Inter-Subject Neuronal Correlation during Natural Article Reading
Date Issued
2015
Date
2015
Author(s)
Liu, Yun-Fei
Abstract
Most neuroimaigng studies on reading comprehension use well-controlled simplified materials and show them in a way different from natural reading. In this study, we used functional magnetic resonance imaging (fMRI) with texts presented in a natural way to reveal brain areas sensitive to word choices and arrangements. Specifically, articles from the New York Times and the Reader’s Digest translated either by human or machine were presented to participants. The correlation of brain activity across participants during article reading was calculated. This experiment design allowed us to study how the common brain activity changes between text genres (New York Times vs. Reader’s Digest) and translations (human vs. machine translations). We found that spatial distributions of correlated fMRI signal across participants were modulated by article genre and translation. Across text genres and translation, we found low inter-subject correlation (ISC) at superior and middle temporal cortices and high ISC at dorsal medial pre- and post-central cortices. Comparing between text gernes, we found ISC values at right anterior temporal lobe were lower during news article reading than during fictional article reading. Comparing between translations, increased ISC values were observed at right inferior frontal cortex when reading texts translated by machine. Taken together, the right hemisphere is more activated when a higher demand in the reading comprehension capacity is required. The experimental approach and results described in this study may be useful for neurolinguists to further elucidate brain processes related to natural language comprehension.
Subjects
inter-subject correlation (ISC)
readability
natural stimuli
right hemisphere
functional magnetic resonance imaging (fMRI)
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-104-R02945001-1.pdf
Size
23.32 KB
Format
Adobe PDF
Checksum
(MD5):c2b9a9080800270c1a5dbcba2a7040f0