KarlPurcellBehavEcon
@ef110econ.bsky.social
Applying behavioural science to help make people happier, healthier, and wealthier with @irrationallabs. All views are personal. Retweets are not endorsements.
Reposted by KarlPurcellBehavEcon
This is all p misleading IMO. I just saw this link so haven't read it yet (but will, though I also think lots of the retweets of this haven't read it as well, given its provenance). I don't see how this counters (eg) Dawson 1999 which goes through 13 millennial cults and finds 12 show the effect 1/n
November 6, 2025 at 2:37 PM
This is all p misleading IMO. I just saw this link so haven't read it yet (but will, though I also think lots of the retweets of this haven't read it as well, given its provenance). I don't see how this counters (eg) Dawson 1999 which goes through 13 millennial cults and finds 12 show the effect 1/n
Reposted by KarlPurcellBehavEcon
How prevalent is misinformation really? Many detection methods rely on URL/Source-level, but they miss a big part of posts that don't include links.
In a new preprint from the group, led by @saminenno.bsky.social, we analyze 10x more posts from German politicians on four platforms.
thread below👇
In a new preprint from the group, led by @saminenno.bsky.social, we analyze 10x more posts from German politicians on four platforms.
thread below👇
November 3, 2025 at 5:06 PM
How prevalent is misinformation really? Many detection methods rely on URL/Source-level, but they miss a big part of posts that don't include links.
In a new preprint from the group, led by @saminenno.bsky.social, we analyze 10x more posts from German politicians on four platforms.
thread below👇
In a new preprint from the group, led by @saminenno.bsky.social, we analyze 10x more posts from German politicians on four platforms.
thread below👇
Reposted by KarlPurcellBehavEcon
Reposted by KarlPurcellBehavEcon
October 31, 2025 at 8:04 PM
Reposted by KarlPurcellBehavEcon
There are ways to address this problem with prompting and tooling (& more recent models do better in these tests), but current LLMs are pretty weak at dealing with time sequences where multiple documents (like court cases) have different time stamps and need to be understood in coherent sequence.
October 28, 2025 at 10:22 PM
There are ways to address this problem with prompting and tooling (& more recent models do better in these tests), but current LLMs are pretty weak at dealing with time sequences where multiple documents (like court cases) have different time stamps and need to be understood in coherent sequence.
Reposted by KarlPurcellBehavEcon
When people learn with ChatGPT instead of following their own searches, they end up knowing less, caring less, and producing worse advice, even when the facts are the same.
Friction is an essential ingredient for learning! Convenience makes us shallow.
academic.oup.com/pnasnexus/ar...
Friction is an essential ingredient for learning! Convenience makes us shallow.
academic.oup.com/pnasnexus/ar...
Experimental evidence of the effects of large language models versus web search on depth of learning
Abstract. The effects of using large language models (LLMs) versus traditional web search on depth of learning are explored. A theory is proposed that when
academic.oup.com
October 28, 2025 at 3:14 PM
When people learn with ChatGPT instead of following their own searches, they end up knowing less, caring less, and producing worse advice, even when the facts are the same.
Friction is an essential ingredient for learning! Convenience makes us shallow.
academic.oup.com/pnasnexus/ar...
Friction is an essential ingredient for learning! Convenience makes us shallow.
academic.oup.com/pnasnexus/ar...
Reposted by KarlPurcellBehavEcon
1. We ( @jbakcoleman.bsky.social, @cailinmeister.bsky.social, @jevinwest.bsky.social, and I) have a new preprint up on the arXiv.
There we explore how social media companies and other online information technology firms are able to manipulate scientific research about the effects of their products.
There we explore how social media companies and other online information technology firms are able to manipulate scientific research about the effects of their products.
October 24, 2025 at 12:47 AM
1. We ( @jbakcoleman.bsky.social, @cailinmeister.bsky.social, @jevinwest.bsky.social, and I) have a new preprint up on the arXiv.
There we explore how social media companies and other online information technology firms are able to manipulate scientific research about the effects of their products.
There we explore how social media companies and other online information technology firms are able to manipulate scientific research about the effects of their products.
Reposted by KarlPurcellBehavEcon
Can AI simulations of human research participants advance cognitive science? In @cp-trendscognsci.bsky.social, @lmesseri.bsky.social & I analyze this vision. We show how “AI Surrogates” entrench practices that limit the generalizability of cognitive science while aspiring to do the opposite. 1/
AI Surrogates and illusions of generalizability in cognitive science
Recent advances in artificial intelligence (AI) have generated enthusiasm for using AI simulations of human research participants to generate new know…
www.sciencedirect.com
October 21, 2025 at 8:24 PM
Can AI simulations of human research participants advance cognitive science? In @cp-trendscognsci.bsky.social, @lmesseri.bsky.social & I analyze this vision. We show how “AI Surrogates” entrench practices that limit the generalizability of cognitive science while aspiring to do the opposite. 1/
Reposted by KarlPurcellBehavEcon
I like this paper doi.org/10.1111/ajps...
American Journal of Political Science | MPSA Journal | Wiley Online Library
Manipulation checks are often advisable in experimental studies, yet they rarely appear in practice. This lack of usage may stem from fears of distorting treatment effects and uncertainty regarding w...
doi.org
October 21, 2025 at 9:24 PM
I like this paper doi.org/10.1111/ajps...
Reposted by KarlPurcellBehavEcon
Introducing SPEAK: A scalable computer-adaptive tool to measure knowledge of early human development, from Caroline Gaudreau, Dani Levine, John A. List, and Dana Suskind www.nber.org/papers/w34349
October 16, 2025 at 1:03 PM
Introducing SPEAK: A scalable computer-adaptive tool to measure knowledge of early human development, from Caroline Gaudreau, Dani Levine, John A. List, and Dana Suskind www.nber.org/papers/w34349
Reposted by KarlPurcellBehavEcon
🚨 out at @apsrjournal.bsky.social 🚨
➡️ We ran a large media literacy experiment to fight misinformation
➡️ 13,500 students, 583 villages in Bihar, India
➡️Created custom misinfo curriculum of 4 months
➡️Partnered w the government to roll it out as an official course in classrooms
hopeful findings👇🏽
➡️ We ran a large media literacy experiment to fight misinformation
➡️ 13,500 students, 583 villages in Bihar, India
➡️Created custom misinfo curriculum of 4 months
➡️Partnered w the government to roll it out as an official course in classrooms
hopeful findings👇🏽
October 15, 2025 at 3:33 PM
🚨 out at @apsrjournal.bsky.social 🚨
➡️ We ran a large media literacy experiment to fight misinformation
➡️ 13,500 students, 583 villages in Bihar, India
➡️Created custom misinfo curriculum of 4 months
➡️Partnered w the government to roll it out as an official course in classrooms
hopeful findings👇🏽
➡️ We ran a large media literacy experiment to fight misinformation
➡️ 13,500 students, 583 villages in Bihar, India
➡️Created custom misinfo curriculum of 4 months
➡️Partnered w the government to roll it out as an official course in classrooms
hopeful findings👇🏽
Reposted by KarlPurcellBehavEcon
This has been up for less than 2 weeks and is already my most read blog post of all 147 to date. It’s always the one you least suspect eh… open.substack.com/pub/julianki...
Applying Bradford Hill Criteria to observational evidence
Towards evidentiary criteria for causal inference in evaluation
open.substack.com
September 30, 2025 at 2:54 AM
This has been up for less than 2 weeks and is already my most read blog post of all 147 to date. It’s always the one you least suspect eh… open.substack.com/pub/julianki...
Reposted by KarlPurcellBehavEcon
Life satisfaction mostly declines with age. Previous findings (esp. the famous U-shaped age-SWB trajectory) were artifacts of misspecified models. doi.org/10.1093/esr/...
September 29, 2025 at 12:14 PM
Life satisfaction mostly declines with age. Previous findings (esp. the famous U-shaped age-SWB trajectory) were artifacts of misspecified models. doi.org/10.1093/esr/...
Reposted by KarlPurcellBehavEcon
Very happy to release this working paper today showing the value of AI in helping shape energy demand.
🚗 Can AI make EV charging cheaper and greener?
CNZ ran the world’s largest AI-managed EV charging trial with 13,000 UK households. Check out the results 👇
Summary & working paper 🔗 www.centrefornetzero.org/papers/ai-in...
CNZ ran the world’s largest AI-managed EV charging trial with 13,000 UK households. Check out the results 👇
Summary & working paper 🔗 www.centrefornetzero.org/papers/ai-in...
September 17, 2025 at 1:14 PM
Very happy to release this working paper today showing the value of AI in helping shape energy demand.
Reposted by KarlPurcellBehavEcon
Instructing LLMs may produce stable personality trait expression and correlations—as self-reported.
However, research by @pengruihan.bsky.social et al finds these traits rarely predict actual behaviour, which suggests LLMs having a personality is an illusion:
buff.ly/NhUiwEo
However, research by @pengruihan.bsky.social et al finds these traits rarely predict actual behaviour, which suggests LLMs having a personality is an illusion:
buff.ly/NhUiwEo
September 14, 2025 at 8:18 AM
Instructing LLMs may produce stable personality trait expression and correlations—as self-reported.
However, research by @pengruihan.bsky.social et al finds these traits rarely predict actual behaviour, which suggests LLMs having a personality is an illusion:
buff.ly/NhUiwEo
However, research by @pengruihan.bsky.social et al finds these traits rarely predict actual behaviour, which suggests LLMs having a personality is an illusion:
buff.ly/NhUiwEo
On my way back to Ireland for 10 days.
I will be splitting time between Kildare and Dublin. Haven't been back for 2 years so haven't seen a lot of people in a while, but would love to change that?
Let me know if you are around and want to meet up for a chat/pint.
I will be splitting time between Kildare and Dublin. Haven't been back for 2 years so haven't seen a lot of people in a while, but would love to change that?
Let me know if you are around and want to meet up for a chat/pint.
September 12, 2025 at 7:59 PM
On my way back to Ireland for 10 days.
I will be splitting time between Kildare and Dublin. Haven't been back for 2 years so haven't seen a lot of people in a while, but would love to change that?
Let me know if you are around and want to meet up for a chat/pint.
I will be splitting time between Kildare and Dublin. Haven't been back for 2 years so haven't seen a lot of people in a while, but would love to change that?
Let me know if you are around and want to meet up for a chat/pint.
Reposted by KarlPurcellBehavEcon
"In a large RCT on a credit card in Mexico, cutting rates and raising minimums barely reduced default. Job loss—more disruptive to cash flow—had far greater impact."
Recently acepted to #REStud, from Castellanos, Jiménez-Hernández, Mahajan, Prous, & Seira:
www.restud.com/contract-ter...
#econsky
Recently acepted to #REStud, from Castellanos, Jiménez-Hernández, Mahajan, Prous, & Seira:
www.restud.com/contract-ter...
#econsky
September 11, 2025 at 6:28 PM
"In a large RCT on a credit card in Mexico, cutting rates and raising minimums barely reduced default. Job loss—more disruptive to cash flow—had far greater impact."
Recently acepted to #REStud, from Castellanos, Jiménez-Hernández, Mahajan, Prous, & Seira:
www.restud.com/contract-ter...
#econsky
Recently acepted to #REStud, from Castellanos, Jiménez-Hernández, Mahajan, Prous, & Seira:
www.restud.com/contract-ter...
#econsky
Reposted by KarlPurcellBehavEcon
Just published in @jpube.bsky.social:
"Frictions in recovering unclaimed property: Evidence from a large-scale natural field experiment"
By Alejandro Zentner & Justin Holz
www.sciencedirect.com/science/arti...
#econsky #publiceconomics
"Frictions in recovering unclaimed property: Evidence from a large-scale natural field experiment"
By Alejandro Zentner & Justin Holz
www.sciencedirect.com/science/arti...
#econsky #publiceconomics
September 9, 2025 at 5:31 PM
Just published in @jpube.bsky.social:
"Frictions in recovering unclaimed property: Evidence from a large-scale natural field experiment"
By Alejandro Zentner & Justin Holz
www.sciencedirect.com/science/arti...
#econsky #publiceconomics
"Frictions in recovering unclaimed property: Evidence from a large-scale natural field experiment"
By Alejandro Zentner & Justin Holz
www.sciencedirect.com/science/arti...
#econsky #publiceconomics
Reposted by KarlPurcellBehavEcon
For anyone who wants to understand web search ranking and the role played by user behavioral data, the Google Search antitrust ruling from today is lucid and detailed in the section on Data Sharing Remedies. (And probably before.)
storage.courtlistener.com/recap/gov.us...
storage.courtlistener.com/recap/gov.us...
storage.courtlistener.com
September 2, 2025 at 11:58 PM
For anyone who wants to understand web search ranking and the role played by user behavioral data, the Google Search antitrust ruling from today is lucid and detailed in the section on Data Sharing Remedies. (And probably before.)
storage.courtlistener.com/recap/gov.us...
storage.courtlistener.com/recap/gov.us...
Reposted by KarlPurcellBehavEcon
Imagine a world where experimental results are increasingly based on AI-generated responses.
(Added ALT text)
(Added ALT text)
August 30, 2025 at 2:23 PM
Imagine a world where experimental results are increasingly based on AI-generated responses.
(Added ALT text)
(Added ALT text)
Reposted by KarlPurcellBehavEcon
Tour de force review on “Economics of Attention” by Loewenstein just published in J Econ Lit
@aeajournals.bsky.social
#behavioraleconomics
@aeajournals.bsky.social
#behavioraleconomics
August 29, 2025 at 11:24 PM
Tour de force review on “Economics of Attention” by Loewenstein just published in J Econ Lit
@aeajournals.bsky.social
#behavioraleconomics
@aeajournals.bsky.social
#behavioraleconomics
Reposted by KarlPurcellBehavEcon
Legislators & staffers have a bias toward thinking the public is evenly divided on policy issues, rather than a conservative bias in perception; prior findings were based on policy questions where the liberal position was more popular
adamzelizer.com/wp-content/u...
adamzelizer.com/wp-content/u...
adamzelizer.com
August 25, 2025 at 9:50 PM
Legislators & staffers have a bias toward thinking the public is evenly divided on policy issues, rather than a conservative bias in perception; prior findings were based on policy questions where the liberal position was more popular
adamzelizer.com/wp-content/u...
adamzelizer.com/wp-content/u...
Reading the great indoors by Emily anthes while outdoors. A book about how buildings & the design of spaces affect our behavior. The exact seat I picked is placed between two different steroes playing different music probably destroying my reading comprehension (and my enjoyment of the music)
August 15, 2025 at 9:12 PM
Reading the great indoors by Emily anthes while outdoors. A book about how buildings & the design of spaces affect our behavior. The exact seat I picked is placed between two different steroes playing different music probably destroying my reading comprehension (and my enjoyment of the music)
Natural experiment of looking at people who move from low walk ability to high walkabilty city finds that built environment can increase physical activity by up to 1100 steps per day!
www.nature.com/articles/s41...
www.nature.com/articles/s41...
Countrywide natural experiment links built environment to physical activity - Nature
By analysing the smartphone data of 2,112,288 participants, in particular observing and comparing the activity of the same individual in two different environments, we find that increases in the walka...
www.nature.com
August 14, 2025 at 10:55 AM
Natural experiment of looking at people who move from low walk ability to high walkabilty city finds that built environment can increase physical activity by up to 1100 steps per day!
www.nature.com/articles/s41...
www.nature.com/articles/s41...
Reposted by KarlPurcellBehavEcon
Come and work with us in BRU! We’re looking for a postdoc to join our team. Please share with anyone interested.
We are currently recruiting for the following position:
Postdoctoral Research Fellow - Behavioural Science
The successful candidate will design, programme and conduct original experiments tailored to the investigation decisions and behaviours of relevance to environmental,
Postdoctoral Research Fellow - Behavioural Science
The successful candidate will design, programme and conduct original experiments tailored to the investigation decisions and behaviours of relevance to environmental,
August 12, 2025 at 9:07 AM
Come and work with us in BRU! We’re looking for a postdoc to join our team. Please share with anyone interested.