FOR RELEASE: Friday, March 24, 2000, 8:36am

Formation of temporal-feature maps in the barn owl's auditory system.

Richard Kempter (Phone 1-415-476-6897, Fax 1-415-476-6897, kempter@phy.ucsf.edu)
Keck Center for Integrative Neuroscience, UCSF,
San Francisco, CA 94143-0732, USA

Christian Leibold (Phone +49-89-289-12193, Fax +49-89-289-12296, cleibold@ph.tum.de)
Physik Department, TU Muenchen,
D-85747 Garching bei Muenchen, Germany

J. Leo van Hemmen (Phone +49-89-289-12362, Fax +49-89-289-12296, lvh@ph.tum.de)
Physik Department, TU Muenchen,
D-85747 Garching bei Muenchen, Germany

Hermann Wagner (Phone +49-241-804835, Fax +49-241-8888133, wagner@bio2.rwth-aachen.de)
Institut für Biologie II, RWTH Aachen,
D-52074 Aachen, Germany

Popular Version of Paper [Y3.002]
Friday morning, March 24, 8:36am
APS March 2000 Meeting, Minneapolis

Sound localization is important to the survival of many animal species, in particular, to those that hunt in the dark. As early as 1948, Jeffress [1] proposed a scheme that combines axonal delay lines from both ears and neuronal coincidence detectors so as to convert interaural time differences (ITDs) in the submillisecond range into a place code. His scheme has had a profound impact on understanding sound localization. Neurons tuned to ITD and maps resembling the circuit envisioned by Jeffress have been observed in many animals [2]. The mechanism, however, that underlies map formation during ontogenetic development has remained unresolved. Here we offer a solution to the problem of how a map of ITDs is set up in the nucleus laminaris of the barn owl, as a typical example [2,3]. We show that an array of neurons is able to represent ITDs in an orderly manner, viz., a map, if homosynaptic spike-based Hebbian learning [4] is combined with a presynaptic propagation of synaptic modifications [5]. The latter may be orders of magnitude weaker than the former. We argue that the algorithm is a key mechanism to the formation of temporal-feature maps, especially on a sub-millisecond time scale.

Imagine a mouse moving through the grass in front of a tree and in this way producing noise. The owl sitting on a branch of the tree and listening, its ears receive the broad band spectrum produced by the mouse. Vertical sound localization is achieved by measuring the intensity difference between the two ears, which are slightly different, both in position and in feather screening. We have concentrated on horizontal sound localization. It is due to a difference in arrival time between the two ears (the interaural time difference or ITD). Sampling frequencies in the range of two to eight kilohertz (2-8 kHz), the barn owl reaches a precision of two degrees, which is equivalent to detecting an ITD of a few microseconds - amazingly good and ten to hundred times better than the membrane time constants of the neurons handling the data, and a few hunded times better than the range of transmission delays from the ear to the barn owl' nucleus laminaris. How, then, does the owl's brain reach that high a precision?

A brain is made up of many neurons. The essence of a neuron can be described as follows. It consists of three parts: (i) an input part, the dendritic tree, (ii) a central processing unit, that emits a pulse with an amplitude of 0.1 Volt, a so-called spike, when its voltage exceeds a threshold, and (iii) an output part, the axon. The axon is a transmission line for the spikes. At its terminations one finds synapses. These pass spikes to the dendritic tree of another neuron. Synapses play a key role in what follows since their strength can be modified through `learning', a kind of programming that is performed by the system itself.

The first stage of horizontal sound localization is to be accomplished in the so-called nucleus laminaris. Here signals from both ears meet. A laminar neuron represents a certain direction in space and, thus, a certain interaural time difference. Its key problem is getting the data from the left and right ear in unison so that they arouse the neuron to fire vigorously. In so doing the neuron would generate spikes but, to this end, it needs enough instantaneous input. The input arrives via the axons coming from other neurons, here from those doing hearing (in the cochlea). A spike traveling along an axon needs a finite amount of time (a delay) to reach a synapse at the axon's termination. The delay varies from axon to axon. Synapses can `learn', however. That is, their efficacy can wax or wane depending on the timing of the arriving spikes in relation to the firing of the receiving neuron - a key element of our theory, as we will see shortly.

Three weeks after hatching, a barn owl's head has reached its final size but sound localization does not function yet. This is not too surprising in view of a huge scatter of the axonal delays. The scatter blocks a well-tuned periodic arrival of the incoming spikes. To see why, let us consider a 5 kHz signal. This is no restriction since the ears sort out the input according to frequencies. The period is 200 microseconds but, because of the axonal delays, the scatter in the transmission times from the ear to a laminar neuron of a youngster is five times as big. The solution to the timing problem is the observation that not genetic coding, which seems implausible in view of many thousands of axons, but a simple training of the synapses at the axon terminations leads to the required fine-tuning.

The training singles out synapses and, hence, axons with the right timing; those that differ by a multiple of the period are also fine. The training (done by hearing) is a kind of selection that is based on the arrival times of incoming spikes as compared to the firing times of the postsynaptic neuron. There is a subtle cooperative process where hundreds of synapses are `steering' a postsynaptic neuron and, in so doing, suppress synapses with the wrong timing but strengthen those that fire in unison with the postsynaptic neuron.

So far, the ITD tuning of single laminar neurons can be understood. A coordinated development of ITD tuning, however, requires an interaction mechanism between them. We propose that the change of a synaptic efficacy through pairing pre- and postsynaptic spikes is not restricted to the stimulated synapse but also give rise to a factor that propagates along the presynaptic axon and affects the properties of the axon (e.g., conduction velocity or safety) and synapses at neighboring neurons. Thus, neurons collectively single out axons with the right timing. The strength of the interaction can be orders of magnitudes weaker than learning at the level of a single neuron. Other types of interaction mechanisms are not feasible [6]. We suggest that ``non-specific axonal learning'' is important for the formation of computational maps whenever time plays a key role.

In view of such a successful system performance, one could ponder about applying similar techniques to industrial applications where an extremely good timing in a sensible surroundings is to be realized.

[1] LA Jeffress, J Comp Physiol Psychol, Vol 41, 35 (1948).
[2] CE Carr, Annu Rev Neurosci, Vol 16, 223 (1993).
[3] M Konishi, Scientific American Vol 268/4, 66 (1993)
[4] W Gerstner, R Kempter, JL van Hemmen, and H Wagner, Nature, Vol 383, 76 (1996)
[5] RM Fitzsimonds and M-m Poo, Physiol Rev, Vol 78, 143 (1998)
[6] R Kempter, C Leibold, JL van Hemmen, H Wagner, Formation of temporal-feature maps by presynaptic propagation of synaptic learning (2000) manuscript in preparation


Richard Kempter's Homepage

[Lay Language Papers]   [Virtual Pressroom]   [Meetings Homepage]   [APS Homepage]