Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Bottom-up vs. top-down processing of Turkish vowel harmony in the VWFA
Poster Session E, Friday, October 2, 11:00 am - 1:00 pm, Wangari Maathai
This poster is part of the Sandbox Series.
Stefan Pophristic1, Özlem Öge-Daşdöğen1,2, Alec Marantz1; 1New York University, 2Atlas Üniversitesi
Introduction. The visual word form area (VWFA) is one of the first brain regions responsible for language-specific processing of visually presented words—yet there is still no clear model of what computations are done in this area. fMRI work on the VWFA has largely shown identical responses to both real and pseudo-words (e.g.VicknierEtAl2007), suggesting that this area is responsible for calculating orthographic regularity rather than linguistic processing (e.g.DaheaneEtAl2005). Magnetoencephalographic (MEG) work has further implicated this area in morphological decomposition of whole words (e.g. splitting “farmer” into “farm+er”), indexed through the M170 component (e.g.SolomyakMarantz2010). However, it is unclear whether this decomposition is motivated by 1) top-down linguistic processing or 2) bottom-up orthographic statistics. Question 1. To tease these possibilities apart, we focus on Turkish vowel harmony (VH), an orthographic/phonological generalization whereby vowels within a word agree in phonological features. While VH applies across morpheme boundaries, 25% of the vocabulary (loanwords) naturally violate the generalization. We employ a 2x2 design in a visual lexical decision task (LDT) with simultaneous MEG recording, where participants are presented with morphologically simple vs. complex words that are harmonic or disharmonic. While disharmonic simple words are grammatical, since VH must apply across morpheme boundaries, the disharmonic complex words are ungrammatical. Thus, while the local vowel-to-vowel statistical relationship is maintained across simple vs. complex disharmonic words, the two conditions differ in grammaticality. By comparing the harmonic vs. disharmonic conditions between simple and complex words, we can test whether linguistic grammaticality has additional effects on VWFA activity beyond local orthographic relationships. If the VWFA is sensitive to morphological properties, then illegal disharmony across a morpheme boundary (complex condition) should be processed differently than legal disharmony (simple condition). Otherwise, if the VWFA is insensitive to morphological properties, then disharmony should be treated equally, regardless of morpheme boundaries. Question 2. Since effects of VH in the VWFA have not yet been tested, it is likewise unclear what the expected directionality of the VH violation is. If VH statistics are tracked as orthographic statistics, the harmonic condition should facilitate processing and lead to decreased VWFA amplitudes, as shown with bigram/trigram variables (SolomyakMarantz2009). Otherwise, if the VWFA engages in form-based lexical access, then conditional well-formedness should reflect lexical competition effects (e.g. CarreirasEtAl1997), resulting in increased VWFA activity. Behavioral Pilot. 30 participants completed the study online. A significant harmony-complexity interaction (p=0.044) revealed that VH differences were driven by the simple (p=0.003) but not complex (p=0.949) condition. No VH effects were found for nonce-words (p=0.882). MEG Study. 9/30 participants have successfully completed the LDT and a functional localizer task (WhiteEtAl2023) with simultaneous MEG recording. To establish an fROI, we will conduct spatio-temporal permutation clustering over the 100-250ms time window in the occipital/temporal lobes (GwilliamsEtAl2018). The dependent variable(s) for linear mixed-effects models are dSPM values averaged by trial/participant over the spatial/temporal extent of the group-level localizer cluster. We test: Q1) whether the interaction term between VH and Complexity is significant; Q2) whether VH positively or negatively predicts the dependent variable.
Topic Areas: Morphology, Phonology