Theoretical models of sexual selection assume that female mating preferences are fixed and variation found between individuals resulting solely from allelic variation at specific loci coding for sexual preferences. For the last decade, an increasing number of studies have demonstrated that individual phenotypic variation in preferences was common across a wide range of taxa and induced by the environmental context and the females’ condition. Further, developmental stages of life are crucial in the formation of behaviours in general and have proven to be determinant to learn sexual preferences in some species that dispense care for their young. However, very little studies have analysed how the early social and physical environments shape female mate choice in species that lack parental care. In this thesis, I used guppies (Poecilia reticulata), firstly, to investigate the influence of various aspects of the social environment provided by males during two ontogenetic phases. Secondly, I explored whether learned preferences in a foraging context during development could be transferred into a mating context. Considering the early social environment, I explored three distinctive features potentially displayed by males and that females might experience while growing. Females were reared with different values of a sexual trait not genetically preferred in the population (orange colour) and different values of a trait for which they had innate predisposition (total colour area). In both cases, females were exposed to the different treatments for the whole developmental period or for its later phase. My results indicated that females changed their sexual behaviours in response to both type of traits experienced, reversing sometimes their genetic preferences. Moreover, the timing of exposure seemed to be a key factor in the acquisition of preferences as females exposed only to the later part of development with different values of total colour didn’t rely anymore on colour patterns to discriminate among males. In a third body of experiment, I examined whether the overall phenotypic variance exhibited by males during whole development, independently of the values of a specific sexual cue, mediated female’s behaviours. In a context of high variance, female became choosier relatively to those experiencing less variance. As a response, males switched mating tactics and attempted more forced copulations. In its final part, my thesis searched for a link that might have arisen, owing to developmental conditions, between preferences using the same sensory modality in two behavioural contexts. Maturing females were given food that was associated to a certain colour and subsequently tested for both their coloured preference in a foraging and a sexual context. Although no foraging preference for the corresponding colour was detected, females that experienced a yellow stimulus preferred yellower males compared to females with other experiences. Taken together these results suggest that developmental conditions and especially the social environment play a pivotal role in the process of mate choice. Under some circumstances, learned mate preferences override genetically-based preferences highlighting the importance of non-genetic mechanisms. Accordingly, it is urgent to integrate in the study of sexual selection and reproductive isolation this dimension. In guppies, for instance, the effect of early social life might contribute to the maintenance of colour pattern polymorphism found in males.