CI Atlas · Speech-Coding Strategies: The Complete Lineage · Module 02

2One Electrode, One Sound: The Single-Channel Era

The first commercial cochlear implants used a single intracochlear electrode and a single processing channel. This module dissects the 3M/House signal chain, shows why one channel cannot exploit the cochlea's tonotopic map, and frames the leap to multichannel place coding as the field's first great supersession.

FThe 3M/House device

The House/3M single-electrode device was the first FDA-approved cochlear implant in 1984 and had several hundred users It was developed by House and Urban in the early 1970s and manufactured by the 3M Company It provided little more than a sensation of sound and sound cadences and was useful mainly as a lip-reading aid and an alert to acoustic events On a historical comparison the 1980 3M House device scored only about 20% sentence recognition in quiet.[1997][2008]

~100Pulse rate (pps) of earliest single-channel implants vs up to 2941 pps in CIS [1993]

+37Percentage-point CIS gain on SPIN sentences over analog coding [1993]

TThe single-channel signal chain

The processing chain was: microphone to amplifier to a band-pass filter (340-2700 Hz) to amplitude-modulation of a 16 kHz carrier to output amplifier to external coil to a single active electrode in scala tympani A single band-passed, amplitude-modulated analog carrier conveyed the speech waveform with no spectral or place decomposition There was no exploitation of cochlear place coding at all Only one single-channel processor drove one implanted cochlear electrode.[2010][1997]

TWhy one channel was not enough

A single channel neglects the place-dependent temporal detail of the cochlear nerve, because different fibres carry different temporal features Because a single site cannot convey place (spectral) information, the device gave essentially no open-set speech recognition except in a few subjects Multichannel/multisite systems in the 1980s gave significantly higher speech reception on average than single-channel predecessors This motivated multi-channel implants that stimulate different nerve populations with different features.[1991][1999]

FParallel multichannel beginnings

The modern era of cochlear implants is traced to Djourno and Eyries in France in 1957 Early multichannel intracochlear stimulation in man was reported by Eddington and colleagues in 1978, establishing the loudness-growth and dynamic-range groundwork Early Vienna work explored percepts elicited by different speech-coding strategies in multichannel devices These threads converged into the multichannel waveform strategies of the 1980s.[1978][1983]

TBy the numbers

FHear it

This is the core idea made audible. The chosen sound is split into 4 frequency bands; each band's envelope is extracted and re-imposed on a band of noise — just as a cochlear implant drives 4 electrodes. With one or two channels speech blurs; by 4–8 it becomes intelligible; more brings diminishing returns — the same plateau the charts show. Try the real sentence, or load your own clip (a WAV/MP3) to vocode actual speech. The bundled sentence is a synthesised text-to-speech sample (two standard Harvard sentences). Synthesised/processed live in your browser; raise the volume gently.

Case 14.2 · One Electrode, One Sound

A historian reviewing CI records finds that a 1984-era single-channel recipient could detect when someone was speaking and use it to support lip-reading, but scored near zero on open-set word lists in quiet.

Which engineering limitation best explains this performance pattern?

Self-assessment — Module 22 questions

Question 1

What carrier did the 3M/House single-channel processor amplitude-modulate?

Question 2

What was the band-pass filter range in the 3M/House processor?

Tracked locally in your browser — see /progress for the dashboard.