Poster C22, Friday, August 17, 10:30 am – 12:15 pm, Room 2000AB

Processing linguistic complexity in Japanese scrambled sentences: an fMRI study

Hyeonjeong Jeong1, Kaoru Koyanagi2, Fuyuki Mine2, Yoko Mukoyama3, Hiroshi Ishinabe4, Haining Cui1, Kiyo Okamoto1, Ryuta Kawashima1, Motoaki Sugiura11;1Tohoku University, Sendai, Japan, 2Sophia University, Tokyo, Japan, 3Musashino University, Tokyo, Japan, 4Higashiosaka Junior College, Osaka, Japan

Japanese is one of the languages that grammatically allows for flexible word order between the subject (S) and object (O). Processing scrambled sentences (e.g., OSV) is considered to enhance higher syntactic computation such as movement, and higher demand for linking syntactic and semantic information than processing their canonical counterparts (e.g., SOV). Previous neuroimaging studies have reported greater activation in the left inferior frontal gyrus and posterior superior temporal gyrus to process scrambled sentences (e.g., Ben-Shachar et al., 2004; Kim et al., 2009). However, the role of each area has not been clarified, being the focus of this research. The current fMRI study attempted to reveal which brain areas are sensitive to syntactic movement and complexity of argument structure for Japanese auditory sentence processing. We hypothesized that processing scrambled sentences with complex argument structure (e.g., a ditransitive verb, which assigns two objects) induce more demanding linguistic computation than processing scrambled simple transitive sentences that assign one object. Participants in this study were 20 healthy, right-handed Japanese native speakers (mean age: 21.95±1.78, 10 females). They were asked to perform a semantic-plausibility judgment task with auditory-presented Japanese sentences during fMRI scanning. We manipulated verb types (transitive: T vs. ditransitive: D) and word order types (canonical: C vs. scramble: S). As a result, 168 simple Japanese sentences were created with 56 semantically implausible sentences as a filler condition. All canonical sentences (S-O-V for T and S-IO-DO-V for D) were created as scrambled sentences (O-S-V for T, IO-S-DO-V or DO-S-IO-V for D). After dividing two sets of stimuli, we counterbalanced canonical and scrambled sentences among the participants. We modeled four regressors of TC, TS, DC, and DS conditions with only correct trials of plausible sentences in each condition. We tested the main effect of the word order type (i.e., scrambling ([TS+DS]–[TC+DC]) and the interaction effect between verb type and word order type as linguistic complexity ([DS–DC]–[TS–TC]). Statistical analyses were performed with SPM12 using a random effects model (corrected to p<0.05 by cluster size). Two major findings emerged. First, analyses of the main effect of scrambling ([TS+DS]–[TC+DC]) revealed significantly greater activation in the left inferior frontal gyrus and supplementary motor area. Consistent with findings of the previous studies (Kim et al., 2009), ours suggest that the left inferior frontal gyrus plays a pivotal role in processing syntactic movement (e.g., scrambling effect) in both transitive and ditransitive sentence conditions. Second, as for the interaction effect (i.e., linguistic complexity) between verb type and word order type, significantly larger differential activation in the bilateral posterior part of superior temporal gyri was found in the [DS–DC] than in the [TS–TC]. This interaction result provides further evidence that syntactic complexity induced by scrambling of more arguments of sentences (i.e., scrambled ditransitive sentences) requires additional involvement of the posterior part of the language area due to the high cognitive demand to link syntactic and semantic information.

Topic Area: Grammar: Syntax

Back