Lightnews — Scholar-powered news

Andrew Little

@anthlittle.bsky.social

6.1K followers 460 following 730 posts

Prof at UC Berkeley. Formal theory, political beliefs, democracy. Associate Editor at ‪@ajpseditor.bsky.social‬ https://anthlittle.github.io/

anthlittle.github.io

Posts Media Videos Starter Packs

Pinned

Andrew Little @anthlittle.bsky.social · Feb 27

New paper! @annemeng.bsky.social's thread lays out the key findings, but the v short version is (1) concessions are rising, (2) but primarily for unfair elections, (3) incumbents concede much more and this is probably causal, (4) weak evidence for conceding being a norm osf.io/preprints/os...

Andrew Little @anthlittle.bsky.social · 8d

May I tap this sign 😉

Andrew Little @anthlittle.bsky.social · 21d

I don’t think this should primarily be described as autocratic rather just extremely stupid and embarassing. Not autocracy, clownocracy.

Acyn @acyn.bsky.social · 21d

Deputy AG Blanche: RICO is available to all kinds of organizations committing crimes and committing wrongful acts… So is it sheer happenstance that individuals show up at a restaurant where the president is trying to enjoy dinner and accost him with vile words?

Reposted by Andrew Little

Lindsey Cormack @dcinbox.bsky.social · Aug 22

Ok - so I'm going to do a real context+write up but for now, here's what some of these things look like.

To start my data reference is DCinbox which is ~208,000 official e-newsletters over the past 15 years.

Reposted by Andrew Little

Xiaobo Lü | 吕晓波 @xiaobolu.bsky.social · Aug 22

The CCP’s triumph over the KMT shocked the world—even the Soviets bet on the KMT with more aid. My book, Domination and Mobilization, asks: how did the CCP survive and prevail?

Domination and Mobilization offers three fresh arguments that explain the reversal of the century. 👇

cup.org/3G6WKB2

Andrew Little @anthlittle.bsky.social · Aug 22

Definitely some discussion of this in global games papers but blanking on specific cites :)

Andrew Little @anthlittle.bsky.social · Aug 22

Sorry just saw this! I think this idea could be strategic in a strict sense: e.g., if I want to participate in political movements that succeed even though I'm not pivotal ("warm glow" effects, etc.), beliefs about whether it will succeed may drive participation.

Andrew Little @anthlittle.bsky.social · Aug 22

For a shorter description of the key idea with some toy models applying it to persuasion and whether "thinking" leads to more or less confidence, we also have a companion AEA P&P piece here:
anthlittle.github.io/files/augenb...

anthlittle.github.io

Andrew Little @anthlittle.bsky.social · Aug 22

In sum, the world is complicated and we need to make simplifying assumptions to understand it. This is a key driver (if not the key driver) of both overconfidence and disagreement in beliefs. My hunch is that this also explains much disagreement in politics.

Andrew Little @anthlittle.bsky.social · Aug 22

The observational data: the Survey of Professional Forecasters lets us test some other predictions from the theory. As predicted, “excess” MSE above what variance alone would imply equals twice the disagreement in individual forecasts. (Honestly we were shocked at how well the data fits the theory!)

Andrew Little @anthlittle.bsky.social · Aug 22

Here is a graphical version of the key result. Participants are unresponsive to changes in across-model uncertainty (left & middle panels), but reasonably responsive to within-model uncertainty (right panel).

Andrew Little @anthlittle.bsky.social · Aug 22

A nice thing about this design is that we can also independently vary across-model uncertainty by changing the prediction date when the trend-line is hidden, and within-model uncertainty by changing the noise when it is shown.

Andrew Little @anthlittle.bsky.social · Aug 22

The experiment: participants predict future “sales” data generated by a linear trend plus noise, and report their uncertainty. Sometimes they see the trendline (so only within-model uncertainty matters). Sometimes they don’t (introducing across-model uncertainty).

Andrew Little @anthlittle.bsky.social · Aug 22

The theory produces many predictions, but the core one is that under broad conditions, across-model uncertainty, overprecision, and disagreement move together. Under stronger conditions, they exactly coincide.

Andrew Little @anthlittle.bsky.social · Aug 22

In particular, we assume people account for uncertainty given their assumptions (“within-model uncertainty”) but neglect the fact that other assumptions could imply different beliefs (“across-model uncertainty”).

Andrew Little @anthlittle.bsky.social · Aug 22

The theory: forming beliefs requires assumptions (or a "model") about how data map to outcomes. We develop a model with strong simplifying assumptions to explore how using models to make simplifying assumptions affects beliefs.

Andrew Little @anthlittle.bsky.social · Aug 22

Very happy to share this paper: Like it says on the tin, we study how simplifying assumptions drive overprecision (excessive certainty) and disagreement (divergence in beliefs), using theory, an experiment, and observational data.

Draft here: osf.io/preprints/os...

Quick thread below.

Andrew Little @anthlittle.bsky.social · Jul 16

Sure but for that goal we could just arbitrarily delay making all decisions, and if you want to make that case you are on your own 😉

Andrew Little @anthlittle.bsky.social · Jul 16

Using LLMs as a sounding board has helped me solve a lot of technical problems, or at least solve them faster than I would have otherwise. But when you do this on a topic you know well you realize they make a ton of mistakes (while exhibiting 100% confidence) that a non-expert wouldn't catch.

ghost malone ergo propter malone @proptermalone.bsky.social · Jul 16

I think it's probably really useful to have LLMs act as a sounding board/sparring partner for experts tbh but Travis Kalanick talking to Grok about quantum physics, or to anyone else about anything else, is not that

Paul Waldman @paulwaldman.bsky.social · Jul 16

My god these guys are such spectacular morons

gizmodo.com/billionaires...

Andrew Little @anthlittle.bsky.social · Jul 9

I've seen editors of journals say this before but being on the other side can confirm: the easiest zero-cost way to speed up the review process is to quickly turn down requests to review that you can't do.

Andrew Little @anthlittle.bsky.social · May 28

Some earlier reporting on problems with the case www.buzzfeednews.com/article/kenb...

The FBI Said It Busted A Plot To Kidnap Michigan’s Governor. Then Things Got Complicated.

The case seemed like a lock — until an informant and one FBI agent were charged with crimes, another was accused of perjury, and a third was found promoting a private security firm. And that wasn’t al...

www.buzzfeednews.com

Andrew Little @anthlittle.bsky.social · May 28

The GOB presidency
youtu.be/APWXorE6h8U?...

Andrew Little @anthlittle.bsky.social · May 28

Maybe a good day to recommend this podcast on the Whitmer kidnapping plot. Those convicted were far from blameless, but closer to "jokers strung along by government informants" than criminal masterminds. podcasts.apple.com/us/podcast/t...

The Michigan Plot I 1. The Man in the Vac-Shack Basement

Podcast Episode · Chameleon: Dr. Miracle · S7 E1 · 41m

podcasts.apple.com

Andrew Little @anthlittle.bsky.social · May 14

Also apologies to @edogrillo.bsky.social for accidentally demoting him to second author!

Andrew Little @anthlittle.bsky.social · May 14

Yeah i've also had moments of "sounds like Grillo and Prato's AJPS," I think for the same reason Adam thought of it

Andrew Little @anthlittle.bsky.social · May 14

I’m generally on team “talking about Abrego Garcia is good both politically and morally” and this is clever but I’d want to see if the treatment durably shifts beliefs before reading too much into it. Would guess this is mostly changing how people interpret the Q.

G Elliott Morris @gelliottmorris.com · May 14

Rare for a public poll, we did a survey experiment to test whether priming respondents about the deportation of Kilmar Abrego Garcia impacts support for Trump's broader immigration agenda. It does. Support for blanket deportations fell 20 points after hearing about Garcia's case.