People are well aware of the fact that background sounds disrupt cognitive performance. However, there is one exception: music. Many people think that music they like has positive effects on their cognitive performance. In a series of experiments, we assessed the effect of classical music on verbal working memory. Music consistently disrupted verbal working memory. However, with ongoing exposure to the music, distraction decreased as participants were able to adapt to the disruptive effects of the music. This decrease in disruption was only observed if the music unfolded in the correct order which preserved the coherent structure of the music piece. Retrospective metacognitive judgements indicated that people who liked the music thought their cognitive performance was less disrupted than people who did not like the music. However, cognitive performance was disrupted irrespective of whether the music was liked or not. Even though people may hold the metacognitive belief that music they like will help them to improve their cognitive performance, objective performance measures indicate that music disrupts verbal working-memory processes irrespective of whether they like it or not.