As a crucial component of the human language system, morphology implicates the formation of words and their inter-connections within a language. With respect to word formation, three strategies were generally employed across languages, that is, inflection, derivation, and compounding. Morphological typology classified world languages in light of these word formation methods (e.g., Brown, 2010). For instance, Chinese is a typical isolating language which relies heavily on compounding (more than 70%) for word formation instead of inflections. As such, Chinese is also called a morpho-syllabic language, where one character/syllable is linked to one specific morpheme in the most cases (DeFrancis, 1989). By contrast, Latin constitutes an example of inflectional language using extensively inflections. In particular, modern English is regarded as a weakly inflectional language with limited word form changes marking number, tense, among others, although it also belongs to the Indo-European language family as Latin. Despite the linguistic typology, morphology and word structure information generally implicates an important processing stage and component in both language comprehension and production processes (Levelt, 1993; Schiller, 2020). Extensive studies shed light upon the notion that the competence in morphological processing could impact complex word comprehension, early literacy development, and reading in a second language (e.g., Ke, Miller, Zhang, & Koda, 2020; Kehayia et al., 1999; Marks et al., 2021; Wang, Cheng, & Chen, 2006). So far, however, the neurobiological basis of morphological processing in the human mind and brain across languages and populations is still poorly understood.
Drawing on electrophysiological and neuro-imaging techniques, extant studies arguably elucidated that morphological processing constitutes an integral representation in the human mental lexicon, which is dissociable from semantics (Bölte, Jansma, Zilverstand, & Zwitserlood, 2009; Zou, Packard, Xia, Liu, & Shu, 2019). Bölte et al. (2009) tracked the brain potentials to German derived adjectives with altered morphological manipulations by using event-related potentials (ERPs) and reported a left anterior negativity (LAN) effect peaking around 450 ~ 500 ms after word onset. LAN has been recognized as the indicator of sensitivity to (morpho)syntactic errors (Morris & Holcomb, 2005; Rossi, Gugler, Friederici, & Hahne, 2006) in existing studies. The authors therefore interpreted the observed LAN as an index of structural difficulty resolving and morphological parsing. Likewise, Gao et al. (under review) identified N400 and LAN effects associated with semantic and morphological constraints, respectively, when Chinese native speakers were reading real compound words, morphologically legal, and illegal nonwords. As such, morphological processing might implicate a late, controlled, and top-down process. In contrast, earlier ERP components, such as P/N250, were also found to be associated with an automatic and form-based morphological decomposition in both English and Chinese word reading (K. K. Chung, Tong, Liu, McBride-Chang, & Meng, 2010; Gao, Hua, He, & Yuan, 2021; Morris & Stockall, 2012). Yet, this early and rapid component is present only in masked morphological priming paradigm with a short atimulus-onset asynchrony (SOA range: 50 ~ 57 ms in the abovementioned studies).
Meanwhile, results from magnetoencephalography (MEG) and functional magnetic resonance imaging (fMRI) suggested an important role of the left frontal and temporal cortices in morphological processing across languages (Bakker, MacGregor, Pulvermüller, & Shtyrov, 2013; Bick, Goelman, & Frost, 2008; Forgács et al., 2012; Fruchter, Stockall, & Marantz, 2013; Lehtonen, Vorobyev, Hugdahl, Tuokkola, & Laine, 2006; Liu et al., 2013; Whiting, Marslen-Wilson, & Shtyrov, 2013; Zhao, Wu, Tsang, Sui, & Zhu, 2021). Within this network, the left frontal gyrus (LFG) tends to be a language-general region for morphology, sometimes accompanied with the basal ganglia and the cerebellum (Leminen, Smolka, Dunabeitia, & Pliatsikas, 2019; Zou, Packard, Xia, Liu, & Shu, 2015). A line of Chinese studies also suggested that the left temporal regions might be crucial in Chinese morphological tasks, which might correspond to the morpho-syllabic salience of Chinese language as compared to alphabetical languages. For instance, a recent fMRI study (Ip et al., 2019) asked Chinese adults to complete morphological and phonological judgment tasks, respectively. Their results revealed that morphological tasks elicited robust activations in the superior temporal gyrus (STG) and middle temporal gyrus (MTG), relative to phonological tasks. In addition, another MEG study (Hsu, Pylkkänen, & Lee, 2019) identified the altered activations in the left posterior temporal cortex, which was modulated by the morphological complexity in reading Chinese two-syllabic words. Note that as the aforementioned studies involved various morphological structures (inflections, derivations, and compounds), their findings on language -general and/or specific neural basis for morphological processing should be interpreted with caution.
So far, however, there has been little discussion about the neural correlates of morphological processing in bilingual brains. As bilingualism becomes increasingly pervasive in the globalization context, insights into bilingual brain would be informative for biliteracy development and pedagogy. It is thus crucial to investigate whether or not bilinguals would employ the same morphological processing strategies for first (L1) and second languages (L2). Even though a number of studies compared the morphological representation and processing patterns of bilingual individuals’ L2 with that of native speakers speaking the same language (e.g., W. Chung, Park, & Kim, 2019; Lehtonen et al., 2009; Marks et al., 2021; Prehn, Taud, Reifegerste, Clahsen, & Flöel, 2018), the relationship between L1 and L2 within bilingual brain has not been well established. This inadequacy in L1/L2 comparisons within bilingual morphological processing could be largely attributed to the confounding effect from typological difference between L1 and L2. Methods have been used to somewhat mitigate this confound in contrasting L1 and L2 among bilingual research, which include recruiting highly proficient bilinguals and using shared language structures between L1 and L2. For instance, Lehtonen et al. (2009) observed language-specific brain patterns for L1 Finnish and L2 Swedish when high-proficient Finnish-Swedish early bilinguals completed a lexical decision task with inflected and mono-morphemic nouns during fMRI scans. Specifically, L1 morphological processing was linked to left IFG and posterior temporal area (PTA), which were however not observed in L2 Swedish. This pattern thus suggests distinct processing strategies for bilingual morphology. In contrast, an increasing body of literature tended to hold that neural mechanisms underlying L2 morpho-syntactic processing are transferred from L1 system when L1 and L2 share the same structures (see the review by Tolentino & Tokowicz, 2011). This notion is primarily in line with the unified competition model (MacWhinney & Kroll, 2005), where the processing of L1-L2-overlapped language structures would employ common cognitive resources and strategies from L1. For example, bilingual morpho-syntactic processing engages similar fronto-temporal network for L1 and L2 (e.g., Vingerhoets et al., 2003; Wattendorf et al., 2001). In addition to shared neural resources, there were also differences between L1 and L2 within bilinguals, generally resulting from L2 proficiency, age of acquisition (AoA), and linguistic variations. With respects to morphological processing, the extent to which bilingual L2 resembles or differs from their L1 brain patterns is inclusive.
Therefore, the current study aims at investigating the shared and distinct neural correlates of L1 and L2 morphological processing of shared structures among Chinese-English bilinguals. A recent study (Ip, Hsu, Arredondo, Tardif, & Kovelman, 2017) contrasted the neural corelates of morphological processing among Chinese-English bilingual children with English monolinguals (6 ~ 13 years) by using an auditory morphological structure task. Interestingly, compared to English monolinguals, bilingual children manifested enhanced brain activation in the MTG for English task, which is closely associated with lexico-semantic processing and second language literacy. Meanwhile, L2 English morphological processing involved greater IFG activation relative to L1 Chinese within bilingual individuals. The results therefore suggested both language-general and language-specific neural resources for L1 and L2 morphological engagement. In particular, the authors interpreted the alternations in IFG between L1 and L2 as a language-specific indicator, which reveals bilingual children’s sensitivity to differing morphological constraints of Chinese and English. However, they employed distinct morphological structures (i.e., English derivation and Chinese compound), which might contaminate the bilingual transfer effect as they claimed. Moreover, morphological awareness elicited from auditory tasks might not fully capture the morphological salience, which relies largely on meaning-to-print association. Therefore, the objective of the current study is to further examine the shared and distinct neural corelates of L1 and L2 morphological processing in bilingual brain by using a visual morphological priming paradigm and a shared morphology. Both electroencephalogram (EEG) and functional near infrared spectroscopy (fNIRS) would be recorded simultaneously to capture the temporal-spatial changes in the brain when adult Chinese-English bilinguals were performing morphological tasks. By using a derivational morphology, which exists in both Chinese and English, we hypothesized the observed temporal-spatial characteristics would denote the processing strategies of L1 and L2, instead of linguistic difference.