🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Get higher SNR in single-trial fMRI. We show substantial impact on downstream analyses. Pre-print: biorxiv.org/content/10.110…. Toolbox (Matlab and Python): glmsingle.org. Led by Jacob Prince and w/ Ian Charest Jan Kurzawski John Pyles Academic Mercenary

You can now read our latest work on bioRxiv: Color-biased regions in the ventral visual pathway are food-selective @sussexvisionlab Sussex Colour Group cvnlab biorxiv.org/content/10.110…

Latest lab work in PLOS One. Denoising methods come with risks. A reminder of the classic bias-variance tradeoff, and to acknowledge, study, and consider bias. journals.plos.org/plosone/articl…

Excited to share our new preprint! "Re-expression of CA1 and entorhinal activity patterns preserves temporal context memory at long timescales" w/ the amazing Sarah DuBrow Hutchinson Lab Brice Kuhl and many others from the NSD crew cvnlab biorxiv.org/cgi/content/sh…

Latest lab work in #JNeurosci. A detailed look at variation of BOLD fMRI signal strength within individual brains, and whether this is a good proxy for local neural activity. w/ Jan Kurzawski Jon Winawer Faruk Gulban jneurosci.org/content/42/38/…

Preprint! How does the brain make sense of the world? Using a variety of converging analyses and recurrent DNN modeling of large-scale 7T data, we propose that the visual system computes rich semantic scene descriptions, rather than category labels only.🧵 arxiv.org/abs/2209.11737

I’m pleased to share a pre-print about activity in the visual word form area! @jdyeatman, cvnlab and I found that words evoke larger responses when task-relevant than when ignored, but the opposite happens for non-letter shapes. biorxiv.org/content/10.110… (1/2)

We've posted this working draft on OSF arguing for the need of further validation of layer-specific fMRI as an accurate and robust marker of neural activity, but we are still refining the text. We welcome comments from the community.

Latest lab work in Nature Communications. We leverage the large-scale Natural Scenes Dataset and encoding models to obtain unbiased estimates of voxel-level orientation tuning. Results demonstrate coarse-scale orientation preferences in V1. w/ Zvi Roth eli merriam nature.com/articles/s4146…

Latest lab work in NeuroImage by Faruk Gulban. We release high-quality mesoscopic (0.35-mm) anatomical measurements in the living human brain. Clear laminar and vascular features. May help us better understand the nature of fMRI signals. doi.org/10.1016/j.neur…

You can now watch videos of all main stage sessions from #ccneuro22 at 2022.ccneuro.org/program.php! Hope that you enjoy reliving this year's conference - and stay tuned for announcements about next year's conference soon. #ccneuro23 is coming...

Latest lab work in Neuron. We review tasks, how tasks affect activity in the visual system, and call for incorporation of tasks into stimulus-computable models. w/ Rachel Denison David L Barack @MichaelJArcaro Kate Bonnen doi.org/10.1016/j.neur…

There is much discussion about capabilities of GPT models. But how accurate are such models as models of language processing in the 🧠? Here, we demonstrate that GPT is accurate enough to noninvasively drive and suppress 🧠activity in the language network of new individuals. 1/

cvnlab is looking to hire postdoc(s) in cognitive/visual/computational/systems neuroscience. More info at cvnlab.net/jobad.html

1/ Excited to share a new NeuroAI preprint: “What can 1.8 billion regressions tell us about the pressures shaping high-level visual representation in brains and machines?” Led by Colin Conwell, with Jacob Prince, cvnlab, and George Alvarez. biorxiv.org/content/10.110…

Pumped to share this work from my PhD, “A single computational objective drives specialization of streams in visual cortex”, just in time for the holidays! With the all star team Eshed Margalit , cvnlab , Daniel Yamins & kalatwit 🔗:biorxiv.org/content/10.110… 🧵: 1/n…