Musical training and semantic integration in sentence processing: Tales of the unexpected.

Abstract

Transfer effects have been noted in a number of areas surrounding the processing of auditory music and language. There is evidence that musical training enhances the ability of listeners to attend to speech in noise (Alain, Zendel, Hutkaa, & Bidelman, 2014; Strait & Kraus, 2011), to segment speech and differentiate speech sounds and phonemes (Dege & Schwarzer, 2011; Francois, Chobert, Besson, & Schon, 2013; Kuhnis, Elmer, Meyer, & Janke, 2013; Schon & Francois, 2011; Strait, O'Connell, Parbery-Clark, & Kraus, 2014), and to detect acoustic and pitch modulations in language (Jantzen, Howe, & Jantzen, 2014; Kuhnis, Elmer, Meyer, & Janke, 2013; Meyer et al., 2011; Moreno et al., 2009; Schon, Magne, & Besson, 2004). These auditory processing transfer effects from music to language have been accounted for by several factors. First, music and language share a number of acoustic parameters (frequency, duration, intensity, and timbre) (Besson, Chobert, & Marie, 2011), which may simply mean that musical training enhances general auditory processing which in turn enhances the processing of speech sounds (Kuhnis et al., 2013). Second, musical and linguistic sounds are processed by overlapping areas of the brain (Brown, Martinez, & Parsons, 2006; Ducreux, Marsot-Dupuch, Lasjaunias, Oppenheim, & Fredy, 2003; Kraus & Chandrasekaran, 2010; Patel, 2011). Third, musical training places higher demands on this shared processing system in terms of precision to achieve the correct tuning (pitch), timbre (sound quality), dynamic (volume) and, in some cases, sound integration with other instruments, which leads to greater finetuning of sound processing (Kraus & Chandrasekaran, 2010; Kuhnis et al., 2013) through repetition and focused attention (Patel, 2011). These music-to-language transfer models are concerned with the basic processing of the auditory components which make up music and language (Besson et al., 2011; Kraus & Chandrasekaran, 2010; Kuhnis et al., 2013). But can effects of musical training be observed on higher levels of linguistic processing, for example relating to the semantic integration of the meaning of words within a sentence?The similar integration processes required by the shared combinatorial nature of music and language, namely the integration of small components into wider units through the application of rules and probabilities, suggest that the answer could be yes (Besson et al., 2011). A number of studies have indeed demonstrated similarities between the structural processing of music and language, for example showing that harmonic incongruities in music and semantic or syntactic incongruities in language elicit similar neurophysiological responses (Besson et al., 2011; Featherstone, Morrison, Waterman, & MacGregor, 2013; Koelsch, Gunter, Friederici, & Schroger, 2000; Koelsch, 2011; Patel, Gibson, Ratner, Besson, & Holcomb, 1998; Patel, 2008; Steinbeis, Koelsch, & Sloboda, 2006). Furthermore, studies have demonstrated shared resources in the processing of musical and linguistic incongruities (Slevc, Rosenberg, & Patel, 2009; Steinbeis & Koelsch, 2008), shared neural areas involved in the processing of complex music and language (Brown et al., 2006; Ducreux et al., 2003; Koelsch et al., 2002; Levitin & Menon, 2003, 2005; Mirz et al., 1999; Schmithorst, 2005; Vuust, Roepstorff, Wallentin, Mouridsen, & Ostergaard, 2006), and adverse transfer effects of specific language impairment on the neurophysiological patterns associated with the harmonic integration of chords within an ongoing musical sequence (Jentschke, Koelsch, Sallat, & Friederici, 2008). These observations are similar to those said to underpin transfer effects from musical training to auditory aspects of language processing (Besson et al., 2011; Kraus & Chandrasekaran, 2010; Patel, 2011), and make the existence of transfer effects between music and language at a higher processing level plausible. …