Lecture 18

Neuroethology
(Lecture 18 - Auditory filters)

"There is nothing in the intellect that was not previously in the senses."

Aristotle (384 - 322 BC)

I chose this quotation to begin this lecture because it highlights the fact that all of our informations about the world comes to us through our senses. But not all of that information is useful - there is a lot of noise that often masks the signal we are trying to detect.

This picture shows how hard it can be to pick the interesting stuff out of the useless noise that bombards our sensory systems. The example below shows a similar phenomenon, but with sound rather than images.

How do you filter out the noise?

Peripheral filters
Tune your sensory system to the specific frequencies at which information is transmitted (ie, visible light, or a specific sound frequency).

Neural filters
Properties of individual neurons that limit their response to appropriate stimulus types.
Networks of neurons that respond as a whole to specific stimulus types (feature detectors).

Forward masking (selective attention)
- Pollack (1988)
- Sobel & Tank (1994)
Both of these studies were carried out on the Omega Neuron (ON1) of crickets, which is an auditory interneuron. Pollack found that the lateral inhibition (see below) exhibited by ON1 could be explained by forward masking, where an intensity-dependent inhibition follows initial excitation of the neuron. The result of this forward masking is that following excitation by a loud sound, any quieter sounds are not detected by ON1. This filters out all but the loudest signals, which are encoded with a very high signal to noise ratio

The Sobel and Tank study showed that the hyperpolarisation that causes the post-exitatory inhibition is regulated by intensity-dependent release of intracellular calcium. A large depolarisation leads to a large release of Ca2+, which opens calcium-gated channels and allows a hypoerpolarising current to inhibit further firing of the cell.

Spike frequency filter
- Lang (1996)
The spike frequency filter hypothesised by Lang is a function of the rapid decay of EPSPs, determined by the cable properties (time constant and space constant) of the membrane. The time it takes each EPSP to decay imposes a lower limit on the frequency of stimulation that will lead to summation and hence an action potential - if the rate of stimulation is too low, the EPSPs decay before the following EPSP arrives and the membrane potential never crosses the critical threshold. High frequency stimulation, on the other hand, will lead to summation and a series of spikes in the auditory interneuron. This kind of filter is often called a high pass filter because it allows only high frequencies to be passed on to the next stage of processing.

Spike synchronisation filter
- Lang (1996)
The spike synchronisation filter seen by Lang in the auditory interneuron AN4 of the locust has also been described in AN4 of grasshoppers (von Helversen, 1972, 1979; Ronacher and Römer 1985). AN4 responds to a sound pulse with an initial inhibition (IPSP) followed by excitation (EPSP). If the sound pulse is long and unbroken, AN4 responds with tonic excitation after the brief, initial inhibition. If the sound pulse in interrupted by gaps as small as 2ms, however, AN4 is inhibited for the duration of the stimulus.

This filtering is also the result of the cable properties of the membrane of AN4. When many receptors fire at once, as is the case at the beginning of a sound pulse, the IPSPs are sufficient to hyperpolarise the cell and inhibit firing. As the sound pulse continues, the synchrony of the individual receptor responses breaks down. Because the initial IPSP is brief, the EPSPs that follow are more likely to sum when the individual receptors are firing at slightly different times from one another. This allows the membrane potential to cross the critical threshold and produces a response in AN4 to a continuous sound pulse. If the pulse is interrupted by gaps the receptors fall into synchronous firing again, which once again leads to a large inhibition of AN4.

Directional Information

The auditory system can determine the direction to a sound source using two kinds of information:

Intensity difference - the difference in sound pressure between one ear and the other.

Temporal difference - the time lag between when sound reaches one ear and when it reaches the other.

An experiment performed by Heiner Römer (1981) found interneurons in the auditory system of the locust which showed lateral inhibition similar to that seen in the cricket ON1. These neurons responded with strong excitation to sounds presented to the ipsilateral ear (same side as the cell body), but with inhibition if the same sound was presented to the contralateral ear (opposite side to cell body). In this way the auditory interneuron amplifies the intensity difference between the ears, making the loud side louder and the quiet sise quieter. In some cases the inhibition followed an initial excitation, but in others the inhibition came first. In these neurons the time difference between the two sides of the auditory pathway is also being exaggerated. These neurons are a nice example of how directional information in a sound stimulus can be detected and amplified within the nervous system.