Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Does the linear position of a morpheme affect early, form-based decomposition? EEG evidence from prefixation, infixation, and suffixation
Poster Session E, Friday, October 2, 11:00 am - 1:00 pm, Wangari Maathai
This poster is part of the Sandbox Series.
Samantha Wray1, Dave Cayado2, Rémi Lamarque3, Rowan Stewart3, Linnaea Stockall3; 1Dartmouth College, 2Royal Holloway, University of London, 3Queen Mary, University of London
Ample neurolinguistic evidence suggests that morphologically complex words undergo early, form-based decomposition within the first ~200ms of word recognition (indexed in MEG by the M170), localized in the left fusiform gyrus, (the Visual Word Form Area – VWFA), such that words like teacher are segmented into meaningful subparts {teach}+{-er}. Existing theoretical accounts have argued that this decomposition process is modulated by linear position of morphemes. Specifically, Beyersmann and Grainger (2022) argued that only morphemes at the edges of the letter-strings (i.e., edge-alignment) are subject to decomposition. Further, asymmetry in morphological priming found in prefixes and suffixes also suggests that the linear position of affixes modulates decomposition (Kim et al., 2015; Meunier & Segui, 2002). Cayado et al. (2023, 2026) showed behavioural masked priming evidence that prefixes, infixes, and suffixes are all successfully extracted during the decomposition process, contra predictions of models that argue for modulation of decomposition by affix linear position and edge-alignment. However, the failure to find an effect may have been due to behavioural masked priming not being sufficiently sensitive, since it reflects not just the initial decomposition that happens within 200 ms after seeing the word, but also subsequent lexical decision processes reflected in the button press ~800 ms later. The present study addresses this problem by using electroencephalography (EEG), allowing us to focus our investigation on the first 200 ms of seeing a written word. We have two main questions: (1) are prefixed, infixed, and suffixed words decomposed into morphemes? (2) Do edge-alignment and affix linear position modulate this decomposition? Twenty native Tagalog speakers will perform a visual lexical decision task while 64-channel EEG records neural activity. This task has three conditions: (a) Prefixation; words with the ni- prefix (e.g., ni- + lason ‘poison’ = nilason); (b) Infixation; words with the <in> infix (e.g. tawag ‘call’ + <in> = tinawag); and (c) Suffixation; words with the -in suffix (e.g. basag ‘break’ + -in = basagin). Following MEG studies that demonstrate decomposition by correlating VWFA activity at ~170ms to stem:whole word transition probability (see Wray et al. 2022 for Tagalog), we aim to replicate this approach in EEG. If early decomposition is affected by the linear position of an affix, we expect the strength or timing of the correlation to differ between affix types. If, however, early decomposition is not affected by linear position, as shown by Cayado et al. (2023, 2026), the correlation should be comparable across all conditions.
Topic Areas: Morphology, Reading