Music listening is highly pleasurable and important part of most people's lives. Because music has no obvious importance for survival, the ubiquity of music remains puzzling and the brain processes underlying this attraction to music are not well understood. Like other (such as food, sex, and money), pleasurable music activates structures in the dopaminergic reward system, but how music manages to tap into the brain's reward system is less clear. Here we propose a novel framework for understanding musical pleasure, suggesting that music conforms to the recent concept of pleasure cycles with phases of wanting/expectation, liking, and learning. We argue that expectation is fundamental to musical pleasure, and that music can be experienced as pleasurable both when it fulfills and violates expectations. Dopaminergic neurons in the midbrain represent expectations and violations of expecta- tions (prediction errors) in response to rewards and salience signals. We argue that the human brain treats music as an alert/incentive salience signal, and suggest that the activity of dopamine neurons represents aspects of the phases of musical expectation and musical learning, but not directly the phase of music liking. Finally, we propose a computational model for understanding musical anticipation and pleasure operationalized through the recent theory of predictive coding.Keywords: music, pleasure, reward, dopamine, anticipationSupplemental materials: http://dx.doi.org/10.1037/a0031126.suppSome people prefer Mozart whereas others prefer Nirvana or Bob Dylan, but common to most of us is that we find great pleasure in listening to music. Indeed, music is consistently rated to be among the top 10 things people find pleasurable in life (Dube & Lebel, 2003). Consequently, researchers have suggested that music listening is rewarding in itself (Huron, 2001, 2003; Vuust & Kringelbach, 2009; Wallin, Merker, & Brown, 2000; Zald & Zatorre, 2011). This might explain why music is so widely used across situations in our everyday life and why music has sustained such prominence in human cultural development (Fitch, 2005; Rentfrow & Gosling, 2003).Similar to the processing of other rewards, the brain structures that mediate musical perception and pleasure are thought to be anatomically and functionally separated (Peretz, 2010). This sep- aration of perception and pleasure has permeated existing music research (Peretz, 2010), and consequently most studies concentrate entirely on one or the other system. This may partly be owing to the advantages and disadvantages associated with different brain imaging methods. Functional MRI (fMRI) and positron emission tomography (PET) have high spatial resolution, allowing the study of emotion processing by midbrain structures. However, the high spatial resolution comes at the cost of the temporal resolution, which is essential for studying online musical anticipation. A high temporal resolution measuring changes on the basis of millisec- onds is an integrated part of electroencephalography (EEG) and magneto-encephalography (MEG). Consequently many EEG and MEG studies describe anticipatory processes in the auditory cor- tices (Brattico et al., 2009; Brattico, Tervaniemi, Naatanen, & Peretz, 2006; Fujioka, Trainor, Ross, Kakigi, & Pantev, 2004; Garza Villarreal, Brattico, Leino, Ostergaard, & Vuust, 2011; Janata, 1995; Koelsch, 2009; Koelsch, Jentschke, Sammler, & Mietchen, 2007; Leino, Brattico, Tervaniemi, & Vuust, 2007; Loui, Grent-t-Jong, Torpey, & Woldorff, 2005; Maess, Koelsch, Gunter, & Friederici, 2001; Patel, Gibson, Ratner, Besson, & Holcomb, 1998; Schiavetto, Cortese, & Alain, 1999; Toiviainen & Krumhansl, 2003; Trainor, McDonald, & Alain, 2002), but few relate this knowledge to the pleasure of music.In this article, we propose a novel framework for understanding musical pleasure. …