there may be several variations of high resolution to come.

On top of all the differences in speech strategies, there is obviously the individualized customization and optimization that the audiologist goes through with each person.  For instance, the exact stimulation rate or number of electrodes used may vary to some degree.

Theoretically, it is possible to run speech strategies like ACE or SPEAK on a Clarion.  Only one small public attempt has ever been made to run a speech strategy similar to SPEAK on the Clarion, which yielded mixed results.  Because SAS and MPS require capabilities that the Nucleus and Med-El do not have, it is not possible to run those speech strategies on those devices.

II. Speech Strategy Preferences

There's no way to predict ahead of time which speech strategy is best for any one person.  Different people do best with different speech strategies and the only way to know for sure which one is best for you is to try them! What we do know is that, on the whole, users have shown a preference for certain strategies over others. Here are the breakdowns for strategy preference among Nucleus 24 and Clarion HiFocus users:

Clarion CII Bionic Ear (source: Advanced Bionics study):
96% High resolution
4% SAS,PPS, or CIS

Nucleus 24 (source: Cochlear Corp study, Apr 1999):
61% ACE
23% SPEAK
8% CIS
8% No preference/Not sure

Naturally, no comparisons exist between ACE or SPEAK vs. SAS, MPS, or High Resolution, since these speech strategies are not available on the same device to enable users to compare them against one another.