Background. Many studies provide evidence of relationship between Music theory and the role of expressive performance in the communication of some structural features of a piece of music. Although an increasing number of studies relates music theory and music perception, only few of them take into account the way in which performance influences this relationship. An experiment by Serafine, Glassman & Overbeeke (1989) tested the cognitive reality of hierarchical structure. In this study subjects had to listen and match a musical excerpt with a rendered reduction either true or false, in terms of Schenkerian theory. However, performance of the excerpts did not received particular attention in the experimental design. Aims. The purpose of this study is to verify the influence of music performance in the listener´s hierarchical representation of a musical piece, in a similar experimental condition such as the one used by Serafine et al. Method. One of the stimuli reported by Serafine was selected. Several commercial recordings of the piece were analysed according to the timing parameter. The interonset interval (IOI) was calculated and the relationships between the IOI of the different versions were compared. According to the differences found, new false rendered structures were composed. In the present study subjects had to compare and to judge the goodness of fit between a) the melody model and b) the true rendered reduction and false reductions. Results. Differences in the IOI profiles, which emphasise in a different way specific points of the fragment, were found. It was predicted that the differences between versions would influence the listener’s response. Results of the listening test are currently being processed. Discussion. The analysis of the IOI profiles revealed that the objective differences found between versions might be related to some structural features of the musical piece. Consequences for music analysis, music performance and for music cognition are discussed.