I am thrilled to share that our paper introducing iSTTC has just been published in PLOS Computational Biology! This work was led by Irina Pochinok, who drove the project from start to finish with great dedication and rigor. Congratulations Irina, well deserved!

For those unfamiliar with iSTTC, here is the short version: intrinsic neural timescales reflect how neural networks integrate information over time. They are short in brain areas specialized in rapid sensory reactions, and longer in those devoted to complex decision-making. I find them fascinating because they are a promising bridge between low-level mechanisms — single-neuron or circuit properties — and high-level cognitive phenomena. And they are remarkably robust: replicated across species as different as worms and humans, and across recording modalities from single-unit activity to fMRI.

The problem is that estimating timescales reliably from sparse or epoched spike data has always been tricky with existing methods. iSTTC improves this: it provides accurate estimates even under challenging conditions, works on both continuous and trial-based data, and substantially relaxes inclusion criteria — meaning you can include many more neurons in your analyses. As a bonus, we discovered that the common practice of estimating timescales from short, epoched recordings leads to errors roughly ten times larger than those from long, uninterrupted recordings. Not our original goal, but a finding we think the timescales community will care about.

I am particularly excited because our lab works a lot with early brain development, where spiking data is notoriously sparse — exactly the regime where iSTTC shines. We built this tool with those datasets in mind, and I cannot wait to start applying it.

If you want to give it a try yourself, the code is freely available on GitHub. We hope many of you will!

A heartfelt thank you to the editors and reviewers, whose feedback genuinely pushed the manuscript to a higher level. One reviewer, in a fun twist, is no longer anonymous: I ran into him (and he presented himself) at a bar during COSYNE this year!

Finally, a well-deserved (if slightly ironic) acknowledgment to Piet Mondrian for the inspiration behind the iSTTC logo. We like to think he would have approved, even though that’s probably highly unlikely.