Sign in

cjl2007.bsky.social

@cjl2007.bsky.social
61 followers 55 following 45 posts
PostsRepliesMedia
Reposted by @cjl2007.bsky.social
Ignacio Saez, PhD @ignaciosaezphd.bsky.social · 04/09/2026
📣 We're hiring! The Saez Lab at Mount Sinai is seeking a postdoctoral researcher with experience in intracranial recordings, electrophysiology, BCIs, mood disorders, and/or computational modeling, to explore the neural basis of human cognition. 🗓️ Apply by September 30, 2026. Please share widely!
02223
Reposted by @cjl2007.bsky.social
Anne Billot, PhD @annebillot.bsky.social · 24/08/2026
📣 New study out today! We show that the brain's language network stays remarkably stable with age — same topography, lateralization, selectivity and connectivity as in younger adults — while the domain-general multiple demand network showed altered activity patterns. www.nature.com/articles/s41...
nature.com
Preserved topography, lateralization, selectivity, and functional connectivity of the language network in older brains - Nature Communications
Here, the authors show that aging affects brain networks differently, with neural changes mirroring the preservation or decline of the cognitive functions they support. The multiple demand network exh...
24216
Reposted by @cjl2007.bsky.social
Caterina Gratton @caterinagratton.bsky.social · 24/08/2026
🚨Job Search announcement 🚨 @psychillinois.bsky.social is searching for an assistant prof in cognitive neuroscience. Come join us! Pls RT 🙏 Full ad: illinois.csod.com/ux/ats/caree... #neuroskyence #psychscisky #neuroimaging
illinois.csod.com
Assistant Professor- Psychology - College of LAS
Duties & Responsibilities
03829
Reposted by @cjl2007.bsky.social
JAMA Psychiatry @jamapsychiatry.com · 13/08/2026
💬 Viewpoint: Longitudinal precision functional mapping combines individualized brain network mapping with repeated imaging to distinguish trait-like illness markers from state-like symptom changes in #Psychiatry. ja.ma/4qartja
A-D compare brain imaging for symptom dynamics & target specificity. Top row (A, B): Sparse/dense neuroimaging uses group map brains & U-shaped symptom severity graphs (T1 vs T1-T9 data). Bottom row (C, D): Sparse/dense PFM uses PFM brain models with similar U-shaped graphs.
051
Reposted by @cjl2007.bsky.social
Caterina Gratton @caterinagratton.bsky.social · 23/07/2026
A special highlight of our work on precision fMRI of the prefrontal cortex
0348
Reposted by @cjl2007.bsky.social
Leah Banellis @leahbanellis.bsky.social · 23/07/2026
Sensing your body (‘Interoception’) is everywhere in theories of mental health. But when we tested how well hundreds of people sensed their heart and lungs, the link to symptoms was almost nowhere to be found. 💣 🚨New @natmentalhealth.nature.com paper 🚨🎉 asks why 🧵 www.nature.com/articles/s44...
1021493
Reposted by @cjl2007.bsky.social
Rodrigo Braga @rodbraga.bsky.social · 14/07/2026
Now published in Journal of Neurophysiology: journals.physiology.org/doi/abs/10.1...
1216
Reposted by @cjl2007.bsky.social
Mike Frank @mcxfrank.bsky.social · 29/06/2026
Most of the statistical tests you learned in stats class — t-test, ANOVA, correlation — are actually special cases of a single thing: linear regression. Ch 7 of Experimentology argues that thinking in models, not tests, is more flexible and a better foundation for theory. 🧵 experimentology.io
330761
cjl2007.bsky.social @cjl2007.bsky.social · 23/06/2026
📩 How to apply Interested candidates should email application materials to Chuck Lynch: 📧 cjl2007@med.cornell.edu Please include: • Current CV • (Optional) brief statement of interest & relevant research background
000
cjl2007.bsky.social @cjl2007.bsky.social · 23/06/2026
Experience with TMS, TMS-fMRI, or other neuromodulation approaches (temporal interference, focused ultrasound, etc.) is a big plus. fMRI / precision mapping experience also highly valued. Please RT/share!
100
cjl2007.bsky.social @cjl2007.bsky.social · 23/06/2026
Postdocs will help lead dense longitudinal brain–mood mapping + personalized TMS interventions targeting circuits controlling mood state shifts. More details: postdocs.weill.cornell.edu/opportunitie...
postdocs.weill.cornell.edu
Postdoctoral Associate in Psychiatry | Office of Postdoctoral Affairs
As a Postdoctoral Associate, you will be working to advance projects using precision functional MRI (PFM) and targeted neuromodulation to study brain–symptom dynamics in unipolar and bipolar depressio...
100
cjl2007.bsky.social @cjl2007.bsky.social · 23/06/2026
Our lab at Weill Cornell is still looking for a Postdoctoral Fellow for BD²-funded projects studying mood dynamics in bipolar disorder w/ longitudinal precision fMRI and personalized TMS targeting. 📍 NYC | Full-time 🧵 Details below
110
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
@gordonneuro.bsky.social @zachladwig.bsky.social @mlueckel.bsky.social @jingnandu.bsky.social @logangrosenick.bsky.social @nilisolomonov.bsky.social + lots of others I can't find on bluesky!
000
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
Paper, code, and data are all below. The OSF project reproduces key analyses and figures in the paper. If you hit snags with the code or the repo, reach out, feedback very welcome. doi.org/10.1016/j.ne... · github.com/cjl2007/pfm-... · osf.io/qx2et
100
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
Grateful to the editors and reviewers at @cp-neuron.bsky.social for a fast, constructive process. And thank you again to my co-authors and collaborators, who made this possible, especially everyone who took time to complete the blinded dataset ratings.
100
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
Two reassurances though on that point. First, we see that NSI degrades as you remove BOLD signal or inject noise. Second, NSI correlates strongly with the independently developed "pBOLD" by Gonzalez-Castillo and colleagues, a template-agnostic measure of BOLD dominance from echo-time dependence.
100
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
Like any measure though, the NSI framework has limitations. One worth flagging: the NSI-usability mapping is anchored to expert judgments, and those judgments could carry a familiarity bias, where data resembling networks we already know gets rated higher (e.g., the mere-exposure effect).
100
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
To sum up: we think NSI offers an objective read on whether a dataset can support PFM, useful both for screening data you have and planning how much to collect. It won't replace eyeballs on data, but it makes that judgment more reproducible, complementing existing and emerging fMRI quality measures.
110
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
Caveat: FC reliability projection is an active area of development, and others are building more formal frameworks for it, so I see our NSI-based approach as a useful starting point, not the last word.
100
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
Both models above are validated in the paper: under nested leave-one-subject-out / leave-one-dataset-out CV, generalization to unseen subjects and sites held up well (usability AUC ≈ 0.92, reliability ≈ 0.87).
100
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
So from a single 10-minute pilot, project reliability at a chosen duration (30 min here) against a target level (R² = 0.7 here) pfm-nsi run --cifti Data.dtseries.nii --reliability --nsi-t 10 --query-t 30 --thresholds 0.7 You get P(R² ≥ 0.7) and a point estimate of R², each with a CI.
100
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
Why this works: we found that reliability accumulates faster in some datasets than others, and NSI tracks that growth rate. Early NSI predicts how quickly reliable FC builds up with additional minutes (see panel B and E from Fig. 6 below).
110
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
A second use: projecting FC reliability. From a short pilot (~10 min) in one or a handful of individuals, you can estimate the reliability you already have, or what you'd reach with more scanning, without collecting the hours of repeat data that split-half reliability normally requires.
110
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
When applied prospectively, you can scan the same individual under competing protocols (or run their data through different preprocessing / denoising strategies), compute NSI for each, and pick the best option before committing to a study.
100
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
So, for a given dataset: pfm-nsi run --cifti Data.dtseries.nii You get a median NSI and a calibrated P(usable), placed against the average expert-derived usability threshold and graded confidence bands (plus per-network and per-structure NSI histograms, if you want).
100
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
Why this works: experts judge usability by whether FC maps show coherent large-scale network structure, and NSI quantifies exactly that. It closely tracks blinded expert ratings and maps to a calibrated probability of usability (see panels C and F from Fig. 5 below).
110
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
A first use: screening datasets for PFM usability. You can take the NSI from an individual's dataset and get a calibrated estimate of whether it can support precision functional mapping.
100
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
The part I'm most excited about: the pfm-nsi code is now public, so you can try it on your own data. github.com/cjl2007/pfm-... (Python + MATLAB implementations). A couple of examples of what you can run, and where we hope it helps:
github.com
GitHub - cjl2007/pfm-nsi: An open-source tool for quality control of precision functional mapping (PFM) and individual-level functional connectivity MRI. It estimates whether a dataset contains suffic...
An open-source tool for quality control of precision functional mapping (PFM) and individual-level functional connectivity MRI. It estimates whether a dataset contains sufficient recoverable large-...
100
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
A few new visualizations since the preprint. For example, this crawling-seed movie samples FC from seeds throughout the brain, alongside each dataset's mean expert rating and NSI. (Cropped to the first 3 min for Bluesky; full 30-min version is online.)
100
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
Everything from the original thread still holds; linking it here for the full background if you haven't seen it. bsky.app/profile/cjl2...
100
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
To make these judgments more objective, reproducible, and standardized across datasets and labs, we introduce the Network Similarity Index (NSI), a measure of how strongly an individual fMRI dataset expresses coherent large-scale network structure.
110
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
In practice, we tend to make this call using indirect proxies like scan duration and head motion, or expert humans "eyeballing" the data (e.g., looking at FC seed maps).
100
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
To recap the problem: interest in precision fMRI & individual-level FC analyses more generally is growing rapidly, but perhaps starting to outpace our ability to say when a dataset can actually support it.
110
cjl2007.bsky.social @cjl2007.bsky.social · 22/06/2026
Happy to share that our paper on objective quality assessment for "precision" fMRI is now out in @cp-neuron.bsky.social A recap of what it does, what's new since the preprint, and code you can run on your own data. cell.com/neuron/fullt...
27622
Reposted by @cjl2007.bsky.social
Caterina Gratton @caterinagratton.bsky.social · 11/06/2026
How can we acquire new knowledge and skills on a daily basis while remaining the same person? I'm excited to share our new preprint led by @hyejinjadelee.bsky.social & Ally Dworetsky tackling this question, just in time for #OHBM2026: doi.org/10.64898/202... 🧵👇 #neuroskyence #PsychSciSky
doi.org
Functional brain organization is stable within individuals across years
Brain regions exhibit dynamic yet highly coordinated activity patterns that form large-scale functional networks measurable through resting-state correlations. While their association with fluctuating...
15920
Reposted by @cjl2007.bsky.social
Evan Gordon @gordonneuro.bsky.social · 11/06/2026
This is one of the most socially important scientific findings I've ever been a part of. SES has a larger cross-sectional effect on brain function than any other variable. Prediction of IQ from brain data is actually mostly just predicting SES. Mind blowing.
1238
cjl2007.bsky.social @cjl2007.bsky.social · 02/06/2026
Very cool, looking forward to trying this out
000
Reposted by @cjl2007.bsky.social
Ignacio Saez, PhD @ignaciosaezphd.bsky.social · 01/06/2026
How does your brain filter out the noise and focus enough to learn efficiently? In our newest preprint, we use intracranial recordings + computational modeling to explore how selective attention shapes learning and state representations in the human brain 🧠 www.biorxiv.org/content/10.6... 1/🧵
bioRxiv preprint titled 'Coordinated human prefrontal dynamics sustain task-state representations during learning'
24115
Reposted by @cjl2007.bsky.social
Rodrigo Braga @rodbraga.bsky.social · 29/05/2026
We are hiring an Associate or Full Professor to lead the neuroimaging research program at Northwestern's Center for Translational Imaging in the Department of Radiology careers.northwestern.edu/psc/hrnu_er/...
077
Reposted by @cjl2007.bsky.social
Caterina Gratton @caterinagratton.bsky.social · 18/05/2026
Very excited about this project, led by @zachladwig.bsky.social We show that an individual’s lateral prefrontal cortex has reliable and highly detailed network organization missed in past group approaches. Full🧵 coming soon!
16123
Reposted by @cjl2007.bsky.social
Zach Ladwig @zachladwig.bsky.social · 19/05/2026
(1/5) Our paper is out now in Neuron! bsky.app/profile/cp-n... The LPFC has fine-scale organization that gets missed in group-averaged data because of inter-individual variability. It is visible in individuals in both resting-state and task data.
311531
Reposted by @cjl2007.bsky.social
Evan Gordon @gordonneuro.bsky.social · 22/04/2026
I always assumed that brain function had to line up with cytoarchitectonics. It turns out I was wrong. Human cortex, especially PFC, is tiled by chains of functional patches that subdivide and interlink architectonic areas into parallel processing streams. www.biorxiv.org/content/10.6...
14234122
Reposted by @cjl2007.bsky.social
Dan Handwerker @danielhandwerker.bsky.social · 07/04/2026
tedana, multi-echo fMRI denoising software, newsletter includes tedana v26.0.3 updates and details of multi-echo fMRI ed course at OHBM 2026. groups.google.com/g/tedana-new...
groups.google.com
Multi-echo fMRI course at OHBM 2026 and tedana v26.0.3
094
Reposted by @cjl2007.bsky.social
Rodrigo Braga @rodbraga.bsky.social · 31/03/2026
Our new paper on brain networks engaged during imagining is out now in Neuron! Here is a download link (free for 50 days): authors.elsevier.com/c/1msNE3BtfH... Congratulations to Nate Anderson for leading this work @rementurus.bsky.social 🧵
38639
cjl2007.bsky.social @cjl2007.bsky.social · 12/02/2026
This work was a true team effort. Grateful to my co-authors and collaborators who made this possible. @jingnandu.bsky.social @gordonneuro.bsky.social @mlueckel.bsky.social + others not on bsky
030
cjl2007.bsky.social @cjl2007.bsky.social · 12/02/2026
Preprint link 👇 biorxiv.org/content/10.6... Feedback welcome — especially from investigators currently applying or planning to apply PFM in new domains, clinical populations, or translational contexts
biorxiv.org
120
cjl2007.bsky.social @cjl2007.bsky.social · 12/02/2026
Collectively, NSI supports two immediate uses: 1️⃣ Screening datasets for PFM suitability 2️⃣ Planning acquisition based on expected reliability gains NSI evaluates intrinsic network structure in the data itself — supporting transparent, reproducible precision fMRI.
130
cjl2007.bsky.social @cjl2007.bsky.social · 12/02/2026
NSI also explains a familiar puzzle: at matched scan durations, individuals show strikingly different reliability trajectories. Higher NSI → faster reliability accumulation and higher achievable reliability. Even a brief “scout” scan NSI predicts how much added scanning will improve reliability.
120
cjl2007.bsky.social @cjl2007.bsky.social · 12/02/2026
Experts: • Inspected FC maps • Assigned continuous quality scores • Made binary PFM usability decisions NSI closely tracked these judgments — outperforming scan duration. We mapped NSI to usability probability to provide a calibrated, prospective decision tool.
120