Musical semantics: Dimensions, processes, and neural correlates 105 litude is taken to refect the increased amount of harmonic integration, reminiscent of the N400 refecting semantic integration of words. That is, at the same position within a chord sequence the N5 is modulated by the degree of t with regard to the previous harmonic context, analogous to the N400 (elicited at the same position within a sentence) which is modulated by the degree of t with regard to the previ -ous semantic context (see also Figure 1). Therefore, the N5 was proposed to be re-lated to the processing of musical meaning, or semantics, although the type of mu-sical meaning had remained unclear (Koelsch et al., 2000).Notably, the N5 can be elicited even when individuals do not pay attention to the music (e. g., while listeners read a book, or a play a video game, for a review see Koelsch, 2009). That is, the neural mechanisms underlying the generation of the N5 operate partly automatically. Thus, similar to the processing of extra-musical mean-ing refected in the N400 priming effects, intra-musical meaning is processed even when individuals do not pay attention to musical information (an exception is the synchronous processing of words and chords under the instruction to ignore the music and to pay attention to the syntax and semantics of the words; Koelsch et al., 2005.N5 and N400. The following will present evidence for the hypothesis that the earlier brain response to irregular chords (the ERAN) is related to syntactic pro-cessing, and the later response (the N5) to the processing of meaning information. With regard to the ERAN refecting syntactic processes, two ERP studies (Koelsch et al., 2005; Steinbeis & Koelsch, 2008b) showed an interaction between the ERAN and the left anterior negativity (LAN; the LAN is an ERP component elicited by morpho-syntactic violations in language). In these experiments, chord sequences and sentences were presented together synchronously, that is, the onset of each chord coincided in time with the onset of a word of a sentence (Figure 3). The chord sequences ended either on a regular tonic, or on a music-syntactically irregular “Neapolitan” chord (as in Figure 2b). The last word of each sentence was (a) either syntactically correct, with a semantic high-cloze probability (a semantic high-cloze probability means that the last word of the sentence was semantically highly ex-pected, for example the word “beer” in the sentence “He drinks the cold beer”), or (b) syntactically correct, but with a semantic low-cloze probability (which means that the last word of the sentence was semantically less expected, for example the word “beer” in the sentence “He sees the cold beer”), or (c) it was morpho-syntacti -cally incorrect (due to a gender disagreement), with semantic high-cloze probability (e. g. “Er trinkt den kuhlen Bier” / “He drinks themasc coldmasc beerneuter”).Results showed three main effects: (1) Music-syntactic irregularities elicited the ERAN (consistent with the results of previous experiments investigating ERP re-sponses to chords only, e. g., Koelsch et al., 2000). (2) Language-syntactic irregular-ities elicited the LAN, consistent with studies investigating ERP responses to words only (Gunter et al., 2000, e. g.). (3) Semantically less expected words elicited an N400 (compared to the semantically highly expected words), consistent with previous