Tessera
Folgen

Calvin Lai

@calvinklai.bsky.social

Social psychologist & professor at Rutgers University. Studying how to reduce prejudice and discrimination. Opinions are my own. he/him calvinklai.com

93 Posts2467 Follower461 Folgt
Society for the Improvement of Psychological Science (SIPS)@improvingpsych.org · 2w ago

Congratulations to the “Project Implicit,” which received🏆2026 SIPS Mission Award! Project Implicit offers online tests to millions of participants, uses those tests to document implicit and explicit biases, and shares massive open datasets to power discoveries in psychology and beyond.

Project Implicit

Project Implicit runs on small donations, most under $5. That’s what keeps the tests online, the data public, and the science accessible to anyone around the world who is interested in learning about…

implicit.harvard.edu

Chokely Carmichael🥋@stewartcoles.bsky.social · 2w ago

Engaging in the peer review process—from any position within it—would quickly disabuse anyone of the notion that the academy is full of groupthink. You can’t get even a small group of experts to agree on what is important or appropriate. Half the time they don’t even agree on what *is*.

Jake Grumbach@jakemgrumbach.bsky.social · 2mo ago

My new theory paper (early draft) A new reason why concentrated wealth destroys democracy: common ownership A standalone media owner maximizes profit. A common owner like Musk (who owns X + SpaceX) will run the media firm at a loss to increase the $ of his total portfolio tinyurl.com/mcekr7cd

Wealth Concentration and Democratic Erosion: A Theory of Media and Common Ownership* Jacob M. Grumbach„ June 18, 2026 Abstract I develop a model of how extreme wealth concentration harms democracy through a common ownership problem. A standalone profit-seeking media firm will maximize advertising revenue and audience size, producing content that is orthogonal to democracy. In contrast, an owner who controls both a media firm and other large firms can use the media firm to produce anti-democratic content, which reduces the profitability of the media firm but maximizes the profitability of the overall business portfolio through greater rents from an authoritarian leader (e.g., preferential regulatory treatment or public contracts). Anti-democratic content grows in the owner’s wealth and in the authoritarian premium, the differential value of concentrated capital under authoritarian governance. I theorize three extensions: an oligopolistic setting in which reduced democratic content is a common pool resource; an endogenous authoritarian premium based in the authoritarian’s targeting problem; and owners that vary in proactively or reluctantly supporting the authoritarian.
Jake Grumbach@jakemgrumbach.bsky.social · 2mo ago

Wealth concentration was already way off the rails before Musk became a trillionaire. The wealthiest megabillionaires in the country have basically been doubling their wealth annually in recent years. Feels near impossible to maintain a liberal democracy in this kind of situation.

Benedek Kurdi@benedek.bsky.social · 2mo ago

How does amount of counterattitudinal information influence whether and how much attitudes change? Rydell et al. (2007) found that implicit evaluations changed slowly and incrementally, whereas explicit evaluations changed abruptly in response to counterattitudinal info.

A plot showing incremental implicit evaluation change in response to 20, 40, 60, 80, and 100 counterattitudinal statements.A plot showing abrupt explicit evaluation change in response to 20, 40, 60, 80, and 100 counterattitudinal statements.
John Holbein@johnholbein1.bsky.social · 2mo ago

Running an experiment to study racial discrimination? Going to use names to signal race? Here's a new tool! This helps you to choose the names you want to use based on: 1.) the strength of the racial signal you want to send 2.) whether you want to hold other factors (e.g., class) constant

Alejandra Caraballo@esqueer.net · 2mo ago

The drop was driven almost entirely by a cratering of support within the Republican party by 21 points. 2022 was the inflection point when this decline started. It's now lower than it was in 2012 before President Obama publicly supported marriage equality.

presents a line chart titled "Perceived Morality of Gay or Lesbian Relations, by Party ID". The chart tracks the percentage of U.S. adults who believe gay or lesbian relations are "morally acceptable" from 2005 to 2026, segmented by political affiliation. Key observations from the data include: Democrats: This group shows the highest levels of moral acceptability, trending upward from 50% in 2005 to 81% in 2026. Independents: This group follows a similar upward trend, starting at 53% in 2005 and reaching 64% by 2026. Republicans: This group consistently reports the lowest levels of moral acceptability, beginning at 31% in 2005 and finishing at 35% in 2026, after reaching a peak in the early 2020s.
Josh Chafetz@joshchafetz.bsky.social · 2mo ago

1/ This is awful for so, so many reasons. Just to pick one: even if you thought it was lawful and appropriate to vet funded research for consistency with the government's priorities, *the president is not the government*.

Jess Calarco@jessicacalarco.com · 2mo ago

This is bad. Bad for researchers. Bad for science. Bad for universities. Bad for professional associations. For example: 1. Every award would be vetted by political appointees, who could overrule peer reviewers and deny the award if it doesn't advance the president's priorities. 1/🧵

Martin Naunov@naunovmartin.bsky.social · 3mo ago

Thrilled this is finally out in its JOP-formatted version! Hope it's useful to anyone working on LGBT politics, (electoral) discrimination, status differentials within & across identity categories, and non-textual experiments. Main findings in quoted thread. www.journals.uchicago.edu/doi/epdf/10....

Bild
Martin Naunov@naunovmartin.bsky.social · last yr.

1/ Much has been said about 🇺🇸 becoming more accepting of gay candidates—especially since Pete Buttigieg’s presidential run. But I ask: What kind of gay candidates no longer face discrimination? And, specifically, how does gender presentation affect electoral viability?

Philip Ball@philipcball.bsky.social · 3mo ago

"a PRS is not an internal predisposition waiting to be realized; it is a risk estimate derived from a model constructed from population-level DNA associations without clearly defined biological mechanisms."

Eric Turkheimer@ent3c.bsky.social · 3mo ago

With @remyfurrer.bsky.social, a new piece in Nature Human Behavior. The standard description of a polygenic score as a "genetic predisposition" essentializes a statistical construct. It's not language policing: PGS are a new kind of entity; now is the time to think carefully about what they are.

Mark Thornton@markthornton.bsky.social · 3mo ago

Where did this year's #psychjobs market end up? Thanks to an unusually high number late postings, not as bad as it was looking earlier in the year. Still, this is the smallest number of jobs posted since 2012, representing a 26% decline from last year (~2/3rds the size of the covid drop in 2020).

Bar graph showing tenure track psychology jobs posted each year since 2010 broken up by area.
Mark Thornton@markthornton.bsky.social · 9mo ago

Another ~monthly update on the state of the #psychjobs market: more jobs continue to trickle in, though at a decreasing pace. We've nearly hit parity with covid in absolute terms, although the prior year baseline was higher this year than that, so we still see a slightly larger relative decrease.

Sol Messing@solmg.bsky.social · 3mo ago

New in Nature: LLMs give "the party line" in the languages of authoritarian regimes. This works when they control the media, which feeds pretraining data. We show more state control over the media means less critical LLMs. 6 studies spanning 38 languages & 13 models. Details ↓

Scatter plot of 38 countries plus China showing the proportion of LLM responses favorable to the regime in the local language vs. World Press Freedom Index score. Lower press freedom predicts more regime-favorable responses; all production models pooled.