hakwan lau
@hakwan.bsky.social
neuroscientist in Korea (co-director of IBS-CNIR) interested in how neuroimaging (e.g. fMRI or widefield optical imaging) can facilitate closed-loop causal interventions (e.g. neurofeedback, patterned stimulations). https://tinyurl.com/hakwan
Pinned
www.cell.com
a while back i threatened to share this. finally online
for detection tasks we often systematically estimates sensitivity wrong. we need to control for unequal variance in models, but we often don't coz it needs extra data
now there's a virtually 'free' way to do it
www.cell.com/iscience/pdf...
for detection tasks we often systematically estimates sensitivity wrong. we need to control for unequal variance in models, but we often don't coz it needs extra data
now there's a virtually 'free' way to do it
www.cell.com/iscience/pdf...
Reposted by hakwan lau
NSD-synthetic, the out-of-distribution companion dataset of NSD consisting of 7T fMRI responses to 284 artificial images, is now published.
#NeuroAI #CompNeuro #neuroscience #AI
doi.org/10.1038/s414...
#NeuroAI #CompNeuro #neuroscience #AI
doi.org/10.1038/s414...
February 12, 2026 at 2:46 PM
NSD-synthetic, the out-of-distribution companion dataset of NSD consisting of 7T fMRI responses to 284 artificial images, is now published.
#NeuroAI #CompNeuro #neuroscience #AI
doi.org/10.1038/s414...
#NeuroAI #CompNeuro #neuroscience #AI
doi.org/10.1038/s414...
Reposted by hakwan lau
February 13, 2026 at 4:06 AM
a while back i threatened to share this. finally online
for detection tasks we often systematically estimates sensitivity wrong. we need to control for unequal variance in models, but we often don't coz it needs extra data
now there's a virtually 'free' way to do it
www.cell.com/iscience/pdf...
for detection tasks we often systematically estimates sensitivity wrong. we need to control for unequal variance in models, but we often don't coz it needs extra data
now there's a virtually 'free' way to do it
www.cell.com/iscience/pdf...
www.cell.com
February 13, 2026 at 3:39 AM
a while back i threatened to share this. finally online
for detection tasks we often systematically estimates sensitivity wrong. we need to control for unequal variance in models, but we often don't coz it needs extra data
now there's a virtually 'free' way to do it
www.cell.com/iscience/pdf...
for detection tasks we often systematically estimates sensitivity wrong. we need to control for unequal variance in models, but we often don't coz it needs extra data
now there's a virtually 'free' way to do it
www.cell.com/iscience/pdf...
so Ned Block just wrote an excellent review on a book, by a journalist, on consciousness, partly based on interviews w/ "gurus, novelists, and cultural figures", w/ some "eccentric arguments" & views
www.science.org/doi/10.1126/...
i suppose Ned was invited to do it. but why should Science care??
www.science.org/doi/10.1126/...
i suppose Ned was invited to do it. but why should Science care??
The elusive nature of consciousness
A writer grapples with neuroscience’s hardest problem
www.science.org
February 12, 2026 at 11:59 PM
so Ned Block just wrote an excellent review on a book, by a journalist, on consciousness, partly based on interviews w/ "gurus, novelists, and cultural figures", w/ some "eccentric arguments" & views
www.science.org/doi/10.1126/...
i suppose Ned was invited to do it. but why should Science care??
www.science.org/doi/10.1126/...
i suppose Ned was invited to do it. but why should Science care??
Reposted by hakwan lau
Excited to share our new preprint! A massive (~400k trials) multi-site effort to test "subjective inflation," the influential but under-tested idea that subjective experience in the unattended periphery can be inflated beyond what objective performance would suggest
www.biorxiv.org/content/10.1...
www.biorxiv.org/content/10.1...
www.biorxiv.org
February 12, 2026 at 4:05 PM
Excited to share our new preprint! A massive (~400k trials) multi-site effort to test "subjective inflation," the influential but under-tested idea that subjective experience in the unattended periphery can be inflated beyond what objective performance would suggest
www.biorxiv.org/content/10.1...
www.biorxiv.org/content/10.1...
Reposted by hakwan lau
Individual differences in artificial neural networks capture individual differences in human behavior https://www.biorxiv.org/content/10.64898/2026.02.10.705061v1
February 12, 2026 at 5:15 AM
Individual differences in artificial neural networks capture individual differences in human behavior https://www.biorxiv.org/content/10.64898/2026.02.10.705061v1
Reposted by hakwan lau
Thrilled that my paper is out in the @nature.com. We explored how the brain builds complex tasks by compositionally combining simpler sub-task representations. The brain flexibly performs multiple tasks by dynamically reusing neural subspaces for sensory inputs and motor actions
rdcu.be/eRVUk
rdcu.be/eRVUk
Building compositional tasks with shared neural subspaces
Nature - The brain can flexibly perform multiple tasks by compositionally combining task-relevant neural representations.
rdcu.be
February 11, 2026 at 10:40 PM
Thrilled that my paper is out in the @nature.com. We explored how the brain builds complex tasks by compositionally combining simpler sub-task representations. The brain flexibly performs multiple tasks by dynamically reusing neural subspaces for sensory inputs and motor actions
rdcu.be/eRVUk
rdcu.be/eRVUk
Reposted by hakwan lau
My first paper is out in #SociologicalScience!
With Jörg Stolz and Ruud Luijkx, we found robust evidence of ideological #bias in #secularization research: researchers' own religiosity is correlated with their probability of finding evidence of religious decline in their publications.
Read more: 👇
With Jörg Stolz and Ruud Luijkx, we found robust evidence of ideological #bias in #secularization research: researchers' own religiosity is correlated with their probability of finding evidence of religious decline in their publications.
Read more: 👇
NEW: Valeria Rainero, Jörg Stolz, Ruud Luijkx, The Faith Factor. How Scholars’ Religiosity Biases Research Findings on Secularization sociologicalscience.com/articles-v13...
sociologicalscience.com
February 11, 2026 at 10:22 AM
My first paper is out in #SociologicalScience!
With Jörg Stolz and Ruud Luijkx, we found robust evidence of ideological #bias in #secularization research: researchers' own religiosity is correlated with their probability of finding evidence of religious decline in their publications.
Read more: 👇
With Jörg Stolz and Ruud Luijkx, we found robust evidence of ideological #bias in #secularization research: researchers' own religiosity is correlated with their probability of finding evidence of religious decline in their publications.
Read more: 👇
Reposted by hakwan lau
LLMs good performance on medical exams does not translate to accurate performance in real-world settings (preregistered n~1,300 study). This can't be explained by current standard benchmarks for medical knowledge & simulated patient interactions.
www.nature.com/articles/s41...
www.nature.com/articles/s41...
Reliability of LLMs as medical assistants for the general public: a randomized preregistered study - Nature Medicine
In a randomized controlled study involving 1,298 participants from a general sample, performance of humans when assisted by a large language model (LLM) was sensibly inferior to that of the LLM alone ...
www.nature.com
February 11, 2026 at 10:30 AM
LLMs good performance on medical exams does not translate to accurate performance in real-world settings (preregistered n~1,300 study). This can't be explained by current standard benchmarks for medical knowledge & simulated patient interactions.
www.nature.com/articles/s41...
www.nature.com/articles/s41...
as a neurotechnologist, i enjoy working on phenotype probes.
my specialization in neuroscience is the development of behavioral assays. although i sometimes do some neuroimaging on the side too.
my specialization in neuroscience is the development of behavioral assays. although i sometimes do some neuroimaging on the side too.
February 10, 2026 at 10:46 PM
as a neurotechnologist, i enjoy working on phenotype probes.
my specialization in neuroscience is the development of behavioral assays. although i sometimes do some neuroimaging on the side too.
my specialization in neuroscience is the development of behavioral assays. although i sometimes do some neuroimaging on the side too.
if an animal fails visual metacognition (e.g. can't rate confidence meaningfully) & yet can do so in other sensory modalities, does it likely lack subjective visual experiences?
i.e. if the animal has normal visual experiences, & generally good metacog, can it fail visual metacog *somehow*? how?
i.e. if the animal has normal visual experiences, & generally good metacog, can it fail visual metacog *somehow*? how?
February 10, 2026 at 10:38 PM
if an animal fails visual metacognition (e.g. can't rate confidence meaningfully) & yet can do so in other sensory modalities, does it likely lack subjective visual experiences?
i.e. if the animal has normal visual experiences, & generally good metacog, can it fail visual metacog *somehow*? how?
i.e. if the animal has normal visual experiences, & generally good metacog, can it fail visual metacog *somehow*? how?
Reposted by hakwan lau
Here is the preprint of @kjedlovszky.bsky.social work in the lab: osf.io/preprints/ps...: experiencing a volatile task (both social and non-social) increases social hallucinations of chasing in moving dots - filling in the mechanistic gaps of how volatility beliefs influence perception
February 10, 2026 at 9:32 PM
Here is the preprint of @kjedlovszky.bsky.social work in the lab: osf.io/preprints/ps...: experiencing a volatile task (both social and non-social) increases social hallucinations of chasing in moving dots - filling in the mechanistic gaps of how volatility beliefs influence perception
Reposted by hakwan lau
Our paper is out in @natneuro.nature.com!
www.nature.com/articles/s41...
We develop a geometric theory of how neural populations support generalization across many tasks.
@zuckermanbrain.bsky.social
@flatironinstitute.org
@kempnerinstitute.bsky.social
1/14
www.nature.com/articles/s41...
We develop a geometric theory of how neural populations support generalization across many tasks.
@zuckermanbrain.bsky.social
@flatironinstitute.org
@kempnerinstitute.bsky.social
1/14
February 10, 2026 at 3:56 PM
Our paper is out in @natneuro.nature.com!
www.nature.com/articles/s41...
We develop a geometric theory of how neural populations support generalization across many tasks.
@zuckermanbrain.bsky.social
@flatironinstitute.org
@kempnerinstitute.bsky.social
1/14
www.nature.com/articles/s41...
We develop a geometric theory of how neural populations support generalization across many tasks.
@zuckermanbrain.bsky.social
@flatironinstitute.org
@kempnerinstitute.bsky.social
1/14
Reposted by hakwan lau
This is finally out in NeuroInformatics "Enhancing fMRI Decoded Neurofeedback with Co-adaptive Training: Simulation and Proof-of-principle Evidence" shorturl.at/XCwve and the updated PyDecNef repo github.com/pydecnef/Pyd...
@hakwan.bsky.social @meganakpeters.bsky.social @neuroaure.bsky.social
@hakwan.bsky.social @meganakpeters.bsky.social @neuroaure.bsky.social
Enhancing fMRI Decoded Neurofeedback with Co-adaptive Training: Simulation and Proof-of-principle Evidence - Neuroinformatics
A significant challenge for neurofeedback training research and related clinical applications, is participants’ difficulty in learning to induce specific brain patterns during training. Here, we addre...
link.springer.com
February 10, 2026 at 8:47 AM
This is finally out in NeuroInformatics "Enhancing fMRI Decoded Neurofeedback with Co-adaptive Training: Simulation and Proof-of-principle Evidence" shorturl.at/XCwve and the updated PyDecNef repo github.com/pydecnef/Pyd...
@hakwan.bsky.social @meganakpeters.bsky.social @neuroaure.bsky.social
@hakwan.bsky.social @meganakpeters.bsky.social @neuroaure.bsky.social
Reposted by hakwan lau
A side note in metas court room drama is this little memo from 2018 discussing the goal of converting ten “academic allies” given that researchers were associating harms to their products.
February 9, 2026 at 11:13 PM
A side note in metas court room drama is this little memo from 2018 discussing the goal of converting ten “academic allies” given that researchers were associating harms to their products.
Reposted by hakwan lau
neuroAI comparisons of ANNs to brains do have a range of problems. Even more than I had realized. And I was worried before: www.biorxiv.org/content/10.1...
www.biorxiv.org
February 9, 2026 at 2:13 PM
neuroAI comparisons of ANNs to brains do have a range of problems. Even more than I had realized. And I was worried before: www.biorxiv.org/content/10.1...
Reposted by hakwan lau
Our latest paper, “Visual language models show widespread visual deficits on neuropsychological tests”, is now out in Nature Machine Intelligence: www.nature.com/articles/s42...
Non-paywalled version:
arxiv.org/abs/2504.10786
Tweet thread below from first author @genetang.bsky.social...
Non-paywalled version:
arxiv.org/abs/2504.10786
Tweet thread below from first author @genetang.bsky.social...
Visual language models show widespread visual deficits on neuropsychological tests - Nature Machine Intelligence
Tangtartharakul and Storrs use standardized neuropsychological tests to compare human visual abilities with those of visual language models (VLMs). They report that while VLMs excel in high-level obje...
www.nature.com
February 9, 2026 at 2:40 AM
Our latest paper, “Visual language models show widespread visual deficits on neuropsychological tests”, is now out in Nature Machine Intelligence: www.nature.com/articles/s42...
Non-paywalled version:
arxiv.org/abs/2504.10786
Tweet thread below from first author @genetang.bsky.social...
Non-paywalled version:
arxiv.org/abs/2504.10786
Tweet thread below from first author @genetang.bsky.social...
Reposted by hakwan lau
🚨Job alert! I'm recruiting a postdoc! If you want to study the time course of task-driven visual perception, please reach out! #neuroskyence #VisionScience #CogSci barnard.wd1.myworkdayjobs.com/en-US/Facult...
Postdoctoral Research Fellow, Cognitive/Computational Neuroscience
If you are a current Barnard College employee, please use the internal career site to apply for this position. Job: Postdoctoral Research Fellow, Cognitive/Computational Neuroscience The Barnard Visua...
barnard.wd1.myworkdayjobs.com
February 7, 2026 at 8:13 PM
🚨Job alert! I'm recruiting a postdoc! If you want to study the time course of task-driven visual perception, please reach out! #neuroskyence #VisionScience #CogSci barnard.wd1.myworkdayjobs.com/en-US/Facult...
from the docs now available online, it may seem like Epstein was especially interested in consciousness.
but i don't think that's true. rather, perhaps consciousness science was especially interested in Epstein.
but i don't think that's true. rather, perhaps consciousness science was especially interested in Epstein.
February 7, 2026 at 9:33 AM
from the docs now available online, it may seem like Epstein was especially interested in consciousness.
but i don't think that's true. rather, perhaps consciousness science was especially interested in Epstein.
but i don't think that's true. rather, perhaps consciousness science was especially interested in Epstein.
Reposted by hakwan lau
🧠📈A new foundation model for #neuroscience! In the latest Deeper Learning blog, @duranrin.bsky.social and @kanakarajanphd.bsky.social describe their state-of-the-art POCO model, which accurately forecasts neural dynamics across individuals and species.
bit.ly/4afqx5A #NeuroAI #neuroscience
bit.ly/4afqx5A #NeuroAI #neuroscience
Forecasting the Brain: Scalable Neural Prediction with POCO - Kempner Institute
Predicting future neural activity is a critical step toward achieving real-time, closed-loop neurotechnologies. To this end, we introduce POCO, a unified forecasting model trained on diverse calcium i...
bit.ly
February 6, 2026 at 5:36 PM
🧠📈A new foundation model for #neuroscience! In the latest Deeper Learning blog, @duranrin.bsky.social and @kanakarajanphd.bsky.social describe their state-of-the-art POCO model, which accurately forecasts neural dynamics across individuals and species.
bit.ly/4afqx5A #NeuroAI #neuroscience
bit.ly/4afqx5A #NeuroAI #neuroscience
tenure in 4 years!! this is so ridiculous. i'll post something about this tomorrow.....
I have news! After 4 fabulous years at Northeastern, this July I will be moving to Duke—with tenure! It’s hard to convey how grateful I am to everyone who has made this possible: from old professors in Mexico and mentors in the US to students, colleagues, and, of course, my amazing wife and family.
February 6, 2026 at 7:47 PM
tenure in 4 years!! this is so ridiculous. i'll post something about this tomorrow.....
it may be difficult to contemplate the significance of this
February 6, 2026 at 7:44 PM
it may be difficult to contemplate the significance of this
Reposted by hakwan lau
🧪
#publichealth
#neuroskyence
Schizophrenia is among the most refractory and devastating of all brain diseases. It makes me furious to hear this sorry ass excuse of an HHS secretary peddle this bullshit. False hope and misdirected resources. Eejit.
www.nytimes.com/2026/02/05/u...
#publichealth
#neuroskyence
Schizophrenia is among the most refractory and devastating of all brain diseases. It makes me furious to hear this sorry ass excuse of an HHS secretary peddle this bullshit. False hope and misdirected resources. Eejit.
www.nytimes.com/2026/02/05/u...
Kennedy Makes Unfounded Claim That Keto Diet Can ‘Cure’ Schizophrenia
www.nytimes.com
February 6, 2026 at 2:36 AM
🧪
#publichealth
#neuroskyence
Schizophrenia is among the most refractory and devastating of all brain diseases. It makes me furious to hear this sorry ass excuse of an HHS secretary peddle this bullshit. False hope and misdirected resources. Eejit.
www.nytimes.com/2026/02/05/u...
#publichealth
#neuroskyence
Schizophrenia is among the most refractory and devastating of all brain diseases. It makes me furious to hear this sorry ass excuse of an HHS secretary peddle this bullshit. False hope and misdirected resources. Eejit.
www.nytimes.com/2026/02/05/u...
Reposted by hakwan lau
To all the non-academics who don't understand the point of academic freedom and don't care to learn about it, let me give you a factor you might care about:
Tenure is the biggest reason you get my services at a steep discount over my private sector value.
Tenure is the biggest reason you get my services at a steep discount over my private sector value.
No Tenure for Oklahoma.
Instead, faculty will be on "renewable contracts “tied to teaching effectiveness, student completion, job placement, and economic alignment.”
#academicsky
Instead, faculty will be on "renewable contracts “tied to teaching effectiveness, student completion, job placement, and economic alignment.”
#academicsky
Tenure Will Be Eliminated at Most of Oklahoma’s Public Colleges, Governor Says
Gov. J. Kevin Stitt, a Republican, directed the state’s two-dozen regional universities and community colleges to phase out the practice. Existing faculty members will be grandfathered in.
www.chronicle.com
February 6, 2026 at 3:16 AM
To all the non-academics who don't understand the point of academic freedom and don't care to learn about it, let me give you a factor you might care about:
Tenure is the biggest reason you get my services at a steep discount over my private sector value.
Tenure is the biggest reason you get my services at a steep discount over my private sector value.
Reposted by hakwan lau
I hear imagination in nonhuman animals is making news today! www.sciencedirect.com/science/arti...
Event representation in Pavlovian conditioning: Image and action
In a typical Pavlovian conditioning experiment, a relatively insignificant event, the conditioned stimulus (CS), is paired with a biologically more me…
www.sciencedirect.com
February 5, 2026 at 10:46 PM
I hear imagination in nonhuman animals is making news today! www.sciencedirect.com/science/arti...