Dans un contexxte de sélection sexuelle, les signaux émis par les mâles sont censés refléter honnêtement leur condition. Etant donné que la condition est un concept multifactoriel, il a été propposé que les différentes composantes des signaux sexuels reflètent différents aspects de la condition de l'émetteur (Hypothèse des messages multiples). Ma thèse a pour but de tester la condition-dépendance des signaux sexuel et l'hypothèse des messages multiples en manipulant différents aspects de la condition mâle ; et d'étudier les moyens de contourner un système de signalement honnête par des tactiques de reproduction alternatives. Nous avons conduit ces expériences chez Hyla arborea, espèce chez laquelle les mâles émettent des signaux comprenant des composantes accoustiques et visuelles impliquées dans le choix du partenaires sexuel. Nos résultats suggèrent que le taux de chant intra-bout, une propriétédynamique, est un indicateur de la condition actuelle, alors que la fréquence dominante, une prpriété statique du chant, semble indépendante de la condition actuelle mais liée à la condition passée. Les composantes visuelles du signal sont indépendantes des variations du statut mutirtionnel à court terme mais pourraient être des indicateurs de la condition passée. L'adoption d'une tactique satellite dépend plus de la condition passée que de l'actuelle. Plusieurs composantes du signal sont liées à différents aspects de la condition du mâle, ce qui supporte l'hypothèsedes messages multiples. CEpendant, les relations entre les composantes et les aspects de la condtion sont faibles et nous poussent à considérer également la dynamique évolutive des systèmes de signalements. / In a sexual selection context, signals displayed by males are supposed to reflect honestly their condition. Given that condition is a mutlifactorial concet, it has been propposed that the multiple compopnenets often involved in sexual signal reflects differents aspect of emitter's condition (multiple messages hypothesis). My thesis aimed at testing the condition-dependance of sexual signals and the muliple messages hypothesis by manipulating different aspects of male condition ; and at investigating ways to circumvent an honest signalling systems through alternative reproductive tactics. We conduced these experiments on the European treefrog (Hyla arborea), a species in wich males display signals with both accoustic ansd visual components used by females during mate choice. Our results suggest that the within-bout call rate, a dynamic call property, is a reliable indicator of current condition while call dominant frequency, a static call property, seems to be independant from current conditions but is linked to past condition. Visual signal components are independant from variations in nutritional status on short term but are linked to male condition in the field and may be indicators of past condition. The adoption of satellite tactic depends more on past condition than on current condition.. As several signal components are related to the emitter'scondition and seem to be independant upon distinct aspects of this condition, our results mainly support the multiple messages hypothesis. However, the weak correlations betxeen signal components and condition aspects found in this study highlight the necessity to consider also the evolutionary dynamics of signalliings systems.

The quadratic class of time-frequency distributions (TFDs) forms a set of tools which allow to effectively extract important information from a nonstationary signal. To determine which TFD best represents the given signal, it is a common practice to visually compare different TFDs' time-frequency plots, and select as best the TFD with the most appealing plot. This visual comparison is not only subjective, but also difficult and unreliable especially when signal components are closely-spaced in the time-frequency plane. To objectively compare TFDs, a quantitative performance measure should be used. Several measures of concentration/complexity have been proposed in the literature. However, those measures by being derived with certain theoretical assumptions about TFDs are generally not suitable for the TFD selection problem encountered in practical applications. The non-existence of practically-valuable measures for TFDs' resolution comparison, and hence the non-existence of methodologies for the signal optimal TFD selection, has significantly limited the use of time-frequency tools in practice. In this thesis, by extending and complementing the concept of spectral resolution to the case of nonstationary signals, and by redefining the set of TFDs' properties desirable for practical applications, we define an objective measure to quantify the quality of TFDs. This local measure of TFDs' resolution performance combines all important signal time-varying parameters, along with TFDs' characteristics that influence their resolution. Methodologies for automatically selecting a TFD which best suits a given signal, including real-life signals, are also developed. The optimisation of the resolution performances of TFDs, by modifying their kernel filter parameters to enhance the TFDs' resolution capabilities, is an important prerequisite in satisfying any additional application-specific requirements by the TFDs. The resolution performance measure and the accompanying TFDs' comparison criteria allow to improve procedures for designing high-resolution quadratic TFDs for practical time-frequency analysis. The separable kernel TFDs, designed in this way, are shown to best resolve closely-spaced components for various classes of synthetic and real-life signals that we have analysed.