Tamas Spisak @tspisak.bsky.social · 28/09/2026Beyond pain: a brain-based biomarker predicts individual pain relief 056
Tamas Spisak @tspisak.bsky.social · 25/08/2026But on the deeper question of what human-aligned AI could ultimately be optimized for - possibly using techniques like the one we propose here: I genuinely have no idea! But thinking about it really highlights just how ill-posed the whole alignment problem is! 130
Tamas Spisak @tspisak.bsky.social · 25/08/2026In the next phase, we’re going towards performance-maxxing in a gaming environment. But we’re also excited to explore pain signatures (e.g. NPS) soon, potentially optimizing stimuli (music?) to reduce pain-related brain responses. 100
Tamas Spisak @tspisak.bsky.social · 25/08/2026Reinforcement Learning from Brain Feedback! Our new preprint by @ggallitto.bsky.social introduces an open-source framework and promising pilot results for using real-time fMRI to optimize (possibly AI-generated) stimuli to elicit a target neural response: doi.org/10.64898/202... 151
Reposted by Tamas SpisakThe Power of Expectations @sfb-trr-289.bsky.social · 03/08/2026⏰ New preprint ⏰ "Layer-specific cortical processing dissociates sensory and cognitive influences on pain" www.biorxiv.org/content/10.6... Collab w. @elh-institute.bsky.social, @tspisak.bsky.social, @ulrikebingel.bsky.social and some people not on here 😍 196
Reposted by Tamas SpisakThe Power of Expectations @sfb-trr-289.bsky.social · 22/07/2026💐✨ Huge congrats to Dr. Balint Kincses ✨💐 He has received the IFORES Career Kickstart grant for "Layers of pain: uncovering the mechanisms in the cortical integration of bottom-up and top-down processes in pain perception and its modulation using ultra-high field fMRI" @elh-institute.bsky.social 081
Reposted by Tamas SpisakThe Power of Expectations @sfb-trr-289.bsky.social · 20/07/2026More than words: Experience shapes #placebo pain relief through additional brain pathways 🧠 We're excited to share our latest publication in @natcomms.nature.com! 🤩 🔗 Read the paper: rdcu.be/fuxmn #neuroskyence #PsychSciSky #Neuroscience #PainResearch #BrainImaging #TeamScience #Reproducibility 11810
Tamas Spisak @tspisak.bsky.social · 19/05/2026Thanks! This package was really meant to be a one-trick pony, and I thought it was important to make that clear from the start. 010
Tamas Spisak @tspisak.bsky.social · 19/05/2026Many of you asked how we create those pretty whole-brain plots with the cortical outline. We just made it available as a Python package: pni-lab.github.io/quickbrain/ 35923
Tamas Spisak @tspisak.bsky.social · 28/04/2026The brain’s “default mode” and “action mode” networks are two sides of the same attractor. Encoding a macro-scale Bayesian prior that biases processing toward internal or external drive. 020
Tamas Spisak @tspisak.bsky.social · 22/04/2026Looking forward to the colloquium in Bochum next Wednesday! www.ini.rub.de/events/ 030
Tamas Spisak @tspisak.bsky.social · 17/04/2026Paper: www.sciencedirect.com/science/arti... Code: github.com/pni-lab/fep-... AI-perspectives: twitter-thread.com/t/2039126149... Application with fMRI data: elifesciences.org/articles/98725sciencedirect.com 000
Tamas Spisak @tspisak.bsky.social · 17/04/2026To me, what’s especially compelling is the scale-free nature of the framework: The same functional form applies across: • single neurons • microcircuits • large-scale brain networks Different substrates, different implementations, same computational imperative. 120
Tamas Spisak @tspisak.bsky.social · 17/04/2026Stochastic + rotational dynamics together enable spontaneous replay: This gives a mechanistic account of “resting-state” dynamics: with low sensory input (likelihood) the system revisits its own priors and consolidates existing memories. 100
Tamas Spisak @tspisak.bsky.social · 17/04/2026With sequential input, expected free energy minimization induces a solenoidal (rotational) flow component: flow moves along the landscape, not just downhill. This supports: • metastable dynamics (NESS) • efficient inference • sequence storage • "planning as inference" 100
Tamas Spisak @tspisak.bsky.social · 17/04/2026It naturally supports continual learning (a key challenge for current AI models). Learning targets unexplained variance → recalling new memories involves recalling the old correlated memories, too → no catastrophic forgetting 100
Tamas Spisak @tspisak.bsky.social · 17/04/2026Self-orthogonalization is gift! It yields: • disentangling latent causes → parsimonious "world model" • improved zero-shot generalization in the spanned subspace by *oscillating* across attractors • context emerging from the urge to orthogonalize • optimal memory capacity, redundancy & robustness. 100
Tamas Spisak @tspisak.bsky.social · 17/04/2026In a related fMRI connectivity work we see: • attractors aligned with connectivity eigenmodes • near-orthogonal attractor structure • yielding a generative model for large-scale brain dynamics In our framework, canonical brain networks (like the DMN) are priors of the brain's generative model. 100
Tamas Spisak @tspisak.bsky.social · 17/04/2026A key result is "self-orthogonalization". Even with correlated inputs, attractors become ~orthogonal. This effectively disentangles latent causes and pushes the system toward a projector-like (eigenmode-aligned) organization. And this is exactly what we see in brain data too! ↓ 100
Tamas Spisak @tspisak.bsky.social · 17/04/2026Learning is not separate from inference; it emerges from the same free energy imperative. When predictions fail, synapses update via predictive coding-like plasticity: • Hebbian → reinforce correlations • anti-Hebbian → suppress redundancy A biologically plausible, multi-scale plasticity rule. 110
Tamas Spisak @tspisak.bsky.social · 17/04/2026At the network level, this yields: → attractors (free energy minima) = prior beliefs → input reshaping trajectories = sensory "likelihood" → dynamics = perception: flow-based, irreversible (MCMC-like) sampling from the posterior 120
Tamas Spisak @tspisak.bsky.social · 17/04/2026Big Picture: To persist, a system must minimize its free energy - and so must all its components! If we follow the "free energy math" across the nested components of a random dynamical system, remarkable properties emerge: perception, learning, memory, spontaneous replay, planning, and much more! 110
Tamas Spisak @tspisak.bsky.social · 17/04/2026Attractor dynamics are a hallmark of brain function. But are they just epiphenomena? Starting from the free energy principle, we show that attractors can actually implement Bayesian priors in self-organizing networks, linking local neural dynamics directly to macro-scale probabilistic inference. 321
Tamas Spisak @tspisak.bsky.social · 17/04/2026Our work with Karl Friston on Self-Orthogonalizing Attractor Neural Networks is now out in Neurocomputing! What does this theoretical model mean for our understanding of the brain? I’ve mapped out the key neuroscience implications below. Read the thread for a neuroscience walk-through ↓ 273
Tamas Spisak @tspisak.bsky.social · 16/04/2026Really exciting to see my pTFCE work re-implemented in Python and taken even further by an independent group. Great to know pTFCE is in good hands! github.com/Don-Yin/pytfce arxiv.org/abs/2603.11344github.comGitHub - Don-Yin/pytfce: Fast probabilistic Threshold-Free Cluster Enhancement in PythonFast probabilistic Threshold-Free Cluster Enhancement in Python - Don-Yin/pytfce 000
Tamas Spisak @tspisak.bsky.social · 19/03/2026Out in @elife.bsky.social: Functional connectivity-based attractor dynamics of the human brain in rest, task, and disease doi.org/10.7554/eLif... 052
Tamas Spisak @tspisak.bsky.social · 12/01/2026Thank you! We have some fresh, initial evidence that the brain is akin to a self-orthogonalizing network as well. elifesciences.org/reviewed-pre...elifesciences.org 110
Tamas Spisak @tspisak.bsky.social · 12/01/2026doi.org/10.7554/eLif...doi.orgFunctional Connectivity-based Attractor Dynamics in Rest, Task, and Disease 000
Tamas Spisak @tspisak.bsky.social · 12/01/2026Brain attractors are approximately orthogonal to each other, suggesting that the brain may function as a self-orthogonalizing attractor network. Check out our revised manuscript about functional connectivity-based brain attractor dynamics in @elife.bsky.social. Link in comment! 2125
Tamas Spisak @tspisak.bsky.social · 07/11/2025www.nature.com/articles/s42...nature.comOn the replicability of diffusion weighted MRI-based brain-behavior models - Communications BiologyStructural connectome-based predictive models can yield replicable brain–behavior associations with moderate samples in several cases, but large datasets remain key for explainability, bias control, f... 011
Tamas Spisak @tspisak.bsky.social · 07/11/2025Replicability of BWAS with functional and structural MRI has been hotly debated. But what about DWI? Our new paper - first authored by @rkotikalapudi.bsky.social - shows that multivariate DWI models of trait-like phenotypes can be replicable, even with moderate sample sizes. 🔗 Link in comment! 111
Reposted by Tamas SpisakOle Goltermann @olegolt.bsky.social · 22/07/2025Serious concerns about a new cortical biomarker for pain sensitivity jamanetwork.com/journals/jam... We (with @tspisak.bsky.social, @christianbuchel.bsky.social) published a commentary on Chowdhury, Bi et al. (2025, JAMA Neurology) raising serious concerns about their reported results. 👇 1/13jamanetwork.comConcern About Predictive Performance of a Pain Sensitivity BiomarkerTo the Editor Chowdhury et al1 evaluated a biomarker for pain sensitivity, combining peak alpha frequency and corticomotor excitability. The authors report outstanding performance (validation set area... 47336
Tamas Spisak @tspisak.bsky.social · 30/05/2025@sfb-trr-289.bsky.social @karlfristonnews.bsky.social 010
Tamas Spisak @tspisak.bsky.social · 30/05/2025As many of you know, I’ve been fascinated by brain attractor dynamics lately. Thrilled to share a new preprint on their link to orthogonal neural representations, co-authored with Karl Friston: arxiv.org/abs/2505.22749 - with implications for both neuroscience & AI! First in a series - stay tuned!arxiv.orgSelf-orthogonalizing attractor neural networks emerging from the free energy principleAttractor dynamics are a hallmark of many complex systems, including the brain. Understanding how such self-organizing dynamics emerge from first principles is crucial for advancing our understanding ... 43010
Tamas Spisak @tspisak.bsky.social · 28/05/2025🚨 New paper out in GigaScience! To avoid common pitfalls in multivariate modeling: combine external validation with pre-registration — freeze your model before testing. For the pros: decide on the fly when to stop training! First-authored by the brilliant @ggallitto.bsky.social 163
