Thomas Henning

Thomas Henning

@thomashenning.bsky.social

Social Decision Neuroscience PhD Student @ Caltech. Neuroscience of AI-Human interactions.

388 followers125 following85 posts

ASU prestige continues to climb!

Aaron Rupar@atrupar.com

Vance: "Some of America's most elite universities bring in foreigners to undercut wages of American residents. There are 9 universities in particular we believe merit an investigation. Those are Harvard, Yale, Stanford, Brown, the University of Pittsburg, UC Davis, Cal Tech, Arizona State, and MIT."

Thomas Henning reposted

"Against this theoretical backdrop, there is a growing body of empirical research documenting that early childhood programs can generate meaningful and lasting improvements in children’s life chances" #Econsky

NBER@nber.org

Several decades of Head Start early childhood care and education research provide insight into variation in program effectiveness, over time, across communities, and by child and family characteristics, from @chloergibbs.bsky.social www.nber.org/papers/w34812

Thomas Henning reposted

Thomas Henning reposted

I know all eyes are on Minneapolis, but there's a fast-approaching nightmare in Springfield, Ohio. Trump is revoking protected status for tens of thousands of Haitians living there on Feb. 3. Reportedly, on Feb. 4, 1000 ICE agents are arriving to remove this population. Ethnic cleansing.

Thomas Henning reposted

Recessions aren’t uniform events. Decomposing the labor market by race, education, and underemployment, using @claudia-sahm.bsky.social-style turning points, the cyclical sensitivities differ sharply—because some people experience recessions first. There’s an inequality to recession. #EconSky

Thomas Henning reposted

For those attending ASSA, EconScience meetings, here is our session on LLMs in experimental economics (as synthetic subjects) 1015am 5 Jan 2026. Audience will be discussants, using Slido app; everyone can submit questions at any time. We're all looking forward to it. @thomashenning.bsky.social

Breaking news; PhD Student Thomas Henning revealed to have said “The philosophical difference between Bayesian and Frequentist stats is fake; given that all statistical methods require assumptions and the debate is only on which of them to directly quantify “

Thomas Henning reposted

A new paper by Nobel Prize-winning economist Claudia Goldin has found that countries where men take on more of the household labor and child care — in other words, those who buck traditionalism’s standards of the provider husband and homemaker wife — have higher fertility rates.

19thnews.orgTo raise fertility rates, it’s not women who need to step up — it’s menNew research found that countries where men do more housework and child care have higher fertility rates.

Why is dollar cost averaging so popular. It makes no sense. EV in the market is highest right now! Unless you believe in market timing (you shouldnt unless you sit in a big Bank/HF and have access to all “public wink wink” data, you should invest what you plan to DCA right away! With few exceptions

Becoming increasingly clear to me that LLM have an “effect bias”. When you ask about the relationship between two things, it defaults to stating there is some relationship (regardless of truth) much more often than stating there is no relationship.

As any researcher can probably tell you (due to the painstaking PhD process) this is like the exact opposite of reality! And yet most humans likely also default to assuming there is an effect! Pattern seeking brain, pattern seeking LLM? Curious where this bias comes from in both humans and LLMs

While it seems to take humans years of deliberate practice to become more skeptical, I wonder what it would take for LLMs to encode the idea that effects between things are often small and non-existent; and frankly even regardless of whether they’re published in a fancy journal.

Hearing many American students are now looking abroad for grad school, something never really seen in prior years. We are watching a brain drain from our country in real time; what an unforced error that will have generationally long reprecussions.

Coinciding with many less applicants from students abroad. To see what was once one of America’s greatest institutions in such disrepair is beyond disheartening. Not sure what the solutions are, if any exist on how we can ever truly recover.

I’ve said it before and I’ll say it again, the best defensive lineman in the world players power forward for the New Orleans Pelicans. Thinking back on Julius Peppers career and contrasting it with Williamson; does Peppers focus on basketball full time with today’s NBA salaries? Incentives matter!

Every year I get more and more teary eyed during commencement. Seeing the culmination of years of work done by your bf colleagues and mentees (and mentors) is a deeply moving feeling. So proud of the entire Caltech class of 2025 both graduates and undergraduates!

This was the first year where I had been able to see the class of undergraduates throughout their entire Caltech career. Seeing the result of their growth both academically and personally puts the value of academia into perspective. At the end of the day this is what it’s about!

Flying into LA after traveling is always depressing; LA should be one of the greatest cities on the planet and instead it’s a destitute suburb. America needs to revitalize and modernize our cities, it’s embarrassing.

Giving a poster presentation Sunday at the Financial AI workshop @ ICLR! Check out the work and give me your suggestions! We find that LLMs (we test 6 models) tend to trade around fundamental value in a fixed FV experimental market; compared to humans who generally create bubbles!

NEP-MAC: Macroeconomics@repec-nep-mac.bsky.social

LLM Trading: Analysis of LLM Agent Behavior in Experimental Asset Markets: Thomas Henning; Siddhartha M. Ojha; Ross Spoon; Jiatong Han; Colin F. Camerer

Should we find this at all surprising? Knowledge is building on top of eachother. As we go deeper the pre-requisites increase. This is like saying Calculus is harder than Algebra; frontier math is getting harder to understand!

Owen Winter@owenwntr.bsky.social

Over the past 80 years, academic writing has become substantially harder to read 📉🧪 www.economist.com/science-and-... I analysed 350k PhD thesis abstracts and found that they've become more complex in every discipline, especially the humanities and social sciences

Not sure what it is about ML but I always tend to spend an inordinate amount of time focusing on the theory/math of a model; And not nearly enough time playing around with the model. I’d like to switch to the latter form of learning, but it’s proven to be difficult mental process to escape

How many other times in history has this happened, and what can be done to get this back in order? It feels different than just misinformation; anyone studying this type of phenomena?

Aaron Sojourner@aaronsojourner.org

The more people are asked about their own experience with the economy, the more positive they are. The more they are asked about far-away aspects of the economy (national), the more negative they are. The more they have to rely on outside info, the more negatively misinformed they are.