New paper in @nature.com with @lukebeehewitt.bsky.social, @isaiasghezae.bsky.social, & @robbwiller.bsky.social: Can LLMs predict results of social science experiments? Across 70 experiments, we find a strong correspondence (r=.85) between predicted and observed effects. doi.org/10.1038/s415... 🧵
Isaias Ghezae
@isaiasghezae.bsky.social
PhD student at Harvard interested in intergroup conflict and political psychology https://isaiasghezae.com
National identity reconfigures brain responses from “them” to “us” Briefly priming national versus ethnic identity increases ventromedial prefrontal cortex activation for ethnic out-group faces, a pattern typically reserved for self and in-group processing. www.pnas.org/doi/10.1073/...
National identity reconfigures brain responses from “them” to “us” | PNAS
How the human brain flexibly adapts social perception by recategorizing out-group (them) to in-group (us) remains unclear. Using functional MRI in ...
pnas.org
More evidence of the non-left/right, anti-system axis in US politics. (Author says this was inspired by my writing at Strength In Numbers!)
🚨New Working Paper!🚨 How do Americans mentally map political coalitions? We often talk as if U.S. politics is just two sorted camps. But when people think about which groups go with which, the picture is more complicated. 🧵👇
🚨New Working Paper!🚨 How do Americans mentally map political coalitions? We often talk as if U.S. politics is just two sorted camps. But when people think about which groups go with which, the picture is more complicated. 🧵👇
New deep dive piece on 2 new big studies on ideology and moderation, including a data exclusive from Blue Rose Research. Short version: Only 8% of "moderates" say they want a moderate party. And when candidates do move to the middle, fx are often negative www.gelliottmorris.com/p/2026-04-07...
Strength In Numbers
Independent, data-driven analysis of politics, public opinion polls, and elections. From author, journalist, and pollster G. Elliott Morris. Click to read Strength In Numbers, by G. Elliott Morris, a…
gelliottmorris.com
🚨🚨 My lab is hiring a Pre-doc Researcher/Lab Manager! 🚨🚨 Full-time pre-doc research opening in my lab at Harvard Psych (start: Summer/Fall 2026). Review begins April 10 Apply: jobs.smartrecruiters.com/HarvardUnive... Please repost + tag folks who might be a good fit! 👇
Research Assistant II
Company Description: By working at Harvard University, you join a vibrant community that advances Harvard's world-changing mission in meaningful ways, inspires innovation and collaboration, and builds...
jobs.smartrecruiters.com
From me and @adambonica.bsky.social: the fact that so many voters trusted Trump over the Democratic Party on prices really shows the limits of swing voters’ understanding of public policy www.bostonreview.net/forum/how-no...
I joked this week that Trump’s presidency has proceeded as if he hired a crack team of consultants to come up with ways to make prices go up. Increasingly feels like the Israeli government is trying its hardest to drive its approval ratings in the US as low as possible.
This essay best captures my experience on how context-aware AI like Claude Code/Codex adds value to my research. AI as a Research Collaborator, Not a Paper Mill modelinganunforeseeablefuture.substack.com/p/ai-as-a-re...
Donald Trump's job approval is 34 points lower than his 2024 vote margin among working-class Hispanics — and working-class whites now on net disapprove of how he's handling his job as president, per new Strength In Numbers/Verasight polling www.gelliottmorris.com/p/workers-tr...
Today in the NYT, I discuss some fresh research on which moderations move the most voters. www.nytimes.com/2026/03/16/o...
New short paper w @jkalla.bsky.social ! Candidates gain from moderation, but less than many theories expect. Many conclude voters must not care about issues. This is wrong. Small *average* effects mask large effects on specific issues & are consistent with widespread issue-based voting 🧵
The sky is not falling; high-quality platforms (Prolific, Verasight, CR Connect) have low rates of apparent bots. osf.io/preprints/ps... But also not zero; vigilance is very much needed!
Another cool data blitz from #SPSP political psych precon. People asume there are two large coalitions in the U.S.: liberals & conservatives. But, work by @isaiasghezae.bsky.social suggests there are instead 3: (1) conservatives, (2) ideological liberals, and (3) a liberal demographic group.