Authors :
Presenting Author: Yu Kitazawa, MD, PhD – Wayne Sate University
Kazuki Sakakura, MD, PhD – Wayne State University; Hiroshi Uda, MD, PhD – Wayne State University; Naoto Kuroda, MD – Wayne State University; Riyo Ueda, MD, PhD – Wayne State University; Ethan Firestone, MS – Wayne State University; Hirotaka Iwaki, MD, PhD – Hachinohe City Hospital; Masaki Sonoda, MD, PhD – Yokohama City University; Takumi Mitsuhashi, MD, PhD – Juntendo University; Shin-ichiro Osawa, MD, PhD – Tohoku University; Kazushi Ukishiro, MD – Tohoku University; Makoto Ishida, PhD – Tohoku University; Kazuo Kakinuma, MD – Tohoku University; Shoko Ota, PhD – Tohoku University; Yutaro Takayama, MD, PhD – Yokohama City University; Keiya Iijima, MD, PhD – National Center of Neurology and Psychiatry; Toshimune Kambara, PhD – Hiroshima University; Kyoko Suzuki, MD, PhD – Tohoku University; Nobukazu Nakasato, MD, PhD – Tohoku University; Masaki Iwasaki, MD, PhD – National Center of Neurology and Psychiatry; Eishi Asano, MD, PhD, MS (CRDSA) – Wayne State University
Rationale:
Prior intracranial studies on English-speaking patients with drug-resistant focal epilepsy have shown that high-gamma activity increases during the comprehension of questions starting with wh-interrogatives, effectively localizing language areas and predicting postoperative language outcomes. Considering Japanese language, where wh-interrogatives can occur both at the beginning and end of questions, we investigated the spatiotemporal attributes of cortical and white matter connectivity modulations during comprehension of sentences beginning and ending with ‘what’, ‘when’, and ‘where.’ Methods:
We included a cohort of 40 Japanese-speaking patients with drug-resistant focal epilepsy (age 6-54 years) who underwent resective epilepsy surgery at two tertiary epilepsy centers in Japan. Extraoperative intracranial EEG recording was performed during an auditory descriptive naming task, where patients were asked to verbally respond to 96 questions. Half of these questions began with 'wh-interrogative,' followed by 'concrete object or adverb,' and then 'verb.' The remaining questions began with 'concrete object or adverb,' followed by 'verb,' and ended with 'wh-interrogative.' This structure ensured that each type of wh-interrogative was represented equally. Using time-frequency analysis, we identified shared and distinct patterns of neural modulation across question structures.Results:
Independent of wh-interrogative type or positioning, we observed high-gamma augmentation in bilateral superior-temporal regions during question listening, and there was enhanced inter-hemispheric functional connectivity through callosal fibers. High-gamma activity in the left posterior inferior frontal gyrus (pIFG) rose at the end of the ‘concrete object or adverb’ phrase, coupled with enhanced intra-hemispheric functional connectivity via the left arcuate fasciculus until response. Subsequent high-gamma augmentation was noted in the bilateral Rolandic regions during verbal responses, with enhanced inter-hemispheric functional connectivity through callosal fibers. ‘What’-initiated questions induced higher high-gamma augmentation in the left pIFG during listening of ‘verb’ phrase than ‘when’ or ‘where’ did.Conclusions:
Our normative atlases demonstrated the neural modulations underlying comprehension of sentences that incorporate uniquely ordered wh-interrogatives. The left arcuate fasciculus between the pIFG and temporal lobe regions might not be utilized as rapidly in understanding questions that begin with a wh-interrogative compared to those starting with a concrete object or adverb. Moreover, the left pIFG seems to be engaged more intensively when processing questions starting with 'what' as opposed to 'when' or 'where.' Funding:
NIH grant (NS64033 to E. Asano)
Children's Foundation (R2-2021-42 to E.Asano)
JSPS (JP22J23281 to N.Kuroda)
JSPS (JP23KJ2197 to R.Ueda)