Processing of auditory signals in the forebrain of the Barn Owl : neural mechanisms of across-frequency integration

Interaural time differences (ITDs) of acoustic signals represent a major cue for sound localization in most animals including humans. Due to the finite propagation velocity of sound and the different path lengths to the two ears, ITDs change with the azimuth of a sound source. Binaural comparison of phase locked spikes is performed in neurons of the auditory brainstem. On this first stage of ITD processing, neurons are sensitive to interaural phase differences (IPDs) in narrow spectral components of the signal. However, IPDs relate in an ambiguous manner to the ITD of the signal. In the barn owl’s external nucleus of the inferior colliculus (ICX), a part of the tectofugal pathway, phase ambiguities are resolved by integrating ITD information across frequency channels. This thesis focusses on ITD processing in the parallel thalamofugal pathway. I report substantial differences in ITD processing in the two pathways that may reflect an impact of behavior on sensory representations. Extracellular recordings were obtained from neurons in the auditory arcopallium (AAr), a high level auditory and sensorimotor integration center in the forebrain of the barn owl (Tyto alba). Responses were characterized with focus on ITD and frequency tuning. ITD curves to pure tones revealed the preferred interaural phase differences at specific frequencies. In neurons with a linear phase-frequency relation, across-frequency integration was described by the neuron’s characteristic delay (CD) and its characteristic phase (CP). The CD is the delay at which the neuron’s relative response strength was consistent across frequencies. The CP indicates the relative response strength at which the CD was exhibited. Besides the classical method of assessing CD and CP from ITD tone curves, I developed a new method allowing for CD and CP estimation on the basis of discrete Fourier transforms of ITD noise functions. Data were collected from 290 AAr units and compared to data sets from 76 ICX units. AAr units were responsive to broad ranges of frequencies. A variety of tuning curve shapes was observed including single peaked curves, curves displaying multiple peaks and curves that were flat over a range of frequencies. In comparison to frequency tuning in ICX units, tuning width in AAr was broader and the distribution of peak values included an important portion of frequencies lower than 3 kHz. In response to ITD noise stimuli, most AAr units featured peak responses at contralateral leading ITDs. Peak widths were broader than in ICX units. ITD tuning curves of AAr units displayed a central peak and side peaks. In a majority of AAr units the main peak was asymmetric displaying a steep slope on the side close to zero ITD and enhanced responsiveness on the side of contralateral leading ITDs. ITD tuning curves were significantly more asymmetric in AAr units than in ICX units. On the populational level, this difference was no longer apparent, when tuning curves were recorded using highpass filtered noise. Hence, low frequencies made an important contribution to the asymmetries of ITD curves in AAr units. Asymmetric tuning curves may be explained by the ITD sensitivity displayed at different spectral components. In ICX units, peak ITDs were consistent throughout the frequency range. This is equivalent to saying that neurons exhibited a CD equal to the best ITD and a CP of zero. In contrast, ITD responses in a majority of AAr neurons displayed an intermediate response level at a consistent ITD value across frequencies. CPs had a mean absolute value of 0.17 cycles, i.e. units displayed slope type integration of ITD information across frequency channels. As linear integration of spectral ITD sensitivity was a good model to explain ITD responses to noise, the observed asymmetric shape of ITD noise curves can be seen as a direct consequence of the prevailing slope type integration in AAr units. Analysis of phase-frequency relations in different frequency ranges, revealed that low frequencies contributed sensitivity to large ITDs. Neurons displaying non zero CPs are typical of the mammalian auditory brainstem, but have not been described in the owl before. While phase locked inhibition and cochlear delays cannot be ruled out as reason for non zero CPs in mammalian systems, they are an unlikely explanation in the owl. I propose that neurons in the auditory forebrain of the owl combine ITD information from lower stages in a systematic way across frequency and across ITD channels to yield non zero CPs. Since the tectofugal pathway seems to subserve sound localization, the thalamofugal pathway is likely to be involved in additional tasks such as recognition of auditory objects, decisions on importance and novelty of sounds as well as control of attention. Therefore, the observed differences in ITD processing in the thalamofugal pathway compared to the tectofugal pathway may represent adaptations to behavioral requirements.

Grantor dc:publisher

Publikationsserver der RWTH Aachen University

Year dc:date

2008

Author dc:creator

• Vonderschen, Katrin

Contributors dc:contributor

• Wagner, Hermann

• info:eu-repo/classification/ddc/570
• Richtungshören
• Vorderhirn
• Schleiereule
• Elektrophysiologie
• Diskrete Fourier-Transformation
• Biowissenschaften, Biologie
• Interaurale Zeitdifferenz
• Sound localization
• interaural time difference
• forebrain
• barn owl
• electrophysiology
• discrete fourier transform

Statement dc:rights

• info:eu-repo/semantics/openAccess

Language dc:language

eng

Identifier

https://publications.rwth-aachen.de/search?p=id:%22RWTH-CONV-112638%22

OAI record oai:identifier

https://publications.rwth-aachen.de/oai2d?verb=GetRecord&metadataPrefix=oai_dc&identifier=oai%3Apublications.rwth-aachen.de%3A50077