INTRODUÇÃO: Em consequência da modernização tecnológica e industrial, surgiram novos fatores que originaram estudos específicos e direcionados sobre os possíveis efeitos nocivos desta evolução no trabalhador. Dentre estes, está o ruído, agente físico que pode causar danos ao organismo humano com efeitos de curto e médio prazo. O ruído, ao atuar sobre o trabalhador pode alterar processos internos do organismo, com consequências fisiológicas e emocionais. A diminuição gradual da acuidade auditiva decorrente da exposição continuada a níveis elevados de pressão sonora é denominada perda auditiva induzida pelo ruído (PAIR), segundo o Comitê Nacional de Ruído e Conservação Auditiva. A PAIR é caracterizada por perda neurossensorial, irreversível, quase sempre bilateral e simétrica, não ultrapassando 40dB(NA) nas frequências graves e 75dB(NA) nas frequências agudas; manifesta-se, primeiramente, em 6kHz, 4kHz e/ou 3kHz, progredindo lentamente às frequências de 8kHz, 2kHz, 1kHz, 500Hz e 250Hz. (MTE, Portaria 19, 1998). Entre os fatores que influenciam no desencadeamento da PAIR, destacam-se as características físicas do ruído (tipo, espectro e intensidade da pressão sonora), o tempo de exposição e a suscetibilidade individual. A PAIR está entre as doenças do trabalho de maior prevalência, com agravante de tratar-se de uma doença irreversível. Acomete com mais frequência o setor industrial. Este estudo avalia a incidência de perda auditiva por ruído e a existência de uma associação positiva entre estes diagnósticos e a exposição a diferentes níveis de pressão sonora em trabalhadores de duas unidades de uma mineradora e pelotizadora de minério de ferro. MÉTODOS: Estudo de coorte histórica com 1.069 trabalhadores divididos em dois grupos, tendo como referência os níveis de pressão sonora aos quais estavam expostos (menor ou igual a 88 dB e acima de 88 dB). A avaliação da audição foi realizada por análise dos resultados das audiometrias realizadas nos exames admissionais e periódicos constantes dos prontuários médicos. Foram calculadas as incidências de perda auditiva de acordo com os grupos de exposição e o tempo médio de início das alterações auditivas em cada grupo. Foram realizadas análises descritivas das variáveis, comparação de médias (teste U de Mann-Whitney), teste de associação de qui-quadrado e, para estimar fatores de risco, modelos de regressão logística univariada e múltipla. RESULTADOS: Entre os participantes do estudo foram identificados 128 casos (11,97%) de perda auditiva. No grupo composto pelos trabalhadores que ficaram expostos aos níveis mais altos de pressão sonora encontra-se uma incidência de 11,84% em alterações de exames audiométricos, enquanto no grupo menos exposto obteve-se uma incidência de 12,4%. O tempo médio para o desencadeamento de alterações audiométricas foi de 3,07 anos (desvio padrão = 1,21) no grupo mais exposto e de 2,88 anos (desvio padrão = 1,53) no grupo menos exposto. Estar na faixa etária acima de 28 anos e 8 meses (RC = 3,33; IC 95%: 2,01 - 5,49) e trabalhar na Unidade 1 (RC = 1,64; IC 95%: 1,09 - 2,46) se mostraram fatores de risco robustos para perda auditiva induzida por ruído. CONCLUSÃO: A perda auditiva ocorre de maneira precoce, prioritariamente até o terceiro ano de exposição e faixa etária, e unidade de trabalho se mostraram os fatores de risco mais relevantes para a perda auditiva induzida por ruído. O estudo mostra a necessidade de uma integração maior entre as áreas de saúde ocupacional e segurança do trabalho na busca de ações preventivas e corretivas, que possam minimizar o surgimento de novos casos ou o agravamento dos já identificados. Este é um estudo preliminar que servirá de apoio a pesquisas futuras sobre o assunto / INTRODUCTION: As a result of technological and industrial modernization new factors have arisen which led to specific studies have focused on the possible harmful effects of this evolution on the worker. Among these factors is noise, a physical agent which can harm the human body with short and medium term effects. Noise can alter internal body processes, with physiological and emotional consequences. The gradual reduction of hearing acuity resulting from a continued exposure to high levels of noise pressure is called noise-induced hearing loss (NIHL), according to the National Committee of Noise and Hearing Conservation. NIHL is characterized by a permanent sensorioneural impairment, almost always bilateral and symmetrical, not exceeding 40dB (NA) at lower frequencies and 75dB (NA) at higher frequencies. It first manifests at 6kHz, 4kHz and/or 3kHz, slowly progressing to the frequencies of 8kHz, 2kHz, 1kHz, 500Hz and 250Hz. (MTE, Ruling 19, 1998). Among the factors which trigger NIHL are the physical characteristics of the noise (type, spectrum and intensity of sound pressure), duration of exposure, and individual susceptibility. NIHL is among the most common occupational diseases, with the aggravating aspect of being irreversible. It is most often found in the industrial sector. This study will evaluate the incidence of noise-induced hearing loss and the existence of a positive association between these diagnoses and the exposure to different levels of sound pressure in workers at two operational units of a iron ore mining and pelletizing organization. METHODS: historical cohort study with 1069 workers, divided in two groups, using as baseline the sound pressure levels to which they were exposed (less than or equal to 88 dB and above 88 dB). The hearing assessment was carried out by means of an analysis of the results of the audiometry exam upon induction and at regular intervals, as recorded in the medical history of the individual. A calculation was made of the degree of hearing loss according to the groups of exposure, and the average time it took for hearing changes to be observed in each group. RESULTS: Among the participants in the study, 128 (11.97%) cases of hearing loss were identified. In the group formed by workers who had been exposed to the highest levels of sound pressures, we found an incidence of 11.84% of changes in the audiometry exams, while in the group subject to lesser exposure we observed an incidence of 12.4%. The average time for the development of hearing impairment was 3.07 years (standard deviation = 1.21) in the group with greater exposure, and 2.88 years (standard deviation = 1.53) in the group with lesser exposure. Robust risk factors for noise-induced hearing loss included being aged over 28 years and 8 months (RC = 3,33; IC 95%: 2,01 - 5,49) and working at Unit 1 (RC = 1,64 ; IC 95%: 1,09 - 2,46). CONCLUSION: Early hearing impairment occurs mainly before the end of the third year of exposure, and age bracket and work location proved to be the most relevant risk factors for noise-induced hearing loss. The study showed the need for a greater integration among the occupational health and safety areas to seek the design and implementation of preventive and corrective actions to minimize the development of new cases or the aggravation of those already identified. This is a preliminary study which will serve as a basis for a more detailed subsequent data analysis

Eine wesentliche Aufgabe auditorischer Systeme besteht in der Erkennung und Klassifikation verhaltensrelevanter Signale. Die akustischen Kommunikationssignale vieler Feldheuschrecken zeichnen sich durch artspezifische Modulationen der Signalamplitude aus, die im Kontext der Partnerwahl zur Erkennung der eigenen Art genutzt werden. Die Kommunikation ist jedoch auch als Basis für sexuelle Selektion von Interesse - einer Abschätzung der Qualität des Senders anhand der akustischen Signale, welche eine Bewertung subtiler Variationen der artspezifischen Musters erfordert. Das Ziel dieser Arbeit bestand darin zu untersuchen, wie amplitudenmodulierte akustische Signale in den Antworten identifizierter auditorischer Interneurone der zweiten und dritten Verarbeitungsstufe repräsentiert werden, insbesondere, wie gut sie anhand dieser Antworten unterscheidbar sind. Dazu wurden (i) sinusförmig amplitudenmodulierte Stimuli genutzt und die Parameter Modulationsfrequenz und Modulationstiefe systematisch variiert, (ii) individuelle Gesänge der gleichen Art, und (iii) im Grundmuster zeitlich reskalierte Gesänge. Lokale Interneurone zeichneten sich aus durch: ein oft sehr hohes zeitliches Auflösungsvermögen, hohe Empfindlichkeit gegenüber Schwankungen der Signalamplitude, sowie gute Unterscheidbarkeit der sinusförmig amplitudenmodulierten Signale und der Gesänge auf der Basis von Spikeantworten. Bei den aufsteigenden Interneuronen nahm die Fähigkeit zur zeitlichen Ankopplung der Spikes an die Amplitudenmodulationen der Stimuli ab, was sich auch in deren reduzierter Unterscheidbarkeit äußerte. Ursächlich hierfür war einerseits die Zunahme der Antwortvariabilität (Jitter der Spikezeitpunkte), aber auch verstärkt auftretende Filtereigenschaften. Auf dieser dritten Verarbeitungsebene kommt es zu einer stärkeren Spezialisierung auf bestimmte zeitliche Aspekte des Stimulus, die als Grundlage einer verhaltensrelevanten Klassifikation von akustischen Signalen interpretiert werden kann. / A central task of auditory systems is the recognition and classification of behaviorally relevant signals. The communication signals of many grasshoppers can be characterized by a species-specific pattern of amplitude modulation, which is mainly used for species recognition in the context of mate finding. Additionally, the communication is also of interest with respect to sexual selection - an evaluation of the signaler''s quality from the signal pattern, which requires the quantification of subtle variations of the common species-specific pattern.The goal of this study was to investigate how amplitude modulated acoustic signals are represented in the responses of identified 2nd and 3rd order auditory interneurons, particularly, how well they can be discriminated on the basis of the responses. For this (i) sinusoidal amplitude modulated stimuli were used and the parameters modulation frequency and modulation depth were systematically varied, (ii) individual songs of the same species and (iii) songs with temporal rescaled basic pattern were presented. Local interneurons can be characterized by: mostly high temporal resolution capacities, high sensitivity to fluctuations of the signal amplitude as well as a good distinguishability of sinusoidal amplitude modulated stimuli and songs on the basis of the spike trains. In ascending interneurons the synchronization to the amplitude modulations decreased, which also appeared in a reduced discrimination performance. This is caused by an increase of response variability (jitter of spike timing) but also by distinctive filter properties of the respective neurons. Neurons on this third processing level exhibit a greater specialization to particular temporal aspects of the stimulus. This can be interpret as a basis of a behaviorally relevant classification of acoustic signals.

INTRODUÇÃO: Em consequência da modernização tecnológica e industrial, surgiram novos fatores que originaram estudos específicos e direcionados sobre os possíveis efeitos nocivos desta evolução no trabalhador. Dentre estes, está o ruído, agente físico que pode causar danos ao organismo humano com efeitos de curto e médio prazo. O ruído, ao atuar sobre o trabalhador pode alterar processos internos do organismo, com consequências fisiológicas e emocionais. A diminuição gradual da acuidade auditiva decorrente da exposição continuada a níveis elevados de pressão sonora é denominada perda auditiva induzida pelo ruído (PAIR), segundo o Comitê Nacional de Ruído e Conservação Auditiva. A PAIR é caracterizada por perda neurossensorial, irreversível, quase sempre bilateral e simétrica, não ultrapassando 40dB(NA) nas frequências graves e 75dB(NA) nas frequências agudas; manifesta-se, primeiramente, em 6kHz, 4kHz e/ou 3kHz, progredindo lentamente às frequências de 8kHz, 2kHz, 1kHz, 500Hz e 250Hz. (MTE, Portaria 19, 1998). Entre os fatores que influenciam no desencadeamento da PAIR, destacam-se as características físicas do ruído (tipo, espectro e intensidade da pressão sonora), o tempo de exposição e a suscetibilidade individual. A PAIR está entre as doenças do trabalho de maior prevalência, com agravante de tratar-se de uma doença irreversível. Acomete com mais frequência o setor industrial. Este estudo avalia a incidência de perda auditiva por ruído e a existência de uma associação positiva entre estes diagnósticos e a exposição a diferentes níveis de pressão sonora em trabalhadores de duas unidades de uma mineradora e pelotizadora de minério de ferro. MÉTODOS: Estudo de coorte histórica com 1.069 trabalhadores divididos em dois grupos, tendo como referência os níveis de pressão sonora aos quais estavam expostos (menor ou igual a 88 dB e acima de 88 dB). A avaliação da audição foi realizada por análise dos resultados das audiometrias realizadas nos exames admissionais e periódicos constantes dos prontuários médicos. Foram calculadas as incidências de perda auditiva de acordo com os grupos de exposição e o tempo médio de início das alterações auditivas em cada grupo. Foram realizadas análises descritivas das variáveis, comparação de médias (teste U de Mann-Whitney), teste de associação de qui-quadrado e, para estimar fatores de risco, modelos de regressão logística univariada e múltipla. RESULTADOS: Entre os participantes do estudo foram identificados 128 casos (11,97%) de perda auditiva. No grupo composto pelos trabalhadores que ficaram expostos aos níveis mais altos de pressão sonora encontra-se uma incidência de 11,84% em alterações de exames audiométricos, enquanto no grupo menos exposto obteve-se uma incidência de 12,4%. O tempo médio para o desencadeamento de alterações audiométricas foi de 3,07 anos (desvio padrão = 1,21) no grupo mais exposto e de 2,88 anos (desvio padrão = 1,53) no grupo menos exposto. Estar na faixa etária acima de 28 anos e 8 meses (RC = 3,33; IC 95%: 2,01 - 5,49) e trabalhar na Unidade 1 (RC = 1,64; IC 95%: 1,09 - 2,46) se mostraram fatores de risco robustos para perda auditiva induzida por ruído. CONCLUSÃO: A perda auditiva ocorre de maneira precoce, prioritariamente até o terceiro ano de exposição e faixa etária, e unidade de trabalho se mostraram os fatores de risco mais relevantes para a perda auditiva induzida por ruído. O estudo mostra a necessidade de uma integração maior entre as áreas de saúde ocupacional e segurança do trabalho na busca de ações preventivas e corretivas, que possam minimizar o surgimento de novos casos ou o agravamento dos já identificados. Este é um estudo preliminar que servirá de apoio a pesquisas futuras sobre o assunto / INTRODUCTION: As a result of technological and industrial modernization new factors have arisen which led to specific studies have focused on the possible harmful effects of this evolution on the worker. Among these factors is noise, a physical agent which can harm the human body with short and medium term effects. Noise can alter internal body processes, with physiological and emotional consequences. The gradual reduction of hearing acuity resulting from a continued exposure to high levels of noise pressure is called noise-induced hearing loss (NIHL), according to the National Committee of Noise and Hearing Conservation. NIHL is characterized by a permanent sensorioneural impairment, almost always bilateral and symmetrical, not exceeding 40dB (NA) at lower frequencies and 75dB (NA) at higher frequencies. It first manifests at 6kHz, 4kHz and/or 3kHz, slowly progressing to the frequencies of 8kHz, 2kHz, 1kHz, 500Hz and 250Hz. (MTE, Ruling 19, 1998). Among the factors which trigger NIHL are the physical characteristics of the noise (type, spectrum and intensity of sound pressure), duration of exposure, and individual susceptibility. NIHL is among the most common occupational diseases, with the aggravating aspect of being irreversible. It is most often found in the industrial sector. This study will evaluate the incidence of noise-induced hearing loss and the existence of a positive association between these diagnoses and the exposure to different levels of sound pressure in workers at two operational units of a iron ore mining and pelletizing organization. METHODS: historical cohort study with 1069 workers, divided in two groups, using as baseline the sound pressure levels to which they were exposed (less than or equal to 88 dB and above 88 dB). The hearing assessment was carried out by means of an analysis of the results of the audiometry exam upon induction and at regular intervals, as recorded in the medical history of the individual. A calculation was made of the degree of hearing loss according to the groups of exposure, and the average time it took for hearing changes to be observed in each group. RESULTS: Among the participants in the study, 128 (11.97%) cases of hearing loss were identified. In the group formed by workers who had been exposed to the highest levels of sound pressures, we found an incidence of 11.84% of changes in the audiometry exams, while in the group subject to lesser exposure we observed an incidence of 12.4%. The average time for the development of hearing impairment was 3.07 years (standard deviation = 1.21) in the group with greater exposure, and 2.88 years (standard deviation = 1.53) in the group with lesser exposure. Robust risk factors for noise-induced hearing loss included being aged over 28 years and 8 months (RC = 3,33; IC 95%: 2,01 - 5,49) and working at Unit 1 (RC = 1,64 ; IC 95%: 1,09 - 2,46). CONCLUSION: Early hearing impairment occurs mainly before the end of the third year of exposure, and age bracket and work location proved to be the most relevant risk factors for noise-induced hearing loss. The study showed the need for a greater integration among the occupational health and safety areas to seek the design and implementation of preventive and corrective actions to minimize the development of new cases or the aggravation of those already identified. This is a preliminary study which will serve as a basis for a more detailed subsequent data analysis

In this thesis I attempt to explain mechanisms of neuronal coding in the auditory system as a form of adaptation to statistics of natural stereo sounds. To this end I analyse recordings of real-world auditory environments and construct novel statistical models of these data. I further compare regularities present in natural stimuli with known, experimentally observed neuronal mechanisms of spatial hearing. In a more general perspective, I use binaural auditory system as a starting point to consider the notion of function implemented by sensory neurons. In particular I argue for two, closely-related tenets: 1. The function of sensory neurons can not be fully elucidated without understanding statistics of natural stimuli they process. 2. Function of sensory representations is determined by redundancies present in the natural sensory environment. I present the evidence in support of the first tenet by describing and analysing marginal statistics of natural binaural sound. I compare observed, empirical distributions with knowledge from reductionist experiments. Such comparison allows to argue that the complexity of the spatial hearing task in the natural environment is much higher than analytic, physics-based predictions. I discuss the possibility that early brain stem circuits such as LSO and MSO do not \"compute sound localization\" as is often being claimed in the experimental literature. I propose that instead they perform a signal transformation, which constitutes the first step of a complex inference process. To support the second tenet I develop a hierarchical statistical model, which learns a joint sparse representation of amplitude and phase information from natural stereo sounds. I demonstrate that learned higher order features reproduce properties of auditory cortical neurons, when probed with spatial sounds. Reproduced aspects were hypothesized to be a manifestation of a fine-tuned computation specific to the sound-localization task. Here it is demonstrated that they rather reflect redundancies present in the natural stimulus. Taken together, results presented in this thesis suggest that efficient coding is a strategy useful for discovering structures (redundancies) in the input data. Their meaning has to be determined by the organism via environmental feedback.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Chez plusieurs espèces, les neurones auditifs sensibles à la durée de la stimulation sont présents au niveau des collicules inférieurs. Toutefois, le décours temporel de leur développement fonctionnel est inconnu. Étant donné que le collicule supérieur est l’un des principaux relais sous-cortical impliqué dans l’intégration des stimuli audio-visuels, nous voulons déterminer si le collicule supérieur du rat contient de tels neurones et s’ils sont sensibles et sélectifs à différentes durées de stimulation auditive. De plus, l'originalité de cette étude est de déterminer les étapes de leur maturation fonctionnelle. Des enregistrements neuronaux unitaires et extra-cellulaires sont effectués dans le collicule supérieur de rats juvéniles (P15-P18, P21-P24, P27-P30) et adultes anesthésiés. La sensibilité à la durée est déterminée lors de la présentation de bruits gaussiens (2-10 dB SPL au-dessus du seuil) de durées variables (3-100 ms). Seulement un faible pourcentage des neurones du collicule supérieur est de type passe-bande (3-9% des neurones parmi les ratons et 20% chez les rats adultes). Une différence significative de la distribution entre les différents types de neurones auditifs sensibles à la durée est présente au cours du développement: les neurones de type passe-haut (63-75%) sont présents en majorité chez les groupes juvéniles alors que 43% des neurones sont de type insensible à la durée de la stimulation auditive chez les rats adultes. Ces résultats montrent qu’une population importante de neurones auditifs du collicule supérieur du rat est sensible à la durée des signaux sonores et qu’un développement fonctionnel important survient au cours du premier mois postnatal. / In many species, duration-sensitive neurons which could possibly analyse species-specific vocalizations are encountered in the inferior colliculus. However, their postnatal development is unknown. Since the superior colliculus is one of the most important subcortical relay implicated in the audiovisual integration, we sought to gain insight on whether the SC of the rat contains such neurons and, if so, whether they are sensitive and selective to duration. Furthermore, the originality of this study is to determine their fonctional maturation. Extracellular single-unit recordings were assessed in the deep layers of the superior colliculus of anesthetized rats throughout postnatal days (P15 – P18, P21 – P24, P27 – P30) and at adulthood. Gaussian noises of different durations (3-100 ms) were presented in free-field in pseudo-random order (2 – 10 dB SPL above threshold). Only few neurons are band-pass: about 3 – 9% across the different age groups (P15 – P30) and 20 % in adults. Thus, in terms of categories of duration-sensitive neurons profiles, a significant difference was evident across postnatal development: long-pass (63-75%) neurons were mostly encountered in the different groups of pups relative to all-pass (43%) in adult rats. These results demonstrate that a large population of neurons in the superior colliculus of the rat is sensitive to noise durations and that important development changes occur during postnatal maturation.

Das Neuron ist das zentrale Element in der Informationsverarbeitung im Nervensystem. In dieser Arbeit werden verschiedene Aspekte der Spikegenerierung sowohl theoretisch als auch experimentell untersucht. Phasen-Rotatoren verschiedener Komplexität werden zur Vorhersage von Spikezeitpunkten vorgestellt. Die Kennlinie eines Neurons wird dabei als wichtiger Parameter für diese Modelle verwendet, damit diese leicht auf echte Neurone anwendbar sind. Die Phasenantwortkurve als ein zweiter wichtiger Aspekt der Spikedynamik wird zur Erweiterung der Modelle verwendet. Solange ein Neuron in seinem überschwelligen Bereich gereizt wird, erweisen sich die Phasenrotatoren als gute Beschreibung des Spikeverhaltens. Es wird jedoch gezeigt, daß bei einer Stimulierung mit Strömen, die um die Schwelle des Neurons herum fluktuieren, diese Modelle, genauso wie alle anderen eindimensionalen Modelle einschließlich des Intergrate-and-fire Neurons, versagen. Feuerraten Adaptation kann in vielen Neuronen beobachtet werden. Es wird ein allgemeines phänomenologisches Modell für die Feuerrate adaptierender Neurone aus den Eigenschaften verschiedene Ionenströme, die Adaptation verursachen, hergeleitet. Dieses Modell ist durch die Kennlinien und einer Adaptations-Zeitkonstanten vollständig definiert. Mit Hilfe des Modells können die Eigenschaften der Adaptation als Hochpassfilter quantifiziert werden. Weiterhin wird die Rolle der Adaptation bei der Unterdrückung von Hintergrundrauschen diskutiert. Sowohl die Phasenrotatoren als auch das Adaptationsmodell werden an auditorischen Rezeptorzellen der Wanderheuschrecke und dem AN1, ein primäres auditorisches Interneuron der Grille {Teleogryllus oceanicus}, getestet. In beiden Fällen stimmen die Modelle gut mit den experimentelle Daten überein. Es wird mit Hilfe der Modelle gezeigt, daß Adaptation in den Rezeptorzellen durch Ionenströme des Spikegenerators verursacht wird, während in dem Interneuron der Eingang schon adaptatiert. Zusätzlich wird der Einfluß der Feuerraten-Adaptation auf die Gesangserkennung analysiert. / The single neuron is the basic element of information processing in nervous systems. In this thesis several properties of the dynamics of the generation of spikes are investigated theoretically as well as experimentally. Phase oscillators of different complexity are introduced as models to predict the timing of spikes. The neuron's intensity-response curve is used as a basic parameter in these models to make them easily applicable to real neurons. As a second important aspect of the spiking dynamics, the neuron's phase-resetting curve is used to extend the models. The phase oscillators turn out to be a good approximation of the spiking behavior of a neuron as long as it is stimulated in its super-threshold regime. However, it is shown by comparison with conductance-based models that these models, as well as all other one-dimensional models including the common integrate-and-fire model, fail, if the neuron is stimulated with currents fluctuating around its threshold. Spike-frequency adaptation is a common feature of many neurons. For various ionic currents, as a possible reason for adaptation, a general phenomenological model for the firing rate of adapting neurons is derived from their biophysical properties. This model is defined by the neuron's intensity-response curves and an adaptation time-constant. By means of this model the high-pass properties of spike-frequency adaptation can be quantified. Also the role of adaptation in supression of background noise is discussed. Both the phase oscillators and the adaptation-model are tested on auditory receptor neurons of locusts and the AN1, a primary auditory interneuron of the cricket {Teleogryllus oceanicus}. In both cases the models are in good agreement with the experimental data. By means of the models it is shown that adaptation in the receptor neurons is caused by ionic currents of the spike generator while in the interneuron it is the input which is already adapting. In addition, the influence of spike-frequency adaptation on the recognition of courtship songs is analysed.

Die Intensität verhaltensrelevanter Signale variiert oft über viele Größenordnungen. Gleichzeitig müssen sensorische Systeme in der Lage sein, über den gesamten relevanten Bereich feine Intensitätsunterschiede aufzulösen. Auf neuronaler Ebene ergibt sich bei Nutzung eines Feuerratencodes aus diesen Anforderungen ein grundsätzlicher Konflikt, da neuronale Antwortbereiche beschränkt sind. Eine Lösung, die in vielen Sinnessystemen beschrieben wurde, ist die Verschiebung von Intensität-Kennlinien, so dass der gesamte Antwortbereich des Neurons zur Verfügung steht, um schnelle Abweichungen vom Mittelwert zu kodieren. Diese Arbeit versucht anhand mathematischer Modelle zu beantworten, wie die Verschiebung von Kennlinien in einem neuronalen Netzwerk entstehen könnte. Ausgangspunkt ist eine Rezeptorpopulation mit Intensitätsbereichsaufteilung und einem begrenzten Verschiebungsbereich der Kennlinien von Einzelrezeptoren, die auf ein Output-Neuron konvergieren. Diese Organisation wurde vom auditorischen System der Grille inspiriert. Modelle, die auf einer Kombination aus einer sättigenden Nichtlinearität und Spike-Frequenz-Adaptation basieren, reproduzieren die Verschiebung der Kennlinien entlang der Intensitäts-Achse. Diese Modelle sind in der Intensitätsdiskriminierung dem Rezeptormodell und der Summe von Rezeptorantworten über große Intensitätsbereiche deutlich überlegen. Die Kennlinien dieser Modelle besitzen zudem weitere Eigenschaften, die in ihrer Kombination übereinstimmend in verschiedenen sensorischen Systemen beschrieben wurden: Insbesondere erklären sie eine zusätzliche scheinbare Verschiebung entlang der Antwortachse, unterschiedliche Steigungen der verschobenen Kennlinien, sowie Steigungsänderungen innerhalb einzelner Kennlinien. Die einfachen, abstrakt formulierten Modelle ermöglichen ein tieferes Verständnis adaptiver Mechanismen über das Modellsystem Grille hinaus. / Intensities of behaviourally relevant signals often vary over many orders of magnitude. At the same time, sensory systems need to ensure high sensitivity to minute intensity differences across the full intensity range. These demands conflict on the neuronal level due to the boundedness of neuronal response ranges. To solve this dilemma, intensity response curves in many sensory system were found to shift towards the actual mean intensity so that the full response range can be used to encode fast fluctuations around the slowly varying mean. Using mathematical models, this study approaches the question how shifts of intensity response curves might arise in small neural networks. The starting point is a population of receptors with stacked response thresholds and limited capacity of adaptive shift that converge onto one output neuron. This organization was inspired by the auditory system of the cricket. A combination of a static saturating non-linearity and spike-frequency adaptation reproduced the desired shift of response curves along the intensity axis. With respect to intensity discrimination, these models are superior to the receptor model and the sum of receptor responses over a wide range of absolute intensities. The response curves generated by these model also displayed details of response curve behaviour consistently observed in numerous experimental studies. In particular, they explain an apparent shift along the response axis, different slopes of the shifted response curves, and changes in the slope within individual response curves. The simple, abstract models allow for a deeper understanding of adaptive mechanisms beyond the auditory system of the cricket.

Ziel der vorliegenden Arbeit war es, die Auswirkungen von externen Störquellen und intrinsischer Variabilität auf die Verarbeitung und Erkennung von akustischen Signalen am Modellsystem der Feldheuschrecke Chorthippus biguttulus zu untersuchen. Damit sowohl die Gesangserkennung am sich verhaltenden Tier als auch die dieser Erkennung zugrunde liegende neuronale Verarbeitung untersucht werden konnte, wurde ein Weibchengesang verwendet, dessen zeitliches Muster durch zufällige Amplitudenmodulationen gestört wurde. Durch die Degradation mit verschiedenen Frequenzbändern konnte überprüft werden, ob bestimmte Modulationsfrequenzen die Signalerkennung stärker beeinflussen als andere. Mit zunehmender Störung der Gesangsstruktur kam es in den Verhaltenstests an Männchen zu einer Abnahme der Erkennungsleistung. Die Stärke der tolerierten Degradation war dabei in der Regel nicht unterschiedlich für die getesteten Degradationsbänder. Die Unterschiede in den neuronalen Antworten, welche entweder durch die artifizielle extrinsische Degradation oder durch interne Fehler in der auditorischen Verarbeitung verursacht wurden, konnten durch eine Spiketrain-Metrik quantifiziert werden. Diese Analyse zeigte, dass die Auswirkung der extrinsischen Signaldegradation von den Rezeptoren über die lokalen Interneurone zu den aufsteigenden Interneuronen abnahm, während es zu einem signifikanten Anstieg der intrinsischen Variabilität kam. Die Stärke der Degradation war dabei erneut nicht unterschiedlich für die getesteten Degradationsbänder. Durch die Bestimmung von neurometrischen Schwellen konnten die Grenzen der Signalerkennung der Männchen mit der Rauschtoleranz der einzelnen auditorischen Neurone verglichen werden. Die kritischen Degradationsstufen, die so ermittelt werden konnten, stimmten teilweise erstaunlich gut überein. Somit sind die Grenzen der Signalerkennung durch die Analyse der Antwortkapazitäten der ersten drei Verarbeitungsstufen relativ gut erklärbar. / The aim of this study was to investigate the effects of extrinsic and intrinsic noise sources on signal recognition and processing within the acoustic communication system of the grasshopper Chorthippus biguttulus. To test both - signal recognition of behaving animals and the underlying auditory processing mechanisms - a female song was used, whose temporal pattern was disturbed by random amplitude modulations. Due to the degradation with various modulation bands, it was possible to test if distinct modulation frequencies have more pronounced effects on signal recognition than others. Behavioural tests on males of Chorthippus biguttulus showed that progressive degradation of the song pattern induced a decrease in recognition performance. The strength of degradation tolerated generally was the same for different modulation bands. The differences between neuronal responses, which were either caused by the artificial extrinsic degradation or internal errors during auditory processing, could be quantified by a spiketrain metric. This analysis showed that the effect of extrinsic signal degradation was much more severe for receptors and local interneurons than for ascending interneurons, whereas there was a significant increase of intrinsic variability with higher levels of processing. The strength of the degradation was again not different for different modulation bands. Signal recognition could be compared with the noise tolerance of individual auditory neurons by determining neurometric thresholds. The average critical degradation levels, to some extend, matched the critical degradation level for behaviour. Thus, by means of analysing the response capacities of neurons from the first three levels of auditory processing, the limits of signal detection are relatively well explained.

Neuronale Adaptation dient dazu, eine Sinnesbahn kurzfristig an die aktuelle Umgebung des Tieres anzupassen. Ihr zeitlicher Verlauf lässt sich in der Antwort einzelner Nervenzellen direkt beobachten. Der Adaptation unterliegen eine Vielzahl verschiedener Mechanismen, die über die gesamte Sinnesbahn verteilt sein können. In der vorliegenden Arbeit wurde der Versuch unternommen, diese unterschiedlichen Betrachtungsebenen zusammenzuführen. Dazu wurden mehrere experimentelle und theoretische Studien durchgeführt. In zwei der vorgestellten Studien wurden Kombinationen aus Strominjektionen und akustischen Reizen verwendet, um intrinsische Adaptation von Netzwerkeffekten zu trennen. Dabei ergab sich in einer experimentellen Studie am auditorischen System der Heuschrecke, dass die Adaptationsmechanismen, die in verschiedenen Teilen der Hörbahn rekrutiert werden, sehr stark von Identität und Funktion der jeweils untersuchten Nervenzelle abhängen. Ähnliche Methoden ermöglichten es, im auditorischen System der Grille präsynaptische Hemmung als Substrat für die wichtige mathematische Operation der Division zu identifizieren. Zusätzlich wurden Modellierungen durchgeführt, bei denen die Frage bearbeitet wurde, wo Adaptation in der Hörbahn wirken sollte, bezogen auf zwei verschieden Aufgaben: die Lokalisation eines Signals und die neuronale Abbildung dessen zeitlicher Struktur. Die Ergebnisse dieser Studie deuten darauf hin, dass die Anforderungen für diese beiden Aufgaben sehr unterschiedliche sind. In einer vierten Studie wurde untersucht, ob die Adaptation in einem auditorischen Interneuron der Grille dazu dient, die gesamte sensorische Umgebung gut abzubilden, oder ob durch die Adaptation eine Abtrennung des jeweils lautesten Signals erreicht werden kann. Zusammenfassend lässt sich sagen, dass sowohl die Adaptationsmechanismen, als auch deren genaue Platzierung innerhalb der sensorischen Bahn wesentlich für Sinnesleistungen sind. / Neural adaptation serves to adjust the sensory pathway to the current environment of an animal. While the effect and time course of adaptation can be observed directly within single cells, its underlying cause is a combination of many different mechanisms spread out along the sensory pathway. The present work has the objective to unite these different levels of understanding of the term adaptation. In order to do so, several experimental and theoretical studies were carried out. In two of these studies, a combination of current injection and auditory stimulation was used, in order to disentangle intrinsic adaptation from network effects. In one of the studies, carried out in the auditory system of locusts, it was revealed that the mechanisms behind adaptation that are activated within different parts of the auditory system depend critically on identity and function of the cell under study. Similar methods enabled the identification of presynaptic inhibition as a possible mechanisms behind the important mathematical operation of division in the auditory system of crickets. Additionally, a modeling study pursued the question, where adaption should work in the auditory system from the perspective of two different tasks of sensory processing: identification of a signal and localization of its source. The results obtained from the model suggest conflicting demands for these two tasks and also present a solution of this conflict. In a fourth study, it was asked wether adaptation in the auditory system of crickets serves to guarantee optimal representation of the entire sensory environment or if it helps to separate one most important signal from the background. In summary, not only which mechanisms of adaptation are at work is of crucial importance for sensory processing, but also the exact placement of these along the pathway.