Lmst

Fabrizio Musacchio boosted:

(1/2) 37 years ago today I submitted my proposal for the World Wide Web 🎂. Today, Rosemary & I spoke with students in New Orleans at Walter Isaacson's Digital History Class at Tulane University. I was asked, as I often am, if I ever could have foreseen where we’d be today. I could not.

🧠 New preprint by Garcia-Garcia et al.: The authors show that cerebellar #GranuleCells do not simply expand #cortical activity into a high-dimensional code. Instead, they preserve low-dimensional cortical #manifold geometry while reorienting it across contexts. This rotation separates similar tasks, reduces interference, and supports flexible dual-task learning.

📄 https://www.biorxiv.org/content/10.64898/2026.03.03.709240v1.full.pdf

#Neuroscience #NeuralDynamics #CompNeuro

Fabrizio Musacchio boosted:

↔️ Follow motion or light?

How does the brain process the complex flood of visual inputs? A new study on larval zebrafish, led by Bernstein member Armin Bahl, provides answers by combining behavioural experiments, advanced brain imaging, and computational modelling.

Read the whole story 👉 https://bernstein-network.de/en/newsroom/news/260305/

@unikonstanz #CompNeuro #BernsteinNetwork

🧠 New paper by Pezon, Schmutz & Gerstner: Linking #NeuralManifolds to circuit structure in recurrent networks.

The study connects two common views of neural activity: low-dimensional #PopulationDynamics (“neural manifolds”) and single-neuron selectivity. Using recurrent network models, the authors show how circuit connectivity constrains both the geometry of neural #manifolds and the tuning of individual neurons.

📄 https://doi.org/10.1016/j.neuron.2025.12.047

#Neuroscience #NeuralDynamics #CompNeuro #RNN

Figure 1 Neuronal activity and circuit structure—terminology

Figure 3 There is no one-to-one relationship between the circuit structure and the neural manifold

Figure 4 Neuronal embeddings distinguish different models implementing the same solution to a decision-making task

🧠 New preprint by Guardamagna et al.: Using large-scale recordings in #rat pups, the authors show that toroidal #manifolds in #MEC emerge by P10, before eye and ear opening, upright gait, and active exploration. Ring-like manifolds appear even earlier, by P9. External spatial experience seems to align these preconfigured internal maps only later, as pups begin to navigate.

📄 https://doi.org/10.64898/2026.03.10.710908

#Neuroscience #GridCells #NeuralDynamics

Figure 1. Toroidal manifolds precede onset of navigation

🧠 New by Codol et al. who show that #MotorCortex dynamics are remarkably conserved across #mice, #monkeys, and #humans. Despite very different #behaviors, #NeuralPopulation activity follows similar dynamical rules on low-dimensional #manifolds. Species differences arise mainly from the geometry of trajectories within this shared #DynamicalSystem.

📄 https://doi.org/10.64898/2026.03.06.709637

#Neuroscience #CompNeuro #NeuralDynamics

Figure 1. Hypothesis of evolutionarily conserved neural
computations through population dynamics.

Figure 2. A multi-species motor neurophysiology datase

Fabrizio Musacchio boosted:

🌍 Vor 75 Jahren erschütterte ein starkes #Erdbeben am 14. März 1951 die Region um Euskirchen, Köln und Bonn. Mit einer Magnitude von bis zu 5,7 verursachte es erhebliche Schäden. Auch in Köln war das Beben deutlich spürbar.

Damals gab es noch keine seismischen Messstationen im Rheinland. So entstand die Erdbebenstation Bensberg – heute Teil des Instituts für Geologie und Mineralogie der Universität zu Köln.

Mehr dazu ▶️ https://uni.koeln/B47RA

#uniköln #unicologne

Kleines weißes Haus mit Satteldach, Holzzaun und Schild „Erdbebenstation Bensberg“ am Eingang.

Gestapelte, aufgerollte großformatige Papierbögen in einem Archivregal.

Nahaufnahme eines Seismogramms mit roten Zickzacklinien auf weißem Papier.

🧠 New paper by Cowley et al: Compact deep neural network models of the #VisualCortex .

The authors show that neural responses in primate visual cortex can be predicted by much smaller, interpretable #DeepNeuralMetworks than previously assumed. By constraining model architecture and parameters, they achieve strong predictive performance while revealing computational structure in cortical processing.

📄 https://www.nature.com/articles/s41586-026-10150-1

#Neuroscience #AI #CompNeuro #DNN

Fig. 1| Identifying compact models of macaque V4

🧠 New preprint by Chericoni et al: #NeuralPopulation activity in the #hippocampus encodes spatial information for different agents (self, prey, gaze) in distinct but related low-dimensional subspaces. The study shows that these representations can be linearly transformed between each other, suggesting a shared geometric code supporting flexible spatial #cognition and multi-agent navigation.

📄 https://arxiv.org/abs/2603.04747

#Neuroscience #CompNeuro #NeuralDynamics

Figure 2. Hippocampal mapping functions for self and prey. A. Bar plot showing the
proportion of neurons with significant tuning for agent positions according to the GLM.
Proportions do not sum up to one hundred as one neuron may be tuned to more than one
agent’s position. Dashed horizontal gray line: 5% chance level. Error bars: standard errors. B.
Venn diagram showing overlap between neurons selectivities. Neurons generally encoded
positions of multiple agents. C. Distribution of correlations between observed and GLM
predicted firing rates for significantly tuned neurons only. Positive correlations indicate that the
GLM reliably captured each neuron’s spatial firing pattern. D. Heatmaps showing tuning of three
example neurons to the positions of the self, chosen prey, and unchosen prey. Neurons often
showed a single hotspot but typically showed more complex features

RE: https://mastodon.social/@thetransmitter/116205669054942061

#Hippocampus #learning #Neuroscience 👇

Fabrizio Musacchio boosted:

A big win for ODF, the EU Commission DG CONNECT has added the ODS template to the CRA Guidance feedback process, reacting promptly in a positive way to our request. Kudos to DG CONNECT.

https://blog.documentfoundation.org/blog/2026/03/10/important-odf-template-added-to-cra-guidance-feedback/

@libreoffice @ODS @ODF @digitalsovereignty @digitalfreedom

Fabrizio Musacchio boosted:

Researchers identified a new class of neurons in the mouse primary visual cortex possessing a two-part receptive field tuned to complex textures and spatial frequencies, challenging the classical model that these early-stage neurons only detect simple transitions in brightness.
#ComputationalNeuroscience #Neurobiology #Neuroscience #ArtificialIntelligence #sflorg
https://www.sflorg.com/2026/03/ns03102601.html

🛰️ New paper by Knowles et al.: Using the #JWST, the authors observed rapid variability in the auroral footprints of #Jupiter’s #moons in Jupiter’s #atmosphere.

IR observations show that the footprints of moons like #Io can change on short timescales, highlighting the dynamic interaction between Jupiter’s #magnetosphere and its satellites. The moon–magnetosphere coupling driving Jupiter’s #aurora is more variable than previously assumed.

📄https://doi.org/10.1029/2025GL118553

#SpacePhysics #PlanetaryScience

Figure 1: JWST/NIRSpec IFU observations of the auroral footprints of Io and Europa, indicated by yellow and purple arrows, respectively. (a) Integrated
radiance with planetocentric latitude at 550 km above the 1-bar level (dotted) and System III (West) longitude (solid). (b) Total
emission with dotted gray lines for magnetic dip angle (degrees). (c) Integrated fundamental emissions with total magnetic field strength in Gauss (blue dotted) and (d) integrated hotband emissions with dash-dotted contours of the magnetic footprints of Io (5.9 ), Europa (9.4 ) and main auroral oval (25.0 ). UTC mid-points are given above.

Figure 2:
temperature and column density spectral retrievals. (a)
temperature, (b) uncertainty of the retrieved temperature, (c)
column density, and (d) its uncertainty.

#Cologne at night.
#WeekendStories #NoFusion

#Galileo’s handwritten notes found in ancient astronomy text. Discovery sheds new light on how famed astronomer came to lead a scientific revolution:

🌍 https://www.science.org/content/article/galileo-s-handwritten-notes-found-ancient-astronomy-text

#Physics #histodons #history

spread pages of a 16th century printing of The Almagest
A historian has discovered that a 16th century printing of The Almagest—a highly influential ancient astronomy text—contains extensive annotations from Galileo Galilei, the astronomer who later overthrew that text’s conception of a geocentric cosmos.

Fabrizio Musacchio boosted:

"Sketch of a novel approach to a neural model", by Gabriele Scheler 2026.
https://arxiv.org/abs/2209.06865

"traditional synapse-centric, weight-based models of memorization are not sufficient or adequate to capture the real complexity of neuroplasticity. [...] We propose a paradigm switch from a synapse-centric model (each synapse learns independently, based on associative coupling) to a neuron-centric model (each neuron uses its intracellular pathways to express plasticity at its synapses and dendritic membrane)."

#neuroscience #CompNeurosci

1/2

🛰️ New paper by Němec et al.: After analyzing ~10 years of #MAVEN wave data, the authors report a #lightning-like electromagnetic #whistler signal in the #Martian #ionosphere. Its frequency dispersion matches theoretical propagation through #Mars’ ionosphere and crustal magnetic fields, suggesting that electrical discharges may occur in the Martian #atmosphere.

🌍 https://doi.org/10.1126/sciadv.aeb4898

#PlanetaryScience #SpacePhysics

Fig. 1. Schematic showing the formation of a whistler.
A short electromagnetic pulse generated by an electric discharge (orange circle) spreads over time due to dispersion in the ionospheric plasma, where higher-frequency signals propagate faster than lower-frequency signals. These waves propagate to spacecraft altitudes (orange square) at locations where the magnetic field lines (black curves) are nearly vertical.

Fig. 2. Frequency-time spectrogram of electromagnetic waves measured by MAVEN.
Wave intensity is color-coded according to the scale on the right. A dispersed whistler is observed in the second half of the time interval at frequencies up to about 110 Hz. The black dashed line shows a theoretical spectral shape for an impulsive source in the Martian atmosphere, calculated from a model with realistic magnetic field and plasma density profiles.

Fabrizio Musacchio boosted:

DNA is a universal language of all life forms on Earth. It is not easy to read because there are multiple layers of meaning. This tool and similar ones will help to make sense of the language, e.g. understand disease potential of mutations. Misuse possible. Discuss! https://www.nature.com/articles/s41586-026-10176-5

#March 2026
#WeekendStories #Cologne #NoFusion