I'm more and more convinced that low-dimensional manifolds in the brain are just an artifact of the experimental designs and analyses we use... 🧠📈 🧪

Great new paper by my colleagues Gloria & Alex, and @lichengzou.bsky.social. Hebbian plasticity accounts for much of representational drift: Drift is ongoing memory storage, not noise. Also reconciles contradictory findings on stability after repeated exposure. www.nature.com/articles/s41... 🧠🧪

BTSP, but not STDP, can account for place field changes in hippocampus, out in Nature Neuroscience today: www.nature.com/articles/s41... 🧠📈 🧪

Please spread the word!! We are accepting applications for the Cold Spring Harbor Asia Computational and Cognitive Summer School in Suzhou, China. Dates: June 29 to July 20 of 2025. Application deadline: March 31. #computationalneuroscience. For the top three reasons to attend, see thread below:

Most neuroscience research is exploratory: they search for *something* rather than testing a hypothesis. It's harking. I used to see that as a problem, but not anymore. The 1st papers generate hypotheses; the real testing comes later. Science only make sense collectively, not individually.

Our new preprint: We found that during representational drift co-activation of neuron pairs is predictive of future effective connectivity. Drift is - in parts - just Hebbian plasticity. But we also need a stochastic process to account for our observations. doi.org/10.1101/2025...

Sometimes I think the most important thing a computational neuroscientist can do to encourage acceptance of their work in the broader experimental community is just to plot their model data in the exact same way the relevant experimental data was plotted. It's very affecting!

𝗧𝗵𝗲 𝗺𝗼𝘀𝘁 𝗯𝗮𝘀𝗶𝗰 𝗶𝗻𝘁𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻 𝘁𝗼 𝗳𝗠𝗥𝗜 𝗻𝗲𝘁𝘄𝗼𝗿𝗸𝘀. Please, absolutely no equations, think of someone math phobic. Much better if a paper. thanks #neuroscience