View article

Cochlear implants (CIs) are prosthetic devices that restore hearing in deaf individuals via electric stimulation of the auditory nerve through an electrode array inserted in the cochlea. The device consists of a microphone and an externally worn speech processor that converts acoustic signals into electric signals. The speech signal coded with electric pulses is then delivered via a wireless transmission system through the scalp to an electrode array, implanted in the cochlea, traditionally in the scala tympani (Grayden & Clark, 2006). While research with CIs dates back to the 1950s (Djourno & Eyriès, 1957), the Food and Drug Administration (FDA) has approved CI use in adults in 1984, in children 2 years and older in 1989, and in children 12 months and older in 2000. The FDA reports that approximately 324,000 people worldwide had received CIs by 2012.

Despite the relatively long history of the implant, the speech signal transmitted via the modern CIs is still inherently degraded in fine spectrotemporal details (eg, Loizou, 1998; Rubinstein, 2004). The device mainly delivers slowvarying amplitude envelopes of speech modulating (usually) fixed-rate digital pulses, delivered at a small number of contact points (electrodes). This degraded signal is recognized and reinterpreted by the brain as speech (Fu, 2002; Shannon, Zeng, Kamath, Wygonski, & Ekelid, 1995). One of the main forms of degradation is the reduced spectral resolution (Friesen, Shannon, Baskent, & Wang, 2001; Fu & Nogaki, 2005; Henry, Turner, & Behrens, 2005). This reduction does not come from the small number of electrodes per se but instead from the channel interactions caused …