Translated title. Original: Gehirnforschung: „Unser Traum ist genauso groß oder größer als der von Musk“
The text highlights brain-computer interfaces as a transformative field, notes the limited current understanding of the brain, contrasts the zebrafish brain with the human brain in neuron counts, and emphasizes the brain's central role in identity and the impact of neurodegenerative diseases like Alzheimer's.
Chinese researchers present evidence that individual neurons can function like minicomputers, with dendrites integrating learned rules and sensory input to guide flexible decisions in mice; results suggest potential paths for more adaptable AI inspired by brain processes.
Neuroscience and AI research intersect through citation patterns, showing AI often draws on broad neuroscience ideas rather than specific neural mechanisms. While the share of AI papers citing neuroscience is small, the total number of neuro-inspired AI papers is still substantial, prompting questions about where to focus effort at the NeuroAI interface in the post-2022 AI boom.
The article discusses the evolving relationship between neuroscience and AI, noting that while AI's core remains neural networks, the brain-AI link is largely conceptual. It questions whether NeuroAI yields genuine mutual benefits and uses OpenAlex data to examine citation patterns between the fields.
A review of tools and concepts for analyzing low-dimensional population dynamics in neural networks, linking physiology, modeling, and neurotechnology, with emphasis on separable population representations and their translational use in brain-computer interfaces.
Translated title. Original: ASCII.jp：利き手を切り替える脳の仕組み、脳内スイッチの正体を解明
A JSTnews feature reports on a study of how the brain switches dominant hands, using mice to show that acetylcholine activity in the opposite hemisphere is necessary for switching, with lateralized changes during training and implications for brain asymmetry and rehabilitation.
FinalSpark says it is growing neurons in cell cultures to build biological neural networks for computing. The company describes large-scale network construction as a challenge and says its lab is working to expand these systems.
The article reports that curiosity shapes brain-like complexity in both natural and artificial systems. Experiments with curious versus non-curious neural networks show that curiosity drives modular, hierarchical brain-like structures, improves transfer to new tasks, and aligns with developmental patterns observed in humans.
Researchers show human brain organoids can classify Braille letters from tactile input with increasing accuracy, exploring organoid-based computing. The work is preliminary, open-loop, and highlights both potential and limitations of using living neural tissue for computing.
Researchers demonstrate that human brain organoids can classify Braille letters from tactile data with increasing accuracy as more organoids are used, but the experiment is open-loop and does not show learning; the work sits within the nascent field of organoid intelligence and cites both potential benefits and current limitations.