Tessera
Folgen

Zoe Purcell 🕵🏻‍♀️

@zoepurcell.bsky.social

Assistant professor working on human reasoning and the human-AI interaction at The American University of Paris and LaPsyDÉ, l'Université Paris-Cité. For biography and publications: https://zoepurcell.org/

16 Posts246 Follower163 Folgt
Zoe Purcell 🕵🏻‍♀️@zoepurcell.bsky.social · 3w ago

How can we collectively contain the threat of dishonest delegation to AI? Our paper "Whistleblowers can contain the unethical externalities of human–AI delegation" is out in PNAS, with @nckobis.bsky.social , Andrew Samuel, and @jfbonnefon.bsky.social. www.pnas.org/doi/10.1073/...

PNAS

Proceedings of the National Academy of Sciences (PNAS), a peer reviewed journal of the National Academy of Sciences (NAS) - an authoritative source of high-impact, original research that broadly spans...

pnas.org

Nils Köbis@nckobis.bsky.social · 7mo ago

Many people worry that AI-assisted writing feels “less authentic” and may erode trust. So we tested it in two preregistered, incentivized trust-game experiments (N = 1,637), which just got published in iScience (Cell Press) More details 👇

Bild
Jérémie Beucler@jeremiebeucler.bsky.social · 8mo ago

Our paper with @zoepurcell.bsky.social, @luciecharlesneuro.bsky.social, and @wimdeneys.bsky.social has been accepted at Behavior Research Methods! 🥳 Here is the updated preprint: osf.io/preprints/ps... Also, the baserater package is now on CRAN: cran.r-project.org/package=base... #psynomBRM

OSF

osf.io

Jérémie Beucler@jeremiebeucler.bsky.social · 10mo ago

1/10 🚨 New preprint: Using Large Language Models to Estimate Belief Strength in Reasoning 🚨 When asked: "There are 995 politicians and 5 nurses. Person 'L' is kind. Is Person 'L' more likely to be a politician or a nurse?", most people will answer "nurse", neglecting the base-rate info. A 🧵👇

Abstract Accurately quantifying belief strength in heuristics-and-biases tasks is crucial yet methodologically challenging. In this paper, we introduce an automated method leveraging large language models (LLMs) to systematically measure and manipulate belief strength. We specifically tested this method in the widely used “lawyer-engineer” base-rate neglect task, in which stereotypical descriptions (e.g., someone enjoying mathematical puzzles) conflict with normative base-rate information (e.g., engineers represent a very small percentage of the sample). Using this approach, we created an open-access database containing over 100,000 unique items systematically varying in stereotype-driven belief strength. Validation studies demonstrate that our LLM-derived belief strength measure correlates strongly with human typicality ratings and robustly predicts human choices in a base-rate neglect task. Additionally, our method revealed substantial and previously unnoticed variability in stereotype-driven belief strength in popular base-rate items from existing research, underlining the need to control for this in future studies. We further highlight methodological improvements achievable by refining the LLM prompt, as well as ways to enhance cross-cultural validity. The database presented here serves as a powerful resource for researchers, facilitating rigorous, replicable, and theoretically precise experimental designs, as well as enabling advancements in cognitive and computational modeling of reasoning. To support its use, we provide the R package baserater, which allows researchers to access the database to apply or adapt the method to their own research.
Jérémie Beucler@jeremiebeucler.bsky.social · 10mo ago

1/10 🚨 New preprint: Using Large Language Models to Estimate Belief Strength in Reasoning 🚨 When asked: "There are 995 politicians and 5 nurses. Person 'L' is kind. Is Person 'L' more likely to be a politician or a nurse?", most people will answer "nurse", neglecting the base-rate info. A 🧵👇

Abstract Accurately quantifying belief strength in heuristics-and-biases tasks is crucial yet methodologically challenging. In this paper, we introduce an automated method leveraging large language models (LLMs) to systematically measure and manipulate belief strength. We specifically tested this method in the widely used “lawyer-engineer” base-rate neglect task, in which stereotypical descriptions (e.g., someone enjoying mathematical puzzles) conflict with normative base-rate information (e.g., engineers represent a very small percentage of the sample). Using this approach, we created an open-access database containing over 100,000 unique items systematically varying in stereotype-driven belief strength. Validation studies demonstrate that our LLM-derived belief strength measure correlates strongly with human typicality ratings and robustly predicts human choices in a base-rate neglect task. Additionally, our method revealed substantial and previously unnoticed variability in stereotype-driven belief strength in popular base-rate items from existing research, underlining the need to control for this in future studies. We further highlight methodological improvements achievable by refining the LLM prompt, as well as ways to enhance cross-cultural validity. The database presented here serves as a powerful resource for researchers, facilitating rigorous, replicable, and theoretically precise experimental designs, as well as enabling advancements in cognitive and computational modeling of reasoning. To support its use, we provide the R package baserater, which allows researchers to access the database to apply or adapt the method to their own research.
Wim De Neys@wimdeneys.bsky.social · last yr.

New preprint: “Folk Thinking, Fast and Slow: Intuitive Preference for Deliberation in Humans and Machines” Pop culture often praises intuition (“Blink”, Steve Jobs). But do we really trust it? Across 13 studies, we find a strong intuitive preference for deliberation. tinyurl.com/8r54dmyn (1/6)

Bild