Sound localization through experience

Our hearing is capable of localizing sounds with great accuracy – we don‘t need to look to work out whether a car is coming from the left or the right. Most animals also have good directional hearing, so that they can locate dangers, prey or mating calls. To localize a sound source, the brain compares the auditory information of the two ears. To this end, it uses two different mechanisms. Firstly, it detects differences in volume between the ears – a sound coming from the side will be blocked by the head, and will therefore be much quieter on the far side of it. Secondly, the brain processes the temporal delay between the sound reaching one ear and then the other. In this task, the brain achieves a temporal resolution in the range of a few microseconds. Benedikt Grothe of the Ludwig Maximilians University in Munich and his colleagues have investigated the neuronal mechanisms that underlie the high performance of the mammalian brain, and discovered how the acoustic experience of newborn gerbils leads to the acquisition of the ability to localize sound.

In 1948, Lloyd Jeffress presented a model according to which specific neurons in the brain are tuned to sounds from a certain direction through defi ned neuronal circuits. This model is based exclusively on excitatory neuronal signals. In 2002, Grothe‘s group found that in mammals, inhibitory neuronal connections, which suppress the activity of downstream neurons, also play a major role in localizing an acoustic source. Jeffress‘s model therefore did not apply for these animals.

Acoustic signals are processed in several stages in successive brain regions. On the lowest level of this hierarchy are the neurons that detect the temporal difference between the ears, located in the medial superior olive (MSO). Grothe and his colleagues have now investigated neurons in the next level in the dorsal nucleus of the lateral lemniscus (DNLL), and found that their response to simple acoustic signals of defi ned frequencies is exactly the same as that of neurons in the MSO. However, cells in the DNLL presumably acquire more specialized functions through neuronal interaction with other brain regions, such as suppressing echo signals or tracking moving sound sources.

Since neurons in the DNLL are far more easily accessible experimentally than those in the MSO, the scientists were for the first time able to collect enough data for comparative studies of development of directional hearing, thus enabling them to draw conclusions about the primary circuits of the MSO. The researchers were able to show that neurons processing signals from both ears are not, or only poorly, tuned to a specific direction in newborn gerbils. The ability to localize sound only develops through hearing experience – a process that is associated with structural changes in the inhibitory input signals to the MSO neurons. Sound localization only develops if the gerbils are exposed to localized sound sources within a certain time frame. If the researchers subjected the gerbils to constant unlocalized noise, directional hearing did not develop normally.