Novel approaches for the investigation of sound localization in mammals

The ability to localize sounds in space is important to mammals in
terms of awareness of the environment and social contact with each
other. In many mammals, and particularly in humans, localization of
sound sources in the horizontal plane is achieved by an
extraordinary sensitivity to interaural time differences (ITDs).
Auditory signals from sound sources, which are not centrally
located in front of the listener travel different distances to the
ears and thereby generate ITDs. These ITDs are first processed by
binaural sensitive neurons of the superior olivary complex (SOC) in
the brainstem. Despite decades of research on this topic, the
underlying mechanisms of ITD processing are still an issue of
strong controversy and the processing of concurrent sounds for
example is not well understood. Here I used in vivo extra-cellular
single cell recordings in the dorsal nucleus of the lateral
lemniscus (DNLL) to pursue three novel approaches for the
investigation of ITD processing in gerbils, a well-established
animal model for sound localization. The first study focuses on the
ITD processing of static pure tones in the DNLL. I found that the
low frequency neurons of the DNLL express an ITD sensitivity that
closely resembles the one seen in the SOC. Tracer injections into
the DNLL confirmed the strong direct inputs of the SOC to the DNLL.
These findings support the population of DNLL neurons as a suitable
novel approach to study the general mechanism of ITD processing,
especially given the technical difficulties in recording from
neurons in the SOC. The discharge rate of the ITD-sensitive DNLL
neurons was strongly modulated over the physiological relevant
range of ITDs. However, for the majority of these neurons the
maximal discharge rates were clearly outside this range. These
findings contradict the possible encoding of physiological relevant
ITDs by the maximal discharge of single neurons. In contrast, these
data support the more recent hypothesis that the discharge rate
averaged over a population of ITD-sensitive neurons encodes the
location of low frequency sounds. In the second study, I
investigated the ITD processing of two concurrent sound sources,
extending the classical approach of using only a single sound
source. As concurrent sound sources a pure tone and background
noise were chosen. The data show that concurrent white noise has a
high impact on the response to tones and vice versa. The discharge
rate to tones was mostly suppressed by the noise. The discharge
rate to the noise was suppressed or enhanced by the tone depending
on the ITD of the tone. Investigating the responses to monaural
stimulation and to tone stimulation with concurrent spectrally
filtered noise, I found that the ITD sensitivity of DNLL neurons
strongly depends on the spectral compositions, the ITDs, and the
levels of the concurrent sound sources. Two different mechanisms
that mediate these findings were identified: monaural
across-frequency interactions and temporal interactions at the
level of the coincidence detector. Simulations of simple
coincidence detector models (in cooperation with Christian Leibold)
suggested this interpretation. In the third study of my thesis, the
temporal resolution of binaural motion was analyzed. Particularly,
it was investigated how fast the neuronal system can follow changes
of the ITD. Here, psychophysical experiments in humans and
electrophysiological recordings in the gerbil DNLL were performed
using identical acoustic stimulation. Although the binaural system
has previously been described as sluggish, the binaural response of
ITD-sensitive DNLL neurons was found to follow fast changes of
ITDs. Furthermore, in psychophysical experiments in humans, the
binaural performance was better than expected when using a novel
plausible motion stimulus. These data suggest that the binaural
system can follow changes of the binaural cues much faster than
previously reported and almost as fast as the monaural system,
given a physiological useful stimulus. In summary, the results
presented here establish the ITD-sensitive DNLL neurons as a novel
approach for the investigation of ITD processing. In addition, the
usage of more complex and naturalistic stimuli is a promising and
necessary approach for opening the field for further studies
regarding a better understanding of the hearing process.