The current study adopts the ERP technique to investigate 1) how people process sentences terminated by an unexpected but semantically plausible word, 2) how people process sentences with the terminal word switched into another language (i.e., code-mixed sentences), and 3) what the differences are between the two switching directions. Related ERP components (i.e., N400 and a late positivity including P600/LPC and the frontal positivity) are briefly summarized in this session,
followed by an explanation about the reasoning of the experiment design of the current study.
N400, a negative component peaking around 400 ms post-stimulus, reflected difficulties in lexical integration (Kutas & Federmeier, 2000). Both low-cloze words and language-switched words at sentence terminal position elicited an N400 effect relative to the most expected word (Alvarez et al., 2003; Federmeier et al., 2007;
Kutas & Hillyard, 1984; Proverbio et al., 2004; Van Petten & Luka, 2012). The fact was that since low-cloze words and language-switched words were less predictable from the given context, they were more difficult to be integrated into the sentence context. With the contention that N400 reflected difficulties in lexical integration, the amplitude of N400 further implicated the degrees of difficulty in lexical integration.
Existing researches had found that the amplitude of N400 component was graded with the cloze probability of the sentence terminal word. The amplitude of N400 was also found to be different between two directionalities of language switching: it was larger in sentences switched from L1 to L2 (or from a more dominant language to a less dominant one) than from L2 to L1, suggesting that integrating a L2 word into L1 sentence context would be more difficult than the other way round.
After the effect of N400, a positive-going wave was observed in sentences terminated by low-cloze words and by language-switched words. As proposed by Van Petten and Luka (2012), the functional processes of this positivity were believed to be different when it was observed in different regions (i.e., frontal vs. parietal). A frontal positivity was elicited by low-cloze words, and might reflect a conceptual or lexical mismatch between the expected word and the presented word (Federmeier et al., 2007;
Thornhill & Van Petten, 2012). As for the positivity observed from parietal electrodes, it was considered a canonical P600/LPC, and was found in sentences terminated by
language-switched words and by semantically incongruous words (Brenders, 2004;
Moreno et al., 2002). The canonical P600/LPC was interpreted as a reprocessing cost.
Sentences that were complicated in nature could elicit a P600/LPC. Still, it should be noticed that P600/LPC seemed to be insensitive to switching directionality, despite the fact that its amplitude could be related to participants’ language proficiency (Moreno et al., 2002).
For the design of the current study, since researchers generally agreed that high constraint sentences provided a context for people to predict the upcoming word (Federmeier et al., 2007; Thornhill & Van Petten, 2012), we would like to adopt high constraint sentences as the experiment materials, and examine how people process sentences with unexpected completions. Specifically, sentence frames will be completed by a word with an unexpected concept and/or in an unexpected language.
In addition, differences between two switching directionality will be another focus in the current study.
The current study will adopt a 2 x 2 x 2 design (cloze probability x language mode x switching). Sentence stimuli will be constructed and counterbalanced across the following eight conditions: high-cloze Mandarin non-mixed sentences (H-MM), high-cloze Mandarin-Taiwanese code-mixed sentences (H-MT), high-cloze Taiwanese-Taiwanese non-mixed sentences (H-TT), high-cloze Taiwanese-Mandarin code-mixed sentences (H-TM), low-cloze Mandarin non-mixed sentences (L-MM), low-cloze Mandarin-Taiwanese code-mixed sentences (L-MT), low-cloze Taiwanese-Taiwanese non-mixed sentences (L-TT), and low-cloze Taiwanese-Mandarin code-mixed sentences (L-TM). See Table 1 for a summary of the experiment conditions.
Table 1: Experiment Conditions for the Current Study
Conditions Cloze Probability Language of Sentence Frame
Language of the Target Word
H-MM High Mandarin Mandarin
H-MT High Mandarin Taiwanese
H-TT High Taiwanese Taiwanese
H-TM High Taiwanese Mandarin
L-MM Low Mandarin Mandarin
L-MT Low Mandarin Taiwanese
L-TT Low Taiwanese Taiwanese
L-TM Low Taiwanese Mandarin
Based on the experiment design, we will first investigate main effects of the experimental factors (i.e., cloze-probability, language mode, and switching). We will also examine whether there are interactions among these factors. Via comparing sentences terminated by a high-cloze word with those by a low-cloze word, how people process a semantic mismatch between an expected word and a presented word could be revealed. We expect to observe a larger N400 effect and a frontal positivity in low-cloze conditions relative to high-cloze conditions. In addition, via comparing sentences terminated by a word in another language with those in the same language, the effect of language switching could be revealed. It is expected to observe a larger N400 and a LPC/P600 in code-mixed conditions relative to non-mixed conditions.
Since existing researches did not dissociate cloze-probability-related N400 from language-switching-related N400, in the current study, we will pay attention to whether there are differences between them in terms of their scalp distribution. For switching directionality, it will be addressed under the interaction between the factors of language mode and switching. We would like to compare sentences switched from
to uncover the mystery of the asymmetric processing cost in previous behavioral studies. Although LPC/P600 may be observed in both directions, a larger N400 is expected be found in sentences switched from Mandarin to Taiwanese relative to the other way round.