T.Taguchi

network analysis, information theory, fMRI, multimodal model, conscious perception

[Murayama lab, RIKEN CBS @mlab_cbs]

新作論文：意識・無意識脳での神経ネットワーク構造の可視化に成功-睡眠中に感覚応答を知覚できない脳の謎にヒント-ハブ細胞の役割とは？大泉ラボ@oizumimとのコラボ。同じデータでマクロからミクロまでネットワーク指標を解析するパイプラインを構築
https://t.co/8mxBzBRTDb
https://t.co/s2rlMjj4Gg

[Masafumi Oizumi @oizumim]

I am looking to hire 1–2 postdocs for next year. Please feel free to reach out if interested in: (1) Inter-brain information translation & control (2) Characterizing brain states via thermodynamics & control theory (3) Comparing qualia structures across individuals and identifying their neural substrates.

Application details coming soon.

[Masafumi Oizumi @oizumim]

来年度に向けてポスドクの方1〜2名募集しています！(1) 異なる脳の間での情報翻訳と制御 (2) 異なる脳状態の熱力学・制御理論的な観点からの特徴づけ (3) クオリア構造の個人間比較とその神経基盤等に興味のある方はお気軽にご連絡下さい！
詳細はこちらをご参照下さい。
https://t.co/AE9tbKrD70

[Shuntaro Sasai (笹井俊太朗) @shuxnys]

【第2弾】研究員（脳解読/脳基盤モデル）

大規模脳データ（EEG、MEG、ECoG等）からのスケーリング則研究と脳基盤モデル構築を行っていただきます。マルチモーダル深層学習で神経デコーディングの最先端を開拓する研究者を募集します。

東京（状況に応じてリモートOK）
年収：600–900万円

[H.Asanuma @H_A_Research]

My first-authored paper has just been published in Scientific Reports!
Our theme:
👉 “To what extent can multimodal large language models (MLLMs) estimate the complex structure of human emotions?”
https://t.co/Rd4eOTSdkA

[Masafumi Oizumi @oizumim]

Excited to share our new preprint! "Principal bundle geometry of qualia: Understanding the quality of consciousness from symmetry", co-first-authored with Ryota Kanai @kanair_jp and co-authored with Chanseok Lim. https://t.co/8CXydqcpG8
We propose that the relational structure of qualia is characterized by principal bundle geometry, which naturally arises from the symmetries that the brain learns from the world.

[Tomoya Nakai @TomNakai]

新しいプレリントを公開しました：
https://t.co/Z5AoWfboNJ

音声とテキストが同時に与えられた時、選択的注意を向けた刺激の意味情報の方が、脳活動によるデコーディング精度が高くなります。

[Ryota Kanai 🌙 （じじぃ） @kanair_jp]

アラヤの笹井さん率いるチームからInterspeech2025に採択された論文をNotebookLMに解説してもらったら、とっても理解しやすかったのでぜひ聞いてみてください。
https://t.co/nwqfIEkG3y

脳波と筋電図を使って、声を出さない「サイレントスピーチ」を実現する研究です。
https://t.co/xJLi8BqIom

[Ken Takeda @kentakeda1248]

Excited to share our new paper in iScience! 🎉
https://t.co/OOvCHLwYcg

We present an unsupervised alignment framework applied to neural recordings, enabling comparison of neural representations across individuals and brain areas—without relying on stimulus correspondence.

[Ken Takeda @kentakeda1248]

Excited to share another new paper in Journal of Neuroscience Methods!
We introduce GWTune, a toolbox for unsupervised alignment using Gromov–Wasserstein Optimal Transport (GWOT).
Open-source and ready to use.
🔗 https://t.co/Avsu3xUIsu
🧰 https://t.co/ZVoPNrzQSa

[Annie G. Bryant @AnnieGBryant]

Info theory offers powerful measures for capturing complexity & interaction among elements of a complex system, like the brain! 🧠 Here's our new unified reference for key info-theoretic time series measures ft. 📊 visuals, ➗equations, & 💬descriptions:
https://t.co/nbBxTNHmhU

[H.Asanuma @H_A_Research]

アーカイブ出ました〜〜！！
マルチモーダルLLMが人間の感情構造をどの程度推定できているか検証した論文です、もしよければぜひ読んでみてください〜〜！
https://t.co/wRLuaPQCpP

[Masafumi Oizumi @oizumim]

Our latest preprint! @H_A_Research @takato1414 @NishimotoShinji
https://t.co/VFYfcKEpkM
We investigated to what extent Multimodal LLMs replicate human high-dimensional emotion structures using unsupervised alignment. MLLMs achieve unprecedented accuracy at capturing emotion structures, but precise alignment remains challenging.

[Jun Kitazono @JunKitazono]

So happy to see Taguchi-san’s paper published! I had the pleasure of closely guiding this work — it’s been incredibly rewarding to watch it take shape over time. A real testament to his perseverance and dedication — well done!

For more details, check out the thread under his original post.

For me, this marks the latest in a series of studies on identifying network cores, particularly in the context of consciousness:
– Informational cores: https://t.co/nRxqlRFQ5G (w/ @kanair, @oizumim)
– Bidirectionally connected cores in a mouse structural connectome: https://t.co/8bAKp1OyLD (w/ Y. Aoki, @oizumim)

Check them out too!

[Masafumi Oizumi @oizumim]

Our paper is out in #JNeurosci! @_ttaguchi @JunKitazono @shuxnys

Bidirectional signaling (feedforward and feedback loops) is ubiquitous in the brain and considered fundamental to brain function—but where exactly in the brain do these strong bidirectional networks emerge, and how are they related to specific cognitive functions?
Dive into the details in this thread by @_ttaguchi 👇
https://t.co/NK5knKjKJI

[T.Taguchi @_ttaguchi]

Our proposed framework offers new insights into the roles of network cores with strong bidirectional interactions, giving us a better understanding of their influence on conscious perception and sensorimotor functions.

For more details, check out our full study! (11/11)

[T.Taguchi @_ttaguchi]

Here, we introduced a new framework to identify subnetworks with strong bidirectional interactions, which we call the “cores” of a network. The framework consists of two steps. (4/n)

[T.Taguchi @_ttaguchi]

Further analyses, including a meta-analysis, suggest that the central cores relate to lower-order sensorimotor functions. The fig shows that regions with strong bidirectional interactions are associated with terms such as “eye movements,” “visual perception,” and “motor.” (10/n)