Lightnews — Scholar-powered news

Reposted by Martin P. Paulus

Kavya Devarakonda, PhD @kdevarakonda.bsky.social · 29d

Excited to share that my PhD work has been published! www.nature.com/articles/s41... 🧠🧪

Amygdala–liver signalling orchestrates glycaemic responses to stress

Nature - Studies in mice show that acute stress activates hyperglycaemia via activation of a medial amygdala–ventral hypothalamic circuit that controls glucose metabolic responses in the...

www.nature.com

1 8 31

Reposted by Martin P. Paulus

Olu Ajilore, MD, PhD @pshrink.bsky.social · Aug 1

www.nytimes.com/2025/08/01/b...

Corporation for Public Broadcasting Will Shut Down

www.nytimes.com

2 1

Martin P. Paulus @mpwpaulus.bsky.social · Jun 11

Now available at BP CNNI, my piece with Murray about Difficult to Treat Anxiety: www.sciencedirect.com/science/arti...

Difficult-to-Treat Anxiety: A Neurocomputational Framework

Anxiety disorders, affecting approximately 1 in 9 individuals globally, impose significant socioeconomic and health burdens, with many individuals fai…

www.sciencedirect.com

1 1

Martin P. Paulus @mpwpaulus.bsky.social · Apr 23

Great work by Hamed to pull this together.

Neuropsychopharmacology @npp-journal.bsky.social · Apr 23

How can we individualize non-invasive brain stimulation protocols to improve treatment outcomes?

This review article highlights 4 potential opportunities, including context, target, dose, and timing

www.nature.com/artic...

Four dimensions of individualization in brain stimulation for psychiatric disorders: context, target, dose, and timing

Neuropsychopharmacology - Four dimensions of individualization in brain stimulation for psychiatric disorders: context, target, dose, and timing

www.nature.com

1 3

Martin P. Paulus @mpwpaulus.bsky.social · Apr 3

And you can listen to the generated podcast here: notebooklm.google.com/notebook/94f... #DTA #MentalHealthResearch #BrainScience

Sign in - Google Accounts

notebooklm.google.com

Martin P. Paulus @mpwpaulus.bsky.social · Apr 3

A labor of love with my friend & long-time collaborator Murray B. Stein. Addressing Difficult-to-Treat Anxiety: A review integrating computational psychiatry, systems neuroscience, & novel interventions. Mechanistic insights for precision psychiatry. www.biologicalpsychiatrycnni.org/article/S245...

1 1 4

Martin P. Paulus @mpwpaulus.bsky.social · Apr 3

The reality? Complex interplay of genetics & environment leads to major challenges & potential for misinterpretation. 🚨 10 crucial recommendations to fix this: www.biologicalpsychiatrycnni.org/article/S245... #Science #Bias #Data

Ten Suggestions for Better Inference in Population Neuroscience Studies

Neuroscience has expanded to a scale where it can inform research about diverse human populations, emphasizing larger samples, better demographic representation, and varied socioeconomic backgrounds (...

www.biologicalpsychiatrycnni.org

Martin P. Paulus @mpwpaulus.bsky.social · Apr 3

Great work by Wes Thompson, Evan White and Chun Chieh Fan at LIBR as well as others: The promise of population neuroscience: understanding brains across diverse groups.

1 1 2

Martin P. Paulus @mpwpaulus.bsky.social · Mar 28

I think you will find that these podcasts are engaging and informative. I hope you give them a try and let us know what you think,

Martin P. Paulus @mpwpaulus.bsky.social · Mar 28

We have asked our investigators to provide us with their most recent publications and we have used a new feature developed by Google called NotebookLM (notebooklm.google.com) to generate podcasts from these publications. www.laureateinstitute.org/featured-pub...

Sign in - Google Accounts

notebooklm.google.com

1 1

Martin P. Paulus @mpwpaulus.bsky.social · Mar 28

I am excited to share a new feature on our LIBR website. In response to a query from one of our research participants who expressed interest in our work and asked whether there were podcasts available about what "LIBR is working on", we have launched a new project.

Sign in - Google Accounts

notebooklm.google.com

1 2

Martin P. Paulus @mpwpaulus.bsky.social · Mar 28

Exploring options to counter US tariffs? Consider the impact of adjusting Treasury holdings. A nuanced approach, but with potential financial leverage. #EconomicStrategy #GlobalFinance #USTreasuries

Reposted by Martin P. Paulus

Russ Poldrack @russpoldrack.org · Mar 16

Amazing story

Nick Byrd, Ph.D. @byrdnick.com · Mar 16

Until this weekend, I did not know that this is how Daniel Kahneman died. I am neither surprised nor disturbed, and I expect this to become more common.

www.wsj.com/arts-cul...

2 10

Reposted by Martin P. Paulus

Jeremy Berg @jeremymberg.bsky.social · Mar 6

Compelling pushback from @jmgrohneuro.bsky.social and the Board of Scientific Counselors at the National Institute on Mental Health (NIMH) regarding termination of tenure-track investigators driven by DOGE.

1/2

19 350 1.1K

Reposted by Martin P. Paulus

Siri Leknes @sirileknes.bsky.social · Feb 18

As the world grows scarier, let's talk about social connection!

We review its role in addiction & how it's influenced by opioids, now out in Biological Psychiatry (special issue on social homeostasis)

It's @loseth.bsky.social's brainchild & covers A LOT🧠

doi.org/10.1016/j.bi...

#neuroscience

Redirecting

doi.org

6 18

Martin P. Paulus @mpwpaulus.bsky.social · Feb 14

Essential for #biomarker discovery in #neuroscience! 🧠🔬 #EVresearch #astrocytes doi.org/10.1002/jex2...

Comparison of Methods for Isolation and Characterization of Total and Astrocyte‐Enriched Extracellular Vesicles From Human Serum and Plasma

Experimental workflow for total EV isolation and AEEV enrichment. A. Human serum or plasma samples were collected from blood and stored at −80°C until isolation of total EV. B. Depiction of several m...

doi.org

Martin P. Paulus @mpwpaulus.bsky.social · Feb 14

A tour-de-force by Leandra Figueroa-Hall: A comparison of EV isolation methods from human serum & plasma reveals a trade-off: Precipitation methods (e.g., ExoQuick) yield more EVs, but size-exclusion (SmartSEC, qEV) offers greater purity.

1 2

Reposted by Martin P. Paulus

Eric Topol @erictopol.bsky.social · Feb 11

A Direct Hit, by @holdenthorp.bsky.social @science.org
www.science.org/doi/full/10....
"This is a moment to unite."

24 380 900

Martin P. Paulus @mpwpaulus.bsky.social · Feb 8

My GPT friend tells me: "NIH’s 15% indirect cost cap can be legally challenged under the APA, contract law, & Takings Clause to block this policy. Congress must intervene to protect U.S. science. #FundResearch #NIH"

1 6

Martin P. Paulus @mpwpaulus.bsky.social · Dec 26

📢 This is an interesting new preprint. 💡 The authors developed a framework using LLMs to boost creativity in scientific research. 🧪 The approach recombines existing knowledge to generate novel ideas, grounded in Boden's theory of combinatorial creativity. arxiv.org/abs/2412.14141

1 2

Martin P. Paulus @mpwpaulus.bsky.social · Dec 21

🚀 Interesting preprint about AI agent simulations? #LMAgent! A multimodal agent society powered by LLMs, simulating 10,000+ agents in e-commerce scenarios. Realistic behaviors, herd effects, & 40% efficiency boost—paving the way for next-gen social science! 🌐🛍️ #AI #Research arxiv.org/abs/2412.092...

1 4

Martin P. Paulus @mpwpaulus.bsky.social · Dec 20

Interesting piece about Illbeing ≠ Wellbeing: the study shows overlapping genetic influences but distinct environmental factors shaping mental health. The authors suggest that integration is key for future interventions. #MentalHealth #Wellbeing

4

Martin P. Paulus @mpwpaulus.bsky.social · Dec 17

OpenAI's o1-preview achieves superhuman performance in clinical reasoning: Diagnoses correct in 78.3% of cases, outperforming GPT-4 (88.6% vs 72.9%). Excels in diagnostic & management tasks, scoring 99% on reasoning tests.
#AIinMedicine
arxiv.org/abs/2412.108...

4

Martin P. Paulus @mpwpaulus.bsky.social · Dec 16

In a new preprint Van Calster et al. provides comprehensive guidance on evaluating predictive AI in healthcare. Recommends core metrics: AUROC, calibration plots, net benefit via decision curve analysis. Essential reading for robust AI model validation! 🔗 arxiv.org/abs/2412.10288 #AIHealth #MedAI

3

Martin P. Paulus @mpwpaulus.bsky.social · Dec 16

Deep learning in medical imaging often takes shortcuts, finding spurious patterns instead of meaningful insights (e.g., predicting diet from knee X-rays 🤯). To mitigate this: diversify datasets, mask confounders, use adversarial training, and validate across distributions. #AIethics

Janelle Shane @janelleshane.com · Dec 14

The authors point out that AI models base their predictions on sneaky shortcut effects all the time; they're just easier to identify when the conclusions (beer drinking) are clearly spurious.

www.nature.com/articles/s41...

The risk of shortcutting in deep learning algorithms for medical imaging research - Scientific Reports

Scientific Reports - The risk of shortcutting in deep learning algorithms for medical imaging research

www.nature.com