Most predator and parasitoid functional response studies have been carried out by using experimental designs where insects are confined to an arena and subsequently exposed to different host densities, which are evaluated individually. In the case of a parasitoid that looks for profitable patches, this design forces it to use the single host density patch available, and therefore the possibility of selection by the parasitoid is not considered at all. A selective functional response, in which the host is distributed in discrete patches at different densities, could be a solution to avoid such a limitation. However, the disadvantage of this design is that it does not meet the independence assumption required to perform a parametric statistical analysis. Nevertheless, the use of nonparametric analyses such as GAM and GAMM models allows the performance of this kind of design, making the relationship between the response and the explanatory variable more flexible, looking for general behavioral patterns. The behavior of the fruit fly parasitoid Diachasmimorpha longicaudata (Ashmead) in a patched condition was assessed in order to demonstrate that nonparametric analyses are useful tools when studying the selective functional response. Results showed that the functional response changed from a “sigmoid curve” to a “bell-shape curve” when the parasitoid had the chance to choose freely among different host densities. The female parasitoid distributed their ovarian load among the eight host densities. The present study suggests that the bell-shape curve displays a general behavior pattern of the parasitoid population.