, 2007; Mizuno and Sugishita, 2007; Caria et al., 2011; Brattico et al., 2011). Previous work has implicated a distributed network of cortical and subcortical (in particular, limbic) areas in mediating the emotional response to music, suggesting that music processing unites cognitive representational and evaluative mechanisms with the more ‘primitive’ neural mechanisms of reward and biological drives (Blood and Zatorre, 2001; Salimpoor et al.,

2011; Omar et al., 2011). From this perspective, music might therefore be regarded as a comprehensive and biologically relevant model stimulus for assessing human frontal lobe functions. More specifically, recognition of emotion in music engages prefrontal and anterior temporal learn more components of the brain network previously implicated in ToM

processing (Blood et al., 1999; Rankin et al., 2006; Mizuno and Sugishita, 2007; Zahn et al., 2007, 2009; Brattico et al., 2011; Eslinger et al., 2011) and damage involving this network has been linked specifically to deficits of music emotion recognition as well as ToM in bvFTD (Omar et al., 2011; Hsieh et al., 2012; Poletti et al., 2012). Most previous studies of music emotion processing in the normal brain and in disease states have assessed the processing of elementary or canonical emotions (e.g., ‘happiness’, ‘sadness’, ‘anger’) or basic affective dimensions such as consonance – dissonance in music Vorinostat supplier (e.g., Gosselin et al., 2006; Koelsch et al., 2006; Mitterschiffthaler Ureohydrolase et al., 2007; Omar et al., 2010, 2011; Caria et al., 2011; Brattico et al., 2011). There is a sense in which all emotional attributions to music involve some degree of mentalising, since musical emotions must be inferred rather than existing explicitly in the stimuli as do animate emotions in facial and vocal expressions. However, behavioural and neuroimaging findings in

autism and other disorders of social conduct suggest that music has complex interactions with mentalising (Bhatara et al., 2009; Heaton and Allen, 2009; Caria et al., 2011). In particular, it has been demonstrated directly that normal listeners are able to make mentalising judgements about composer agency from musical pieces, and such judgements have functional magnetic resonance imaging (MRI) correlates in the same medial prefrontal and anterior temporal network mediating other kinds of ToM attributions (Steinbeis and Koelsch, 2009). Music is an abstract stimulus yet is widely accessible and highly effective in conveying certain kinds of emotional signals: whereas actual social interactions are often highly complex with many potentially relevant variables, music might allow such interactions to be presented in a reduced, surrogate form that isolates elements critical for mentalising (Warren, 2008). In particular, music may code multi-component or ambiguous feeling states as abstract representations.