0.7 Music, biological evolution, and the brain  (Page 10/30)

Neuroimaging of healthy individuals has challenged a domain-specific view of tonality processing, however. This challenge commenced with a study that directly compared brainwave (“event-related potential,” or ERP) activity associated with syntactic processing of language and tonal-harmonic processing of music and found a surprising degree of overlap (Patel et al., 1998). Subsequently, neuroimaging studies using a variety of techniques (e.g., MEG, fMRI) have suggested overlap in brain areas involved in linguistic syntactic processing and musical tonal-harmonic processing (e.g., Maess et al., 2001; Koelsch et al., 2002; Tillmann et al., 2003; Patel, in press(b)). The apparent paradox between data from neurological patients (which support a domain-specific view of tonality) and from neuroimaging (which support a non domain-specific view) led to the “shared syntactic integration resource hypothesis” or SSIRH (Patel, 2003). The SSIRH posits that language and music rely on domain-specific structural knowledge stored in long-term memory (e.g., knowledge of words and their syntactic features or chords and their harmonic relations), but that integration of words or musical tones into hierarchical structures during auditory processing relies on shared, limited neural resources (see Patel, in press (a), for further details and Patel, 2008, Ch. 5, for a full treatment). The SSIRH posits that cases of neurological dissociation result from damage to domain-specific representations, while the similar brain responses seen in neuroimaging studies of healthy individuals reflect shared processes of structural integration operating on these domain-specific representations. Crucially, the SSIRH makes testable predictions, including the prediction that simultaneous structural integration demands in language and music should lead to processing interference. To date, these predictions have been supported by both behavioral and neural studies (Koelsch et al., 2005; Steinbeis and Koelsch, 2008; Fedorenko et al., 2009; Slevc et al., 2009).

Recently, further evidence for overlap between linguistic syntactic processing and musical tonality processing has emerged from clinical studies, including neuroimaging research on specific language impairment (Jentschke et al., 2008), intracranial EEG studies of epileptic patients (Sammler, 2009), and behavioral studies of agrammatic Broca’s aphasia (Patel et al., 2008). The studies of music perception in aphasia have focused on patients with left hemisphere brain damage and “agrammatic comprehension,” i.e., difficulty understanding the meanings of sentences based on their grammatical structure, rather than difficulty understanding the meanings of individual words. For example, such patients, if told the sentence “The girl on the chair was greeted by the man,” would understand that the sentence referred to a girl, a chair, a man, and an act of greeting, but would be unsure of who did what to whom. See Patel et al., 2008, for further discussion. Given this growing evidence for links between tonality processing and linguistic syntactic processing, it is worth stepping back and asking why such connections should exist. After all, instrumental music and linguistic sentences serve different communicative ends and are built from distinct raw materials (e.g., musical tones vs. syllables). Furthermore, the hierarchical structures that organize tones vs. words have been argued to be quite different (Jackendoff, 2009, though see Rohrmeier, 2007, for a different view). Why then would the processing of structural relations in music and language engage similar brain mechanisms?