Listeners can easily say if a singer sounds “right” or if a music performance is “correct”. Recent studies highlighted the relevance of pitch interval deviations along melodies (i.e., compression or enlargement of pitch intervals) when evaluating pitch accuracy of singing performances. Also, it has been shown that listeners are highly sensitive to variations in time (i.e., compression or enlargement of inter-onset intervals). However, as is true for several types of judgments (e.g., beauty or obscenity), the definition of “correctness” in music lacks precision. In the first part of this presentation, I will present a series of experiments designed to examine the boundary between in- and out-of-tune melodies as well as between on- and off-beat performances. Besides highlighting listeners’ sensitivity to pitch and time deviations, these studies provide evidence about the implicit development of the normative notion of “correctness” as a category, its modulation by musical training as well as by expectations. In the second part, I will focus on the process involved in correctness judgments and its specificity/similitudes across music dimensions. Finally, recent findings and on-going experiments will be discussed with the objective to better understand what drives listeners appreciation of artistic performances.