Lightnews — Scholar-powered news

Marianna Zhang

@mariannazhang.bsky.social

500 followers 290 following 120 posts

developmental cognitive scientist & (terminated) NSF postdoc fellow @ NYU | social categories, development, language, climbing, caving | 🗽🌬️🌉🗽 | she/她 mariannazhang.github.io

Posts Media Videos Starter Packs

Pinned

Marianna Zhang @mariannazhang.bsky.social · Apr 26

yesterday, my postdoc funding (salary and research funds) was cancelled by the National Science Foundation, effective immediately. I received the same generic, vaguely threatening, typo-ridden email as many of my colleagues who have had their awards terminated recently. (1/n)

email starting, "The U.S. National Science Foundation (NSF) has undertaken a review of its award portfolio. Each award was carefully and individually reviewed, and the agency has determined that termination of certain awards is necessary because they are not in alignment with current NSF priorites."

41 280 610

Marianna Zhang @mariannazhang.bsky.social · 12d

this is also difficult for PhD programs at institutions on the quarter system, where first-year students have only 3 weeks to get to know their advisor, settle in, and submit a proposal on the typical GRFP timeline! (most wait for 2nd year)

Reposted by Marianna Zhang

Erica D. Pratt @pratted.bsky.social · 12d

Announcing such a major change to NSF GRFP this late in the cycle is incredibly cruel. 2nd yr students (mine included) have been working so hard on their proposals despite ongoing uncertainty. They are driven and passionate about being outstanding scientists and helping those coming up behind them.

This means individuals in the following statuses at the time of application are eligible:

Undergraduate in the final (senior) year of a bachelor’s degree program
Bachelor’s degree-holder with NO enrollment in a graduate degree program (non-degree graduate coursework allowed)
Individual enrolled in a joint bachelor’s-master’s degree program with at least three undergraduate years completed
First-year graduate student in their first graduate degree program with less than one academic year completed in the degree program (according to institution’s academic calendar)
Individuals enrolled in joint bachelor’s-master’s degree programs are considered graduate students. For GRFP, joint bachelor’s-master’s degrees are defined as degrees concurrently pursued and awarded.
Not be a current NSF employee.

Applications that do not meet eligibility requirements will be returned without review as being ineligible for a fellowship.

7 77 160

Reposted by Marianna Zhang

Kristina Olson @olsonista.bsky.social · 27d

Princeton is running its Presidential Postdoctoral Program again this year. If you are interested in applying to work with me, please reach out by October 10 so that I can consider all interested applicants and support a candidate for my lab this year.
dof.princeton.edu/faculty-deve...

Presidential Postdoctoral Research Fellows

We encourage early career scholars to pursue a career in academia by supporting their postdoctoral work at Princeton

dof.princeton.edu

19 16

Reposted by Marianna Zhang

Dan Garisto @dangaristo.bsky.social · 28d

New: A legal challenge to NSF's 1600+ grant terminations just hit a roadblock. Judge Cobb (appointed by Biden) ruled against the plaintiffs, largely on the basis that district courts no longer have jurisdiction for contract-based APA violations. storage.courtlistener.com/recap/gov.us...

The Court finds that it likely lacks jurisdiction over Plaintiffs’ retrospective APA claims,
so it must deny Plaintiffs’ motion for preliminary relief as to those claims. In analyzing Plaintiffs’
remaining claims, the Court concludes that Plaintiffs have failed to show irreparable harm flowing
from their prospective APA claims and have not shown a likelihood of success on the merits of
their constitutional claims. As a result, the Court will deny Plaintiffs’ motion for a preliminary
injunction without considering the other preliminary injunction factors.
Accordingly, Plaintiffs’ Motion for a Preliminary Injunction, ECF 3, is DENIED. A
separate order accompanies this memorandum opinion.

1 3 11

Reposted by Marianna Zhang

Cientifico Latino, Inc. @cientificolatino.com · 29d

Applying to the NSF-GRFP (or another fellowship) on a tight deadline?

We built a 7-week guide + timeline to get you from draft to submission. It’s not too late — you’ve got this! ✨

🔗 cientificolatino.com/apply-in-7-weeks

#NSFGRFP #GradSchool #Fellowship

65 64

Marianna Zhang @mariannazhang.bsky.social · Aug 20

not sure, but if I had to guess, maybe they're archiving it so they can make changes and post a new solicitation, based on new agency priorities/budget?

Marianna Zhang @mariannazhang.bsky.social · Aug 20

this happened to a friend of mine who applied for a postdoc fellowship too! heartbreaking

Marianna Zhang @mariannazhang.bsky.social · Aug 13

the judge declined to intervene on the terminations for now, but the case is still ongoing! www.courtlistener.com/docket/70390...

State of New York v. National Science Foundation, 1:25-cv-04452 - CourtListener.com

Docket for State of New York v. National Science Foundation, 1:25-cv-04452 — Brought to you by Free Law Project, a non-profit dedicated to creating high quality open legal information.

www.courtlistener.com

1 2

Reposted by Marianna Zhang

Carl T. Bergstrom @carlbergstrom.com · Aug 8

1. Federal funding for research promotes tech, biomedical, and scientific discovery in the US, and provides training for the sci/tech workforce that has brought immeasurable wealth to the US over the past 75 years.

Today's executive order includes a provision that will obliterate both functions.

Improving Oversight of Federal Grantmaking

By the authority vested in me as President by the Constitution and the laws of the United States of America, and to improve the process of Federal

www.whitehouse.gov

25 550 1K

Marianna Zhang @mariannazhang.bsky.social · Jul 31

I think the effect of generics ends up being mostly in the semantics but there's also a lot of feature variability, which we're digging into in future work!

Marianna Zhang @mariannazhang.bsky.social · Jul 31

I noticed that too - overall the literal listener only does a bit better than null, but for the generics condition specifically, it does quite a bit better (the orange dot is much closer to the diagonal)!

Marianna Zhang @mariannazhang.bsky.social · Jul 31

the effect of generics is a replication of prior work, but the effect of specifics wasn't obvious to us! specifics could have increased inductive potential, particularly when they invoke the category label ("This Zarpie...."), or could have had no effect since they aren't explicitly about categories

Marianna Zhang @mariannazhang.bsky.social · Jul 31

yes other aspects of a category could matter too! it's interesting though that in our study language changes how coherent we think the same social category is. language also shapes whether we think of a category as more inherent/transient (eg "carrot eater" vs "people who eat carrots")

1 1

Marianna Zhang @mariannazhang.bsky.social · Jul 31

this project was a fun intro to computational modeling 🤖 for me & a dream collab w/ @markkho.bsky.social, @marjorierhodes.bsky.social & Sarah-Jane Leslie! learn more at our 🧑‍🏫 #CogSci2025 talk this Fri 4pm (Language and Computation 3) or 📄 proceedings paper escholarship.org/uc/item/2rs3...

blue banner reading CogSci 2025, Friday August 1, 2025 4-4:22pm, with large text "Learning about Inductive Potential from Generic Statements" with a picture and icon of Marianna Zhang in the bottom right

1 7

Marianna Zhang @mariannazhang.bsky.social · Jul 31

capturing the effect of generics requires a semantic understanding of how generics are related to categories 🐯, while the effect of specifics requires pragmatic reasoning 🤔(speaker could have said a generic but chose to say a specific)! (5/6)

three scatterplots each plotting prevalence (model) on x-axis, prevalence (data) on y-axis. scatterplots left to right: null model (reasoning about individuals and features only), literal listener (reasoning about categories and literal meaning of generics), pragmatic listener (reasoning about categories and what the speaker meant). in the null model plot, the dots are all in the same vertical line, with generics and specifics being off the diagonal. in the literal listener plot, the orange dot (generics) is brought onto the diagonal. in the pragmatic listener plot, the blue dot (specifics) is also brought onto the diagonal. the r^2 increases with each plot.

1 4

Marianna Zhang @mariannazhang.bsky.social · Jul 31

our model predicts that learners generalize a novel feature of a category member 🐯 to other members 🐅 *more* after hearing generics about the category, and *less* after hearing specifics. both were confirmed by adults learning about a novel social category! (4/6)

text on left says "adults generalized a novel feature more after hearing generics, and less after hearing specifics". plot on the right shows with generic, baseline, specific on the x-axis and prevalence on the y-axis. small transparent dots indicate participant and trial level responses, large dots with bars indicate means with 95% CIs. generic is higher than baseline which is higher than specific.

2 7

Marianna Zhang @mariannazhang.bsky.social · Jul 31

in our model, a learner learns about a category's coherence and linked features by reasoning about what a speaker meant to communicate about the latter. (coherence = our formalization of inductive potential, an overhypothesis about a feature of an individual being category-linked) (3/6)

a conversation between an adult and a child on the left side of the screen, with a tiger on the savanna in the background labelled "context". a speech bubble above the adult says "Tigers roar". a thought bubble above the child reads "tiger representation" and contains "coherence = ?" with arrow pointing to a set containing "roaring". on the right, a graphical model with category coherence pointing to category-linked features pointing to informative speaker. speaker and individual point to utterance. individual and utterance are observed (grayed) nodes.

1 3

Marianna Zhang @mariannazhang.bsky.social · Jul 31

our proposal: people learn about the inductive potential of categories (how well they support generalization) from language 🗣️, including generic language about categories and specific language about individuals (2/6)

same picture as in previous post, but speech bubbles labelled specific language ("That pedestrian came over to help" and "This pedestrian wasn't looking where he was going") on the "low" end of the inductive potential continuum, and speech bubbles labelled generic language ("Tigers roar" and "Tigers are solitary") on the "high" end of the continuum

1 3

Marianna Zhang @mariannazhang.bsky.social · Jul 31

📣 new paper! people use some categories to generalize (e.g., we generalize something we learn about one tiger 🐯 to other tigers 🐅), but not others (e.g., we don't generalize from one pedestrian 🚶 to other pedestrians 🚶‍♂️). how do people learn what categories allow for generalization? 🧵

continuum of inductive potential from low (relatively minimal categories whose members are dissimilar) to high (coherent meaningful categories whose members are similar) above a cartoon child. an icon of a tiger appears under "high" inductive potential, with "closes eyes when happy" appearing as a feature of a tiger in a zoo, with an arrow pointing to the tiger icon, and a dashed arrow extending it to a tiger on a savanna. an icon of a pedestrian appears under "low" inductive potential, with "closes eyes when happy" appearing as a feature of a woman on a street, with Xs over arrows pointing to the pedestrian icon, and to a different pedestrian.

2 15 45

Marianna Zhang @mariannazhang.bsky.social · Jul 30

Kusina Ni Tess for massive Filipino breakfast! Tu Lan cash-only pho, Garaje for casual counter-serve. for sit-down, Tin for vietnamese, hed verythai for very nice set menus

Marianna Zhang @mariannazhang.bsky.social · Jul 24

i like knots 3d and their app (on sale for free occasionally) for this too! knots3d.com/index

Knots 3D : Interactive Knot Animations

Welcome to Knots 3D, where learning how to tie knots becomes an immersive 3D experience! Our comprehensive library features over 200 animated knots with step-by-step tutorials that transform complex r...

knots3d.com

1 2

Marianna Zhang @mariannazhang.bsky.social · Jul 17

the athletic was actually acquired separately and the sports dept eliminated in favor of it

1 1

Marianna Zhang @mariannazhang.bsky.social · Jul 10

congrats grace!! I found the excerpt captivating to read, and would be so excited to read a book!

Reposted by Marianna Zhang

Dr. David Miller 🏳️‍🌈 @davidimiller.bsky.social · Jul 10

🧪 BREAKING (good news): Senate subcommittee says NO! to Trump's proposed slashes to NASA & NSF funding.

Today, the subcommittee said to keep NASA + NSF funding at $33.9 billion, the same as in FY24.

See 7:15 below. Full Senate appropriations committee meets tomorrow about it.

🧵 1/3

Subcommittee Markup of the Commerce, Justice, Science, and Related Agencies Appropriations Act | United States Senate Committee on Appropriations

United States Senate Committee on Appropriations

www.appropriations.senate.gov

18 370 1K

Reposted by Marianna Zhang

KMW @kmwil.bsky.social · Jul 2

We can’t depend on a filibuster to “solve” a bad budget for either the NIH or NSF. You need to follow this advocacy alert and make sure those budgets are better than the huge cuts proposed. www.ascb.org/society-news...

Protect NIH and NSF Funding – Your Voice Is Urgently Needed - ASCB

In the next six weeks, Congressional committees will decide how much funding the National Institutes of Health (NIH) and the National Science Foundation (NSF) will receive for Fiscal Year 2026. While ...

www.ascb.org

4 3