The neural encoding of productive phonological alternation in speech production: Evidence from Mandarin Tone 3 sandhi

Encoding category-level and context-specific phonological information at different stages: An EEG study of Mandarin third-tone sandhi word production

“…Another recent EEG study (J. Zhang et al, 2022) also found that T3 sandhi words elicited a larger positive amplitude than non-sandhi words in the left anterior regions around 336-520 ms after participants were cued to produce the target words, and this effect was not modulated by lexical frequency, which is aligned with the computation account. Other studies 1 The tonal values are based on Chao's (1948) five-point-scale tonal notation system, with 5 standing for the highest pitch value and 1 for the lowest pitch value.…”

“…Similar to Nixon et al (2015), we employed a picture naming task in this study, which could cover the entire production process from conceptualization to articulation (Indefrey and Levelt, 2004) and is more naturalistic than the implicit priming task or the syllable or morpheme combination tasks. Moreover, as Nixon et al (2015), we used overt immediate naming instead of covert or delayed naming (e.g., C. Zhang et al, 2015;J. Zhang et al, 2022), because previous research has suggested that covert or delayed production may involve different processes from overt production (e.g., Ganushchak et al, 2011;Laganaro and Perret, 2011) and previous fMRI research found that the effect of the T3 sandhi could only be observed in the overt production rather than covert production (e.g., Chang & Kuo, 2020).…”

“…Zhang et al, 2015; J. Zhang and Lai, 2010;J. Zhang et al, 2022; hereafter, we will label T3 as the underlying form and T2 as the surface form for convenience).…”

“…Moreover, this account also suggests that T3 sandhi will be only applied to real words and the application of T3 sandhi will be sensitive to word frequency, as the word frequency would affect the strength of the representation of the surface sandhi tone in memory and thus the subsequent retrieval process in production (see discussion in J. Zhang et al, 2022).…”

“…However, the tasks used in previous productivity and neurolinguistic studies typically require participants to combine isolated syllables into sequences, which may encourage participants to focus more on the underlying form and resort more to the computation mechanism (see discussion in Chien et al, 2020;J. Zhang et al, 2022).…”