iiinova — where your digital self keeps evolving

Jayson Jeganathan @jaysonjeg.bsky.social · 25/02/2025

On native meshes, gyral/sulcal differences were minimal or non-existent. In general, one good test would be to see if your outcomes (like myelination) co-varied with sulcal depth across the whole brain. If they do, then need to consider whether the relationship is artefactual or real.

020

Jayson Jeganathan @jaysonjeg.bsky.social · 25/02/2025

I'm proud of this work, and I hope this paper helps you to avoid spurious conclusions in your work. Check out the full paper (tinyurl.com/gyral) for a detailed breakdown and for other downstream analyses that can be impacted (FC fingerprinting, hyperalignment, etc). (5/5)

tinyurl.com

Spurious correlations in surface-based functional brain imaging

Abstract. The study of functional MRI (fMRI) data is increasingly performed after mapping from volumetric voxels to surface vertices. Processing pipelines commonly used to achieve this mapping produce...

0150

Jayson Jeganathan @jaysonjeg.bsky.social · 25/02/2025

The new Onavg template promises to even out variability in inter-vertex spacing across the cortex. We tested this. This template reduces variability by 90% in the common surface space, but only by about 20% in subject-specific surfaces :( (4/5)

170

Jayson Jeganathan @jaysonjeg.bsky.social · 25/02/2025

We then explored the consequences. Adjacent sulcal vertices have highly correlated fMRI time series just because they're closer. This can trick functional parcellations into putting parcel boundaries on gyri instead. (3/5)

140

Jayson Jeganathan @jaysonjeg.bsky.social · 25/02/2025

This new visual by @phogat_richa shows it all. In this single-subject flattened cortex, we see mesh vertices and triangles with gyri (red) and sulci (blue). The triangles are clearly smaller in sulci (blue). All fsaverage and fsLR surfaces look like this. (2/5)

282

Jayson Jeganathan @jaysonjeg.bsky.social · 25/02/2025

Do you work on surface MRI such as HCP data? Our paper is finally out in Imaging Neuroscience, showing biases in surface fMRI. The revised ms has many new insights! TLDR: Vertices are much closer to each other in sulci, resulting in (fake) high spatial autocorrelation. (1/5)

15322

Jayson Jeganathan @jaysonjeg.bsky.social · 21/12/2024

Wow. Most recent openAI model o3 smashes the prev leaderboard for ARC reasoning challenge (an AGI benchmark). It is also now the 175th best coder in the world (CodeForces challenge), and can solve 25% of hard-for-humans FrontierMath Qs (up from 2%). What does this all mean?

120

Jayson Jeganathan @jaysonjeg.bsky.social · 13/12/2023

A new podcast episode about a very persuasive theory of consciousness (from my favourite podcast host!) #consci #neuroscience open.spotify.com/episode/0ldz...

000

Reposted by Jayson Jeganathan

Carl T. Bergstrom @carlbergstrom.com · 13/10/2023

1. Imagine we land a space probe on one of Jupiters’ moons, take up a sample of material, and find it is full of organic molecules. How can we tell whether those molecules are just randomly assembled goo or the outcome of some evolutionary process taking place there? 🧪

20419169

Jayson Jeganathan @jaysonjeg.bsky.social · 25/09/2023

Hi! I'm a psychiatry trainee and PhD student. Looking forward to engaging in conversations on human brain imaging, neuroscience, consciousness, and psychiatry! #HiSciSky #compneuro #neuroimaging #consci #interoception

2203