Tom Costello
@tomstello_
Followers
1,721
Following
607
Media
128
Statuses
1,266
research psychologist + assistant professor @AmericanU .
Cambridge, MA
Joined October 2017
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
Georgia
• 326639 Tweets
#BUS_KnockKnockKnock_KhonKaen
• 135208 Tweets
Bama
• 104063 Tweets
スプリンターズS
• 78277 Tweets
Beck
• 52220 Tweets
Ryan Williams
• 47189 Tweets
#صباح__الخير
• 22534 Tweets
#RIZIN48
• 17072 Tweets
サトノレーヴ
• 16829 Tweets
ナムラクレア
• 16217 Tweets
ママコチャ
• 16208 Tweets
ショアキーパー
• 13748 Tweets
マッドクール
• 13186 Tweets
中山11R
• 11858 Tweets
Ashton Jeanty
• 10156 Tweets
Pinned Tweet
Our new paper, out on (the cover of!) Science is now live!
🚨Out in Science!🚨
Conspiracy beliefs famously resist correction, ya?
WRONG: We show brief convos w GPT4 reduce conspiracy beliefs by ~20%!
-Lasts over 2mo
-Works on entrenched beliefs
-Tailored AI response rebuts specific evidence offered by believers
1/
36
182
567
6
25
137
People with authoritarian political views — whether they are on the far left or far right — have characteristically similar psychological profiles.
Our new JPSP paper takes a data-driven approach to understanding left-wing authoritarianism:
A thread:
1/
41
325
1K
Messed around* and made the best EFA table I've ever seen.
*spent far, far, far too much time -- like, we're talking a full day.
27
27
436
It's not just grit: most (read: a vast majority of) highly cited scales in psychology are subsumed by the Big Five. That is, they can probably be considered [sometimes interstitial] facets of the Big Five.
(Bainbridge, Ludeke, & Smillie, 2022, JPSP)
6
99
353
Happy to share some big personal news:
I’ll be starting as an Assistant Professor in the Department of Psychology
@AmericanU
this fall.
35
10
230
Fun announcement: I’ve accepted a postdoc out of
@mit
in the labs of
@DG_Rand
and
@GordPennycook
@UofRegina
. I’m so pleased to be joining this fantastic group; I’ve been an admirer of them for years! 🥳🥳
14
4
229
New proposition to improve psychological science: all psych researchers are paired with an analytic philosopher who make them define the constructs they are studying
20
31
198
Very happy to share this new paper, forthcoming @ JPSP. We meta-analyze relations between left/right political views and cognitive rigidity. Are rightists more rigid than leftists (or vice versa)? A controversial question. And the answer is complicated.
6
58
176
1 / 🧵Thrilled to present a new article in
@TrendsCognSci
exploring the limitations of the left-right spectrum in political psychology. We propose a more nuanced, data-driven approach (w/ co-authors
@LeorZmigrod
&
@ArberTasimi
).
()
5
44
155
If you’re looking to run LLM-human interaction experiments, we just posted a Qualtrics template + guide here:
STAY TUNED for a much more sophisticated, highly customizable, plug and play chatbot designed for research—in the coming weeks.
🚨WP🚨
Conspiracy beliefs famously resist correction, right?
WRONG: We show brief convos w GPT4 reduce conspiracy beliefs by ~20pp (d~1)!
🡆Tailored AI evidence rebut specific arguments offered by believers
🡆Effect lasts 2+mo
🡆Works on entrenched beliefs
48
244
762
4
30
140
Our latest paper () describes what is, to my knowledge, the most comprehensive meta-analytic test of the rigidity-of-the-right-hypothesis (i.e., that cognitive, motivational, and ideological "rigidity" resonate with political conservatism). 🧵
5
50
133
Do popular self-report measures of intellectual humility (IH) actually assess IH’s conceptual core--metacognitive abilities? Our new working paper raises concerns.
w/
@GordPennycook
@ChristieANewton
@hauselin
12
37
126
Still, as we argue, the nearly universal exclusion of left-wing features from previous descriptions of authoritarianism, with little consideration of LWA’s and RWA’s commonalities, has severely limited our ability to understand what authoritarianism truly IS.
10/11
3
9
115
I made a website for my data visualizations to live.
I created these during my phd; I'd have never imagined enjoying any kind of visual medium prior to grad school, but here we are (my fine motor skills must be in the 1st percentile, tbh).
5
13
120
These traits include a preference for social uniformity, prejudice toward different others, willingness to wield group authority to coerce behavior, cognitive rigidity, aggression toward perceived enemies, outsized concern for hierarchy, and moral absolutism.
4/11
1
19
113
I’ll be on the faculty job market this year. I’m a personality-social psychologist. Website here:
If you’re hiring, have any tips, or just want to say good luck, I would love to hear from you!
16
27
108
Now typo free: I’m excited to share our review on the psychology of authoritarianism—its present, future, and discontents—led by Danny Osborne with John Duckitt and Chris Sibley. Out yesterday in
@NatRevPsych
1
29
106
As an interdisciplinary researcher who has always considered personality psych my home base, I am over-the-moon-happy to share that I've received the 2022 J.S. Tanaka Personality Dissertation Award from
@ARPtweets
. And it means I get to attend ARP (for the first time)!
14
3
107
I've been studying the psychology of authoritarianism in the political left (left-wing authoritarianism) for the last 3 yrs. Our latest preprint describes that work and proposes a new way of thinking about (and measuring) left-wing authoritarianism (1/?)
6
30
90
If you'd like to try the intervention described in this paper for yourself (i.e., an AI that provides factual counterpoints to conspiracy theories), try below. You'll be an early tester, so please forgive any bugs.
🚨WP🚨
Conspiracy beliefs famously resist correction, right?
WRONG: We show brief convos w GPT4 reduce conspiracy beliefs by ~20pp (d~1)!
🡆Tailored AI evidence rebut specific arguments offered by believers
🡆Effect lasts 2+mo
🡆Works on entrenched beliefs
48
244
762
4
20
90
E.g., communism, unlike fascism, proposes egalitarian solutions to social injustices, which may attract individuals who hope to alleviate these injustices. Accordingly, left-wing authoritarians may come to believe that authoritarianism is “a necessary evil.”
9/11
3
9
82
Notable differences between LWA and RWA/SDO also emerged. E.g., left-wing authoritarians were lower in dogmatism and cognitive rigidity, higher in negative emotionality, and expressed stronger support for a political system with substantial centralized state control.
5/11
3
15
83
!!!!!!! I defended! What a lovely thread. Thank you for your brilliant mentorship and guidance,
@ArberTasimi
Thrilled that my first graduate student, Tom Costello (
@tomstello_
), successfully defended his dissertation, "The Divergent Foundations of Social and Economic Political Ideology”!
A thread about Dr. Costello, who is pictured here w/ members of his committee...
4
2
77
12
2
83
New working paper! We put GPT4 from
@OpenAI
into
@Qualtrics
and used it to talk people out of conspiracy theories in an RCT.
@DG_Rand
's thread is fantastic -- we'll have even more details soon.
🚨WP🚨
Conspiracy beliefs famously resist correction, right?
WRONG: We show brief convos w GPT4 reduce conspiracy beliefs by ~20pp (d~1)!
🡆Tailored AI evidence rebut specific arguments offered by believers
🡆Effect lasts 2+mo
🡆Works on entrenched beliefs
48
244
762
3
12
81
We sincerely hope that our work will allow researchers to better understand authoritarianism in all of its forms.
11/11
9
5
77
Today I got to meet my name doppelgänger —
@tomcostellonbc
— AKA the reason I will never show up in google searches.
3
1
78
We compare LWA, RWA, and social dominance orientation across six samples and more than 60 criterion-related variables. We find they reflect a constellation of personality traits, cognitive features, and motivations that might be considered the “heart” of authoritarianism.
3/11
2
11
70
I was on
@NBCNightlyNews
tonight talking about our work (
@DG_Rand
@GordPennycook
) on countering conspiracies with AI.
Thanks for such a thoughtful interview,
@HallieJackson
.
Check it out!
Here’s the paper:
Meet Debunkbot. It may be able to change your mind when it comes to conspiracy theories.
@HallieJackson
tries it out.
10
13
20
7
16
71
Really happy to say that I'm one of the recipients of the SPSP Heritage Dissertation Research Award! Some sorely needed good news after a rough few weeks :)
6
4
68
We introduce a new conceptualization of left-wing authoritarianism (LWA) w/ three dimensions: anti-hierarchical aggression, top-down censorship, and anti-conventionalism.
We develop an LWA measure based on this conceptualization and conduct systematic tests of its validity.
2/
4
7
66
Now online: Before the 2020 election, Ariel Malka and I surveyed 6,000 Americans about authoritarianism and democracy. Opposition to democracy was rare, but many endorsed authoritarian actions that helped their preferred candidate or party.
1
25
64
Scott Lilienfeld on grant culture. This is such a great observation:
(Full text here: )
1
5
65
Our results raise the compelling possibility that a shared psychological core underlies different kinds of auth! But, prematurely describing LWA and RWA in this way risks blurring potentially important distinctions between them.
8/11
1
4
61
Exciting new preprint! We delve into the role of cognitive and emotional rigidity in political beliefs and extremism. But the question is - *which* kinds of political beliefs, extremism, and rigidity are interconnected? Collaborating with Irwin Waldman and
@GordPennycook
.
1
20
59
nbd
@hankgreen
thinks our paper is cool. in unrelated news, I've come around -- social media is totally very good for my brain and, also, all humanity.
🚨WP🚨
Conspiracy beliefs famously resist correction, right?
WRONG: We show brief convos w GPT4 reduce conspiracy beliefs by ~20pp (d~1)!
🡆Tailored AI evidence rebut specific arguments offered by believers
🡆Effect lasts 2+mo
🡆Works on entrenched beliefs
48
244
762
2
2
58
, the intervention described in this
@ScienceMagazine
paper, is now updated to include
@OpenAI
's gpt-o1 (aka strawberry)
only a minority of respondents will get gpt-o1, due to costs, but feel free to give it a spin and test your luck!
After an evidence-based dialogue with GPT-4 Turbo, an
#AI
model, conspiracy theorists substantially reduced their belief in favored conspiracies, and this newfound skepticism remained for months, according to a new Science study. Unraveling webs of conspiracy beliefs at scale has
52
135
383
4
10
57
Are social and economic political views psychologically dissimilar? In a new paper, authored with my late advisor, we find that the unique elements of social and economic ideology oppositely relate to numerous psychological constructs. Out now in SPPS!
3
8
55
announcing vegapunk:
our plug + go interface for online research involving interactions between AI models and human subjects. It’s like having a hyper-knowledgeable (research) confederate for your experiments and surveys.
4
13
57
We also found that LWA, RWA, and SDO predicted aggression towards people with different political views and helping towards people with similar views in a well-validated behavioral paradigm (with large effect sizes, i.e., roughly 10% of the total variance).
6/11
1
7
54
In a nationally representative sample, LWA was also correlated with participation in violence during the largely peaceful protests held in response to the killings of unarmed black americans by police/civilians in 2020.
7/11
2
4
52
Fascinating paper on violent extremism out today.
In ~70k terrorist attacks around the globe, right-wing attacks were ~2x as likely to be deadly than left-wing, but left-wing attacks were ~7.5x more frequent. Islamist acts most deadly + prevalent of all.
1
9
48
@PTetlock
Scott Lilienfeld taught almost this exact class for a year or two before he died. A fantastic class I was lucky to TA for. Sharing the syllabus:
2
9
51
The media blitz continues…now internationally!
Thanks for having me on
@BBCWorld
@CFraserBBC
Clip to follow tomorrow
2
4
50
Shared some new work on a general psychological account of authoritarianism at ISPP yesterday (and also talked with my hands a lot as usual).
Slides are here if anyone is curious:
2
0
46
An updated version of our AI-chatbot-conspiracy-intervention working paper is posted. The bigger changes in thread below:
(great explainer in Dave's thread + listen to me talk about the paper here: )
🚨WP🚨
Conspiracy beliefs famously resist correction, right?
WRONG: We show brief convos w GPT4 reduce conspiracy beliefs by ~20pp (d~1)!
🡆Tailored AI evidence rebut specific arguments offered by believers
🡆Effect lasts 2+mo
🡆Works on entrenched beliefs
48
244
762
2
11
45
Someone (on Twitter) just called me a morally stunted nerd who invented a bunch of astrology. Does that mean I’m finally a real personality psychologist?
4
2
45
Are conservatives highly dogmatic and certain, even 100% certain, about their beliefs? How about ideological extremists on both L and R? Yes. But it’s complicated. See more in my new paper with
@ShaunaBowes1
!
Now out in SPPS:
2
17
40
What are the (political) attitudinal correlates of anti-democratic attitudes? A few key findings from a new preprint led by Ariel Malka (with
@ChrisPolPsych
and me) based on 15 national (U.S.) surveys and two large experiments.
1
11
38
Although we knew it was coming, I’m truly heartbroken to hear this. Scott's mentorship has profoundly altered the course of my life and shaped the way I approach the world from a scientific and ethical perspective. I am amazingly lucky to have had him as an advisor and mentor.
It is with great sadness that I share the news that Scott Lilienfeld passed away last night. His wife Candice asked that I share the news with his professional community. The field has lost a giant, & more importantly, for many, a kind & thoughtful friend. My heart is heavy.
118
154
841
1
1
36
It's January 11th, and a strong contender for my favorite paper of 2023 has emerged.
If you want the cutting-edge in political psychology, read this paper! Kudos to the authors:
@mjbsp
,
@svallabhaa
, and Turner-Zwinkels.
1
9
33
Not enough people know that B.F. Skinner tried to develop a pigeon-guided missile system during WW2.
The pigeons were put in the missiles and pecked at the target to stay on track.
3
3
32
Be suspicious of claims about phenomena w/ low base rates that use percentages (e.g., our *thing* made *good outcome* increase 400%). This usually means something like 1 in 10,000 odds becoming 4 in 10,000. Sounds less important this way, right?
4
4
29
We're hiring in cognitive psychology
@AmericanU
-- join me in DC! The call is broad; please do apply if you're interested, but a bonus for experimental research touching on computation and/or politics.
Apps due Sept 15
More details:
0
13
28
This paper (led by
@ShaunaBowes1
) is jam packed with data relating to the psych of conspiracy beliefs (52 constructs spanning personality, motives, and psychopathology). We also hve a shiny app so u can take a gander under the tinfoil hat for yourself:
A project 4 years in the making! Thrilled to share results from our meta-analytic review (accepted at Psychological Bulletin) on the motivational and personological correlates of conspiratorial ideation
@ArberTasimi
@tomstello_
🧵
12
46
153
1
4
27
It sometimes feels like every three or four social-political psychology papers I read reference system justification theory. But this analysis, at least, suggests the experimental evidence for SJ is mostly (like, 75%) bunk. Big if true, as the kids say
2
1
26
These are gorgeous - would love to see someone do this with scientific figures rather than just journalism.
We just published our third Data Vis Dispatch with >20 of the best small and big
#dataviz
projects from last week:
Thanks to
@rosemintzers
for finding most of these gems, and thanks to newsrooms and individuals around the world for creating them! 👏👏👏
2
69
407
0
2
25
Many other mods are included, but in the end, we show that moving beyond the question of *if* conservatives are essentially rigid, to *when* and *why* they might or might not be, will provide a more realistic account of the psych. underpinnings of political opinions.
1
2
20
Just found out that a song I made back in college has 1 million plays !?!?! on Spotify. !?!
I’d say I should have kept at it but life/time is a garden of forking paths that my ADHD-self is not equipped to jump between fluidity. Lol.
3
0
22
For those wondering about my take on the Atlantic piece, this thread is an excellent guide to the literature on, and the status of, left-wing authoritarianism in behavioral science. I do have some qualms about the practice of using the parenting measure to assess LWA.
2
2
21
Full clip of my segment on AI + the psychology of conspiracies for The Context with
@CFraserBBC
@BBCNews
Paper discussed was in collaboration with
@DG_Rand
@GordPennycook
Segment starts around 18m!
0
3
21
In this new paper (!!), Chris Patrick and I use a genetic algorithm to construct abbreviated versions of the Left-wing Authoritarianism Index and evaluate them with multidimensional item response theory.
Just accepted @ Journal of Personality Assessment
1
2
20
Now officially available online:
People with authoritarian political views — whether they are on the far left or far right — have characteristically similar psychological profiles.
Our new JPSP paper takes a data-driven approach to understanding left-wing authoritarianism:
A thread:
1/
41
325
1K
1
2
20
New paper in press at American Politics Research! Ariel Malka and I examine support for different “flavors” of anti-democratic attitudes in the American public (n ~ 6k).
PS: I’m traveling, so more substantive thread to follow in a few days.
0
2
17
Our paper (very much led by
@ShaunaBowes1
) on conspiracy beliefs was one of the top ten most downloaded at Journal of Personality. Super cool!
@WileyPsychology
#TopDownloadedArticle
1
1
19
Twitter is…so bad now. It feels newly rudderless and bizarre. It’s not useful —for academics, at least—in the way that it was. Swamped by sticky, strange animus.
6
1
19
In light of the remarkable celebration of Scott that we've seen on Twitter over the last day and a half, it seems worth reposting this link for donations to support the Scott O. Lilienfeld APS Travel Award, which was established earlier this year.
0
9
18
A piece about one of my papers made the front page of reddit today (!?) and there seems to be a 50/50 split between comments saying that the findings are obvious and comments saying that the findings can’t possibly be true.
1
1
19
Here's the code:
NOTE -- it's not optimized and not a function; you'll have to modify it to fit your data. Hope this is helpful!
4
0
19
If you could press a button and radically improve the lives of 10,000 people in the far past* (1k+ yrs) or far future (1k+ yrs), which would you choose?
*assume the present is unchanged
Past people
185
Future people
243
13
9
16
New preprint:
We use machine learning (a genetic algorithm) to develop short (25- and 13-item) versions of our new left-wing authoritarianism measure (Costello et al., 2021, JPSP).
Spoiler alert: it worked surprisingly well!!
2
5
18
This was also my last collaboration with my late PhD advisor. It reflects much of what I learned from him, e.g., the importance of subjecting influential psychological theories to rigorous tests. Finally making this paper public means a lot to me. I hope you'll check it out!
2
1
17
I have a new paper out (with
@ShaunaBowes1
and our advisor, the twitterless Scott Lilienfeld). We look at whether and to what extent authoritarianism, need for certainty, and personality traits are related to different kinds of belief in determinism.
2
5
18
Most notably, whereas social conservatism (traditionalism) was robustly associated with rigidity, associations between economic conservatism (e.g., free-market) and rigidity indicators were inconsistent, small, and negative (but not significant) outside of the United States.
1
1
16
Considering that this literature vastly overrepresents American samples, I'll underscore the importance of this finding—a “rigidity-of-the-economic-right hypothesis” does not seem to hold much water beyond the United States.
1
1
16
E.g., correlations b/w economic conservatism and rigidity constructs were usually not significant. The point estimate for economic cons was negative (but not sig) outside of the US.
Social conservatism and rigidity, by contrast, were consistently positively correlated!
1
2
16
A provocative paper on the moral implications of consciousness research. I found the attached paragraph especially compelling.
“The scientific study of consciousness cannot, and should not, be morally neutral”, by
@lubianiker
and myself together with a power team of 9 courageous scientists, is finally out as a preprint. I’m *very* excited about this one. A Friday thread.
3
38
118
1
4
16
I have little doubt that the legacy of his scholarship will be great and long-lasting—skepticism and critical thinking are likely to be evergreen topics, as is personality—but his legacy among students and close colleagues will be one of warmth, empathy, and intellectual courage.
0
0
16
I found a lizard in my house—in the dead of winter. Cannot catch it for the life of me. New roommate?
3
1
15
Huh - social psych measures of authoritarianism (RWA/SDO) were tepid correlates of antisemitism, but anti-hierarchical aggression (introduced in our LWA paper) was antisemitism's most robust correlate. Shouldn't be taken as a "gotcha" (re ideological asymmetries), but (1/2)
New paper with
@dr_d_allington
&
@LouiseKatz59
@HSScomms
.
Press release
Antisemites are likely to
i. embrace conspiracy fantasies
ii. support repression against political opponents
iii. want to overthrow the social order.
0
8
8
1
0
15
@DeonTBenton
Not sure how helpful number of publications is as a metric. As an ABD this round, I had 8 first author peer reviewed papers, including a JPSP, and didn’t even get a zoom interview. Which is fine! But I think it’s about more than counting pubs.
1
0
15
New paper from
@Trent_Oll
: a canonical measure of authoritarianism doesn't predict Trump support (vs. other GOP candidates). One explanation for this finding IMO is that the "auth" measure ≠ a tendency to favor authoritarian things! See my quote tweet.
Shared some new work on a general psychological account of authoritarianism at ISPP yesterday (and also talked with my hands a lot as usual).
Slides are here if anyone is curious:
2
0
46
1
3
14
NYC is just yankee hotel foxtrot now
@LolOverruled
my friend sent me this and said you can just see everyone in other buildings staring out the windows and her office is passing out respirators
0
43
2K
1
0
14
My girlfriend put together this incredible quarantine birthday surprise for me - board games, whiskey, and cake. Here’s hoping 26 is a better year than 25! But, uh, TBD.
1
0
14
Nice write-up of one of our lab's recent papers at PsyPost:
1
7
12
Will be having a conversation with
@sbkaufman
this afternoon at 5 (EST) to talk about psychological science + human flourishing, live-streamed by . Should be both thought provoking and much fun!
Stream link:
3
3
13
I would not have gotten this job without the support and mentorship of my doctoral (
@ArberTasimi
and the late Scott Lilienfeld) and postdoctoral (
@GordPennycook
@DG_Rand
) advisors. As well as the advice and solidarity of my colleagues and friends.
3
0
13
Which personality traits are related to sexual objectification? Find out in our new paper, out this morning!
@ashleylwatts
@EmoryPsychology
@APA_Journals
1
4
13
I enjoyed writing this article outlining one of our recent papers — hope you’ll check it out!
A recent study from
@tomstello_
and
@ShaunaBowes1
mapped the spectrum of certainty about political beliefs among its participants: the results show more psychological commonalities than you might think
0
11
10
0
2
12
Just read this
@NYMag
story on anti-democratic attitudes in the Republican Party—and their dark implications. I don’t disagree.
Here’s what Ari Malka and I found in a recent paper on the prevalence of anti-democratic views in the American electorate. 🧵
1
3
13
Can authoritarianism exist in people who oppose established hierarchies of authority? Our work developing a measure of left-wing authoritarianism (across 5 samples and >6000 people) says yes. Poster to be presented at the SPSP pol. psych pre-conference: .
2
4
12
Super jazzed (pun intended) that my talk was accepted for
#SPSP2020
in New Orleans. I’ve really been looking forward to sharing this work. 🎷🎺
2
3
11
This looks fascinating! Can’t help but wonder about how it relates to the ideas in
@danwilliamsphil
recent (phenomenal) treatise on partisan motivated cognition. If we think others choose beliefs, how does that bear on how we treat members of rival partisan groups?
We often blame people for the beliefs they hold.
Does this mean we think people can choose their beliefs, like they can choose their actions?
We find children and adults generally think we can select our beliefs, but think we are restricted by evidence and morality.
Out now!
2
28
90
1
4
11
Just discovered that you can give yourself blue eyebrows and a green mustache using Zoom. My SPSP talk tomorrow is about to get super weird!!!!
0
0
12
Hot off the virtual press (by which I mean PsyArXiv)! Authoritarianism and allied traits are associated with a variety of deterministic beliefs in two large community samples (Total N = 20,489).
@ShaunaBowes1
, Scott Lilienfeld and I had fun writing this one.
@EmoryPsychology
"Escape from Freedom": Some Implications of Authoritarianism for Belief in Determinism
0
1
1
1
2
10