Basilar Membrane Properties and Auditory Signal Processing

Properties of the Basilar Membrane

The basilar membrane has varying physical properties along its length:

• Base: Thick, stiff, and narrow (ideal for high frequencies).
• Apex: Thin, less stiff, and wide (ideal for low frequencies).

High frequencies do not move fluid effectively, whereas low frequencies do. Therefore, it is more efficient for high frequencies to cross the membrane early (at the base) and for low frequencies to cross later (at the apex).

Evidence for Active Amplification in the Inner Ear

Experiments measuring sound energy input and output provide evidence for active amplification within the inner ear:

1. Sound energy output was observed to be higher than sound energy input, suggesting an active amplification process.

Outer hair cells are likely responsible for this amplification:

1. They are electromotile.
2. Their contraction changes the relative position of the tectorial and basilar membranes.
3. This movement stretches the inner hair cells to a greater degree than would otherwise occur.

Sound Encoding Processes

Hair cells at different locations on the basilar membrane encode specific frequencies.

Feedback Mechanisms in Hearing

Notch Filtering

Used in the pinna to detect the elevation of sound.

Negative Feedback

Provides protection from loud sounds. Loud sounds cause contraction of muscles in the middle ear, reducing sound transmission.

Positive Feedback

Active amplification by the outer hair cells.

Binaural Hearing: How Signals from Both Ears are Used for Localization

The ears detect slight differences in:

1. Timing: Effective for low frequencies (<1.5 kHz) because nerve firing is synchronized with the cycle-by-cycle structure of the acoustic waveform. This makes it easier to detect differences in spike timing between the two ears.
2. Intensity: Effective for high frequencies (>3 kHz). Low frequencies wrap around the head, whereas high-frequency sounds create a sound shadow behind the head.

Phase of a Sinusoidally Modulated Sound

• Only possible for low-frequency sounds.
• Coding: At every peak of the phase, the nerve fires.
• Firing rate is dependent on loudness (louder sounds = higher amplitudes = higher firing rate at the peak).

Auditory Nerve Responses: 1kHz vs. 10kHz Tone

• 1 kHz: Firing in synchrony with the phase.
• 10 kHz: Firing continuously (not in synchrony with the phase).

Low frequencies are easier to localize based on timing differences. The firing for high-frequency sounds is not synchronized with the phase, so there would not be a difference in the timing of firing between the two ears. Localization of high-frequency sounds relies on intensity differences.

Cochlear Implant Sound Processing

A cochlear implant requires:

1. A microphone.
2. The ability to encode sound into the "language" of the cochlea (decomposition into different frequencies).
3. Electrodes placed accordingly in the cochlea (electrodes for high frequencies at the base and electrodes for low frequencies at the apex).

Decomposition occurs in a "stimulation sequence."

Cochlear implant coding differs from normal coding:

• Stimulation of all hair cells within a particular frequency bandwidth occurs, so there are no subtle differences in the firing of individual hair cells.
• Spikes are not synchronized with the phase of a particular signal.

Why Low-Frequency Sounds Mask High-Frequency Sounds

Loud, low-frequency sounds also slightly move the base of the basilar membrane, where high-frequency sounds are detected. This does not happen at the apex of the membrane with loud, high-frequency sounds.