AbstractThe effect of models differing in releasing quality is tested with respect to acoustical features of the directed song, and is compared to song not induced by releasers (so called “undirected song”). A gradual change of several parameters (for example number of introductory elements, number of motifs per strophe, speed of performance of the first motif) was found, indicating a gliding increase of intensity from song not releaser‐provoked, via song released by models poor in releasing stimuli to song induced by living fe