This study shows that disinformation spreads, not as “false facts”, but as narratives. Compare that to how mis- and disinformation are often operationalized in the big-N quantitative and experimental research in the Nature/Science journals — that’s a major weakness in the field.
Thanks for the feedback! We discussed this too, and tried to address by reporting integrative data analyses & reporting simple effects estimates w CIs
E.g., we observed assigned groups were rated 1.3 pts more trustworthy on avg (of 7), and the exclusion manip moderated this by 0.11 pts [.08, .13]
We conclude by arguing that the next wave of research on motivated belief formation should move beyond the question of whether it exists, and start asking questions about which motivations and under what conditions.
Of course, effect sizes weren't v large--these minimal groups weren't very rich and likely didn't have the motivating force of, say, political groups.
But they do help move theory forward by showing that group identity--even at its most cleanly manipulated--can motivate epistemic processes.
Figure showing A) trustworthiness ratings a function of social group and exclusion or B) in-group preference in trustworthiness as a function of excusion. Participants who worked alone or faced exclusion trusted their arbitrary in-group more--and particpants who faced social exclusion were more likely to face this exclusion.
Ps in both conditions trusted their assigned in-group more. But the participants who first faced social exclusion were more likely to preferentially trust their in-group more.
Social motivations to affiliate with groups make people more likely to trust even these arbitrary groups.
Ok--maybe this still not motivated? Ppl may just coldly tend to trust in-group members.
In the final exp, we experimentally manipulated social motives. P's wrote a short bio and picked others to work with. Then we said the task was cancelled bc nobody picked them or because of a computer error
Moderated mediation model where the effect of source group on belief is fully mediated by source trustworthiness. The first path (source group to trustworthiness) is moderated by collective identification).
We randomly assigned people to the "orange" or "purple team", and had them evaluate trivia claims from in-group or out-group members (or neither).
Ppl were more likely to trust, and thus believe statements endorsed by, even these arbitrarily assigned group members--esp when they valued their group
Critically, it's impossible to parse these explanations out bc in real-world groups, motivations and priors/information environments/etc. are confounded.
So we tested these accounts using minimal groups!
But others say the jury's still out on whether all this is really motivational (as opposed to purely cognitive). For example, maybe we just expect we'll agree with our in-group members, or we already have similar views on the topics they might endorse.
Social psychologists argue that motivations can shape what we believe. For instance, when an in-group member (e.g., political copartisan) endorses a claim we might be motivated to endorse it as well. These motivations can even come at the expense of accuracy, leading ppl to endorse misinfo
New preprint with @jayvanbavel.bsky.social: Novel social identities shape belief in true and false information
In 3 exps (N = 1,459) we randomly assigned people to one of two novel groups. People trusted (and then believed claims more from) their in-group
osf.io/preprints/ps...
Thread 🧵
And of course, I could not have possibly gotten here without the help of so so many people who have guided me along the way--in particular @lkfazio.bsky.social and @jayvanbavel.bsky.social !
For folks interested in applying to work with me as a grad student, see below for:
More info about SBU's Cognitive Science PhD program: www.stonybrook.edu/commcms/psyc...
More info about my lab and work: rmpillai.com
Photo of me next to a bronze statue of Wolfie the Seawolf, Stony Brook's mascot
I'll be starting as an Assistant Professor of Cognitive Science at Stony Brook in Fall '26!
My lab will take a social cognitive approach to studying how we form/update our beliefs and why we disagree about what is true vs false
I'll also be reviewing grad apps! See below for more info
"Vanderbilt is failing in its moral responsibility to push back against rising authoritarianism and political coercion."
My effort to convince Vanderbilt (and all universities) to take a stand for the rule of law, the Constitution, and our democracy.
www.tennessean.com/story/opinio...
I'm excited to finally have a preprint of this paper up, a few years in the making.
In it we argue that industry-driven manipulation of social media research is well underway and that norms and institutions in the field are ill-prepared to resist tech's influence.
arxiv.org/abs/2510.19894
New from me and @brendannyhan.bsky.social
Why Trump's "Compact for Academic Excellence" is a devil's bargain and should be rejected in all forms
www.msnbc.com/opinion/msnb...
Coauthors: Laura Globig, @steverathje.bsky.social, Anni Sternisko, @remi-theriault.com @jayvanbavel.bsky.social
CC: @centerconflictcooperation.com
New preprint: From vaccinations to masking, politics predicts health behaviors. In this review we discuss these differences in light of classic psychological theories of group identity and behavioral decision-making.
We welcome any feedback on this working paper!
osf.io/preprints/ps...
🚨 10 Days Until Stand Up for Science, and some HUGE NEWS! - Take Action With Us! 🚨
We’re excited to announce 32 official Stand Up for Science events happening across the country on March 7th!
🧵⬇️
In a new article now out in Collabora: Psychology, @rmpillai.bsky.social and I find that the mere repetition of information increases estimates of its consensus, a finding we refer to as an “illusory consensus effect.”
S+P Psychologists, Michael Kraus put together an excellent letter calling on social and personality psychologists to meet the moment re protecting students' rights to free speech. He would love faculty signatures! If you're a PhD Student, please send this around!
docs.google.com/document/d/1...