TY - JOUR
T1 - Distinct role of spatial frequency in dissociative reading of ideograms and phonograms
T2 - An fMRI study
AU - Horie, Shizuka
AU - Yamasaki, Takao
AU - Okamoto, Tsuyoshi
AU - Kan, Shigeyuki
AU - Ogata, Katsuya
AU - Miyauchi, Satoru
AU - Tobimatsu, Shozo
N1 - Funding Information:
This work was supported in part by Grants-in-Aid for Scientists , No. 18659259 (to S.T.) and No. 20591026 (to T.Y.), from the Ministry of Education, Culture, Sports, Science and Technology in Japan . We would like to thank Mr. Seiji Itoi and Mr. Takahiko Koike for their technical contributions.
PY - 2012/11/1
Y1 - 2012/11/1
N2 - It has been proposed that distinct neural circuits are activated by reading Japanese ideograms (Kanji) and phonograms (Kana). By measuring high-density event-related potentials, we recently reported that spatial frequency (SF) information is responsible for the dissociation between Kanji and Kana reading. In particular, we found close links between Kana and low SF (LSF) information and between Kanji and high SF (HSF) information. However, it remains unclear which brain regions contribute to this dissociation. To determine this, we performed functional magnetic resonance imaging while presenting unfiltered or spatially filtered Kanji and Kana word stimuli to healthy native Japanese subjects. Fourier analysis revealed that Kanji and Kana stimuli were characterized by HSF and LSF information, respectively. When presented with either type of unfiltered stimulus (Kanji or Kana), the bilateral inferior temporal (IT, BA 37) regions were activated compared to the resting condition. Kana but not Kanji reading also activated the bilateral inferior parietal lobules (IPL, BA 40). When we compared Kanji and Kana reading directly, the left IT region was significantly activated by Kanji reading, while significant activation of the left IPL was observed during Kana reading. In response to filtered HSF stimuli, the Kanji reading minus Kana reading comparison revealed significant activation of the left IT region but not the left IPL. Conversely, significant activation of the left IPL but not the left IT region occurred in the Kana reading minus Kanji reading comparison for filtered LSF stimuli. These results suggest that Kanji and Kana engage a relatively overlapping network, within which the left IT is more involved in Kanji processing, while the left IPL contributes more to Kana processing. The preferential engagements of these brain regions could reflect the close links between Kana and LSF information, and between Kanji and HSF information. Therefore, this study provides further evidence that SF contributes to the dissociation between Kanji and Kana reading.
AB - It has been proposed that distinct neural circuits are activated by reading Japanese ideograms (Kanji) and phonograms (Kana). By measuring high-density event-related potentials, we recently reported that spatial frequency (SF) information is responsible for the dissociation between Kanji and Kana reading. In particular, we found close links between Kana and low SF (LSF) information and between Kanji and high SF (HSF) information. However, it remains unclear which brain regions contribute to this dissociation. To determine this, we performed functional magnetic resonance imaging while presenting unfiltered or spatially filtered Kanji and Kana word stimuli to healthy native Japanese subjects. Fourier analysis revealed that Kanji and Kana stimuli were characterized by HSF and LSF information, respectively. When presented with either type of unfiltered stimulus (Kanji or Kana), the bilateral inferior temporal (IT, BA 37) regions were activated compared to the resting condition. Kana but not Kanji reading also activated the bilateral inferior parietal lobules (IPL, BA 40). When we compared Kanji and Kana reading directly, the left IT region was significantly activated by Kanji reading, while significant activation of the left IPL was observed during Kana reading. In response to filtered HSF stimuli, the Kanji reading minus Kana reading comparison revealed significant activation of the left IT region but not the left IPL. Conversely, significant activation of the left IPL but not the left IT region occurred in the Kana reading minus Kanji reading comparison for filtered LSF stimuli. These results suggest that Kanji and Kana engage a relatively overlapping network, within which the left IT is more involved in Kanji processing, while the left IPL contributes more to Kana processing. The preferential engagements of these brain regions could reflect the close links between Kana and LSF information, and between Kanji and HSF information. Therefore, this study provides further evidence that SF contributes to the dissociation between Kanji and Kana reading.
UR - http://www.scopus.com/inward/record.url?scp=84866170427&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84866170427&partnerID=8YFLogxK
U2 - 10.1016/j.neuroimage.2012.03.046
DO - 10.1016/j.neuroimage.2012.03.046
M3 - Article
C2 - 22480729
AN - SCOPUS:84866170427
SN - 1053-8119
VL - 63
SP - 979
EP - 988
JO - NeuroImage
JF - NeuroImage
IS - 2
ER -