You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

Poster B17, Wednesday, November 8, 3:00 – 4:15 pm, Harborview and Loch Raven Ballrooms

Asymmetric Binarity as a Cognitive Universal: The Rhythm of Syntactic Structure

Danielle Fahey1, Dirk-Bart den Ouden1;1University of South Carolina

Introduction: In all languages and all musical traditions, universal rules group elements hierarchically, with elements organized asymmetrically. In language, Merge theoretically combines elements in phrasal structures, with one element governing the other, recursively, to form sentences [1]. In rhythm, a similar asymmetric hierarchy of beats is proposed in the Generative Theory of Tonal Music [2]. Just like syntactic processing, assigning beats into rhythmic strings, ‘beat induction,’ is automatic and unconscious [3]. Merge and beat induction assign structure not only to grammatical phrases, but also to inherently unstructured sequences. Overlapping perisylvian brain regions supporting both language [4] and rhythmic [5] processing have been implicated in these mechanisms. Because overlapping cortical regions have shown support for similar roles in both domains, this project tested whether structurally comparable linguistic and rhythmic input would be processed similarly in core regions of the language network, inferior frontal and posterior superior temporal cortex. Methods: 19 right-handed adult English-native speakers underwent Magnetic Resonance Imaging (MRI) while auditory stimuli were presented. Neuroimaging data were analyzed using SPM12, with region-of-interest (ROI) analyses of the inferior frontal gyrus (IFG) opercularis, triangularis and orbitalis, and the posterior superior temporal gyrus (pSTG). Linguistic stimuli provided participants with three types of input: (1) grammatical phrases, (2) semi-randomly sequenced words, containing words in an ungrammatical order (no hierarchical phrases), (3) ‘jabberwocky’ stimuli, containing phonotactically-valid English non-words, conjugated by inflectional morphemes. Rhythmic stimuli provided participants with (4) rhythmically-patterned, (5) arrhythmic, and (6) regular beats. Rhythmically-patterned beats provided rhythmic ‘tunes.’ Arrhythmic beats varied in tempo and were characterized by random beat sequences. Regular beats sounded like ticking clocks, without differences between beats in the acoustic signal. These sets of linguistic and rhythmic stimuli should be processed similarly due to their structural congruence: (1) and (4) are structured sequences; (2) and (5) are unstructured sequences that will violate grammaticality rules; and (3) and (6) are not inherently structured, but should be assigned hierarchical structuring. Results: A RM ANOVA showed a 4-way interaction (p<.05) between hemisphere, ROI, domain, and ‘grammaticality’. Follow-up tests in the left IFG operculum revealed higher activation for (2) than for (3) (p=.025), for (3) than for (1) (p=.072), for (4) than (6) (p=.035) and for (4) than (5) (p=.021). Tests in the pSTG also revealed higher activation for (4) than (6), and for (4) than (5) (both, p<.001). Conclusion: These results support prior research suggesting overlap in brain regions responsible for both rhythmic and syntactic processing, specifically in the IFG operculum. However, the function of these regions for linguistic and rhythmic input is likely different, as linguistic and rhythmic ‘grammaticality’ were not processed symmetrically within each region. This analysis suggests that the left IFG operculum and pSTG are domain general in their overall cortical roles, yet domain-specific in the type of processes they support within the domains of language and rhythm processing. References: [1] Chomsky, N. (1995). [2] Lerdahl, F. & Jackendoff, R. (1983). [3] Honing, H. (2012). [4] Friederici, A. D., & Gierhan, S. M. (2013). [5] Danielsen et al., (2014).

Topic Area: Grammar: Syntax

Back to Poster Schedule