Neurolinguistic circuitry for two different scripts of language, such as phonological scripts (PhonoS) versus logographic scripts (LogoS) (e.g., English versus Chinese, resp.), recruits segregated neural pathways according to orthographic regularity (OrthoR). The purpose of this study was to identify the effect of VSF for cortical representation according to different OrthoR to represent Hangul versus Hanja as PhonoS versus LogoS, respectively. A total of 24 right-handed, native Korean undergraduate students with the first language of PhonoS and the second language of LogoS were divided into high- or low-competent groups for L2 of LogoS. The implicit word reading task was performed using Hanja and Hangul scripts during functional magnetic resonance imaging (fMRI) acquisition. Fluctuations of fMRI BOLD signal demonstrated that the LogoS was associated with the ventral pathway, whereas PhonoS was associated with the dorsal pathway. By interaction analysis, compared with high-competent group, low-competent group showed significantly greater activation for Hanja than for Hangul reading in the right superior parietal lobule area and the left supplementary motor area, which might be due to neural efficiency such as attention and cognition rather than core neurolinguistic neural demand like OrthoR processing.
Neurolinguistic circuitry for two different scripts of language, namely, phonological script (PhonoS) versus logographic script (LogoS) (e.g., English versus Chinese, resp.), recruits segregated neural pathways according to orthographic regularity (OrthoR) on human brain cortical mapping using functional magnetic resonance imaging (fMRI) [
Given the better understanding of neurolinguistic circuitry of PhonoS versus LogoS as depicted with fMRI study, further work investigated whether learning the second language (L2) is related to assimilation versus accommodation with respect to native first language circuitry (L1) [
Along with different OrthoR, the spatial pattern between scripts affects cortical activation on neurolinguistic fMRI and Korean script system provides an excellent opportunity to identify the effect of VSF given that the pair of Korean Hangul and Hanja has a unique property of very similar spatial pattern. Meta-analysis demonstrated different visual word form area (VWFA) of peak response and amplitude between LogoS of Chinese and PhonoS of the alphabetic script [
Hence, the purpose of this study was to identify the extent and degree of which VSF affects cortical representation for Hangul versus Hanja as PhonoS versus LogoS, respectively, given different OrthoR. The native Koreans with a first language (L1) of PhonoS and a second language (L2) of LogoS were divided into high-competent (High-C) versus low-competent (Low-C) groups for L2 of LogoS. We hypothesized that the relative higher activation in Low-C group, if any, could more likely be due to neural efficiency such as attention and cognition rather than core neurolinguistic neural demand. The whole brain cortical topology of VSF effect could provide a platform on language processing such as in bilingualism using assimilation versus accommodation between L1 and L2 languages.
A total of 24 right-handed, native Korean undergraduate students at a local university participated in the fMRI study. The participants were divided into two groups, those familiar with Hanja (High-C group, mean
The fMRI experiment had a block design consisting of Hanja and Hangul trials that were randomly intermixed across the run [
LogoS and PhonoS used for the word reading experiment. LogoS words were selected based on level 5 proficiency guided by the KAPHE, which is equivalent to a middle school educational level in Korea. PhonoS words were chosen based on those most frequently used in daily life. Within each activation block, a word was displayed for 1 s. A total of 30 words were presented in each block.
A 3 T whole-body Siemens scanner (Siemens, Magnetom Verio, Erlangen, Germany) was used for image acquisition with an interleaved T2
The first five images were discarded from analysis to eliminate the nonequilibrium effects of magnetization. Functional volumes were realigned, coregistered to anatomical images, and normalized to the Montreal Neurological Institute (MNI) template using a transformation matrix acquired from the T1 anatomical image normalization based on the SPM T1 template. Finally, spatial smoothing was performed with an 8 mm full-width at the half-maximum Gaussian kernel. The resulting time-series was high-pass filtered with a cut-off time window of 128 s to remove low-frequency drift in the blood-oxygen level dependent (BOLD) signal. Then, the design matrix was temporally convolved with a canonical hemodynamic response function for a better fit.
A voxel-based general linear model was used at the single subject level to estimate the parameters associated with the conditions of interest (Hanja reading or Hangul reading versus fixed cross as a baseline) along with six motion parameters as covariates of no interest. With the parameters acquired from the first level analysis, a group level analysis was performed using a
The main effect analysis of script showed increased cortical activation in both High-C and Low-C (
BOLD activation for LogoS versus PhonoS reading (
Contrast | Region | BA | Cluster size |
|
|
|
|
---|---|---|---|---|---|---|---|
LogoS > PhonoS | Fusiform_L | 37 | 518 | −32 | −58 | −18 | 5.68 |
Occipital_Mid_L | 19 | 589 | −32 | −88 | 16 | 5.17 | |
Parietal_Sup_L | 7 | 934 | −28 | −60 | 48 | 5.35 | |
Precentral_L | 9 | 789 | −44 | 2 | 32 | 5.37 | |
Insula_L | 13 | 618 | −32 | 18 | 12 | 4.62 | |
Precentral_L | 6 | 113 | −28 | −6 | 54 | 3.80 | |
Supp_Motor_Area_L | 6 | 334 | −8 | 4 | 62 | 4.15 | |
Fusiform_R | 37 | 514 | 34 | −52 | −16 | 5.13 | |
Temporal_Inf_R | 37 | 227 | 48 | −66 | −12 | 5.13 | |
Parietal_Sup_R | 7 | 890 | 26 | −66 | 50 | 4.62 | |
Insula_R | 13 | 576 | 34 | 22 | 6 | 4.55 | |
Precentral_R | 9 | 255 | 44 | 4 | 32 | 4.95 | |
Cllm_6_L | 160 | −32 | −54 | −24 | 5.68 | ||
Cllm _6_R | 682 | 32 | −62 | −26 | 5.13 | ||
|
|||||||
PhonoS > LogoS | Occipital_Mid_L | 39 | 1746 | −44 | −76 | 32 | 4.74 |
Temporal_Mid_L | 21, 38 | 797 | −50 | 4 | −34 | 4.34 | |
Frontal_Sup_L | 8, 9, 10 | 2048 | −28 | 34 | 36 | 5.16 | |
Frontal_Inf_Orb_L | 11 | 536 | −50 | 36 | −16 | 4.74 | |
Cingulum_Mid_L | 24 | 167 | 0 | −18 | 42 | 4.11 | |
Temporal_Mid_R | 21 | 554 | 56 | −64 | 20 | 4.34 | |
Temporal_Inf_R | 20, 21 | 99 | 62 | −26 | −18 | 3.60 | |
Paracentral_Lobule_R | 5 | 1140 | 2 | −44 | 66 | 4.21 | |
Frontal_Sup_R | 8 | 282 | 24 | 30 | 52 | 3.89 | |
Precuneus_R | 31 | 53 | 12 | −42 | 44 | 3.59 | |
Cllm _Crus2_L | 113 | −44 | −78 | −48 | 4.23 | ||
Cllm _Crus2_R | 563 | 30 | −92 | −38 | 5.21 | ||
Cllm _9_R | 59 | 8 | −58 | −64 | 3.70 |
Main effect analysis of script (LogoS versus PhonoS reading) for all subjects (
Different script systems recruit a different cortical involvement during reading written word depending on their OrthoR [
The analysis of the interaction effect showed substantially different cortical activation depending on Hanja competency at the threshold of approximately uncorrected
Interaction effect analysis of PhonoS versus LogoS and LogoS competency.
Contrast | Region | BA | Cluster size |
|
|
|
|
---|---|---|---|---|---|---|---|
LogoS-PhonoS in High-C > LogoS-PhonoS in Low-C | Caudate_R | 82 | 12 | 16 | 10 | 2.99 | |
Caudate_L | -6 | 10 | 8 | 2.40 | |||
|
|||||||
LogoS-PhonoS in Low-C > LogoS-PhonoS High-C |
Parietal_Sup_R | 7 | 302 | 20 | −64 | 48 | 2.62 |
Supp_Motor_Area_L | 8 | 788 | 14 | 22 | 40 | 2.52 | |
Frontal_Inf_Tri_R | 46 | 64 | 52 | 24 | 28 | 2.03 | |
Frontal_Sup_L | 6 | 57 | −22 | −10 | 48 | 2.29 | |
Frontal_Inf_Tri_L | 47 | 60 | −44 | 24 | 4 | 2.58 |
3D surface rendered images of main effect analysis overlaid by the LogoS VSF difference between Low-C and High-C. The 3D rendered images showed the results of the main effect analysis of script overlaid by the greater activation in the Low-C than High-C for LogoS than for PhonoS reading. The overlaid surface rendered images clearly showed the right occipital cortex extending to the right SPL (red dotted circle, posterior and superior view) and right VLPFC (red arrow, superior view) including the SMA (blue dotted circle, superior view) were predominant in the Low-C (uncorrected
Regarding OrthoR, the pairs of Korean Hangul (PhonoS) and Hanja (LogoS) are similar to the pairs of Japanese Kana and Kanji while the Korean writing system has an interestingly unique property. The spatial pattern between Hangul and Hanja script is very similar. Unlike the letters of the Latin alphabet, the Hangul is packaged into “square” units of subsyllable, each of which transcribes a syllable. It is actually composed of several consonant and vowel letters with the form of left to right or from top to down direction. The spatial features of Hanja have geometrical relationship between radicals, which is similar to that of Hangul [
Additionally, controlling the age at which language acquisition induces different neural activation is needed. Regarding the age of acquisition, a spatially segregated region was found activated in Broca’s area but not Wernicke’s area in late L2 learners during L1 and L2 reading, whereas this phenomenon was not observed in early L2 learners; they shared a substantially overlapped region (both Broca’s area and Wernicke’s area) during L1 and L2 reading [
Since the 14th century, Hangul (PhonoS) and Hanja (LogoS) have been the two basic scripts used in Korea, often combined for the purpose of better communication while Hangul and Hanja represent shallow and deep OrthoR, respectively. Also, because of the government policy, younger Korean generation currently is being exposed to Hanja starting in middle school, which leads to control of the age of language acquisition, consequently avoiding the potential confounder inducing different neural activation [
To delineate the effect of VSF on the cortical representation, an implicit word reading task was performed for Hangul and Hanja in the subjects with L1 as Hangul divided into High-C versus Low-C group during fMRI acquisition. Compared with High-C group, Low-C group showed greater involvement of the right hemisphere including the right SPL, left SMA, and right IPcG during Hanja compared to Hangul reading. Given Low-C group less proficient in Hanja, the areas of greater activation in the Low-C might not be due to intrinsic linguistic processing such as OrthoR but increased demand for attention and cognition such as VSF [
While phonological awareness is closely related to reading and learning PhonoS such as English, due to visual-spatial complexity, visuospatial processing is crucial for reading in LogoS (e.g., Chinese, Korean Hanja) [
The experimental studies suggested that the right parietal cortex has an important role in visuospatial processing such as visuospatial attention and mental rotation of an object in space [
For cortical mapping to represent Hangul (PhonoS) versus Hanja (LogoS) in the subjects with the first language (L1) PhonoS divided into two groups as high versus low competency to L2 LogoS (High-C versus Low-C), Low-C group demonstrated activation and lateralization in the right SPL and VLPFC as well as left SMA, likely due to neural efficiency such as visuospatial attention and cognition rather than core neurolinguistic neural demand like OrthoR processing.
The authors have no conflicts of interest.
Byung-Gee Bak and Young-Bo Kim contributed equally to this work.
The authors would like to thank Dr. Michael DiGianvittorio (Johns Hopkins University School of Medicine, Baltimore, MD, US) for his assistance with the preparation of this manuscript. The authors would also like to acknowledge KAPHE for helping with data collection and Jemin Ahn for helping with data analysis. This work was supported by the National Research Foundation of Korea Grant funded by the Korean Government (NRF-2013S1A5A2A03045081) and GRRC (Gyunggi Regional Research Center, Artificial Intelligence Health Care Research Center).