Jan Fousek

Phil Kragel @pkragel.bsky.social showing how they can use EmoNet to predict amygdala responses and predict reinforcement learning behavior. #ABIM2025 #neuroskyence

True, I was wondering if publishing benchmarks as extensible platform would be a viable alternative (citable versions). Both cloud and local execution possible. People can then add and help maintain their methods…

Yes, do the hemodynamics please! :) Would also love to hear your thoughts on changes of HRF across brain areas / age / brain states..

This project is a result of the enthusiastic work of a group of students from the EITN Fall School in Computational Neuroscience 2023, which I had the privilege to tutor. We had a blast, and I hope our paths will cross in future again.

Second, we show that the amplitude decrease on the sensor level reflects decrease in inter-hemispheric synchrony on the source-level. Both these source-level observations could be used as pointers to follow-up studies on experimental datasets. 🥁

But having the access also to the source-level activity, we can peek into where are these changes coming from. First, we show that the rate of change of the slow-wave properties is proportional to the rate of change of node strength.

Indeed, we were able to replicate the empirical observations across the across the synthetic aging trajectory, including the decrease in slow wave frequency, increase in variability, and decrease in amplitude.

We swapped out the neural mass model for one capturing the sleep-like dynamics (AdEx), and then followed the properties of the simulated slow waves across the aging trajectory bu decreasing gradually the inter-hemispheric connections in representative young connectomes.

In VAB - using brain network modeling - we linked the disruption of the interhemispheric structural connections to change in resting state dynamics and cognitive decline.

doi.org/10.1016/j.ne...

Here we link this decrease of the excitatory drive to the network changes across the aging trajectory. To do this we tapped into our previous work on the Virtual Aging Brain (VAB).

Frontiers | Slow and Fast Neocortical Oscillations in the Senescence-Accelerated Mouse Model SAMP8 The senescence-accelerated mouse prone 8 (SAMP8) model is characterized by accelerated, progressive cognitive decline as well as Alzheimer’s disease (AD)-lik...

What is cool about this is, that it is also consistent with the decrease of down-state spontaneous activity in the Senescence-Accelerated mice.

doi.org/10.3389/fnag...

This paper provided not only the empirical observations, but also a computational model showing that these changes can be replicated by decreasing external excitatory drive at the level of a neuronal population.

During aging, the slow waves are getting flatter and occur less frequently and with more variability. This was shown on a sample of over 2300 subjects using EEG head band during sleep.

doi.org/10.1101/2023...

A Whole-Brain Model of the Aging Brain During Slow Wave Sleep Age-related brain changes affect sleep and are reflected in properties of sleep slow-waves, however, the precise mechanisms behind these changes are still not completely understood. Here, we adapt a p...

Happy to share our latest work here : slowing down of the sleep waves during ageing can be explained by alterations in structural connectivity.

doi.org/10.1523/ENEU...

Let's have a 🧵, dear #neuroskyence, shall we?

Great to see Czech Republic going green on the EBRAINS map! Bigger community means exciting new opportunities to get creative on the frontier of digital neuroscience. 🧠🎊

Excalidraw. It is simple, but often when I feel it is too simple, it is because the figure I’m trying to do is overcomplicated… Inkscape otherwise.