Do Large Language Models Know What Humans Know?
Sean Trott, Cameron Jones, Tyler Chang, James Michaelov, Benjamin Bergen
July 2023 Cogn SciSynopsis of Social media discussions
The posts collectively reflect interest and support for the research, with examples like references to the False Belief Task and phrases such as 'comparing human and GPT-3 performance' suggest deep engagement. Overall, the tone and content emphasize curiosity and acknowledgment of the study's importance in advancing understanding of AI and human cognition.
Agreement
Moderate agreementMost discussions show support for the idea that large language models like GPT-3 can differentiate mental states, although some acknowledge limitations.
Interest
Moderate level of interestPosts express curiosity and enthusiasm about the research, especially highlighted by comments on the significance of the findings.
Engagement
Moderate level of engagementParticipants reference specific experiments like the False Belief Task and discuss the implications, demonstrating active engagement.
Impact
Moderate level of impactThe discussions emphasize the potential influence of this research on AI understanding and cognitive science, indicating moderate impact.
Social Mentions
YouTube
2 Videos
Bluesky
1 Posts
1 Posts
49 Posts
News
18 Articles
Metrics
Video Views
137,291
Total Likes
2,567
Extended Reach
265,456
Social Features
71
Timeline: Posts about article
Top Social Media Posts
Cameron Jones
Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cognitive Science https://t.co/L0PowtiMBS @cogsci_soc @Sean_Trott @jamichaelov @tylerachang
view full post
Tilbe Göksun
Does GPT-3 have a theory of mind?
Check out the blog published in CogSci Blogs based on this paper:
https://t.co/79HXlCKVSl https://t.co/kWeXlzLEpY
view full post
Akira Murakami
"while statistical learning from language exposure may in part explain how humans develop the ability to reason about the mental states of others, other mechanisms are also responsible."
Trott et al. (2023). Do Large Language Models Know What Humans Know?
https://t.co/5h8FNTRImk
view full post
Philosophy of Futures Studies
"Do Large Language Models Know What Humans Know?" | @Sean_Trott, @camrobjones, Tyler Chang, @jamichaelov, Benjamin Bergen | #AI #Philosophy #FuturesStudies #ArtificialIntelligence #Cognition (1/n)
https://t.co/tT2PgqzY4s https://t.co/CpPFYgm7bJ
view full post
Uğurcan Vurgun, PhD, MA
Do Large Language Models Know What Humans Know?
by Sean Trott, Cameron Jones, Tyler Chang, James Michaelov, Benjamin Bergen
https://t.co/NAOBmDIYWe
view full post
Posts referencing the article
Assessing Large Language Models' Understanding of Human Beliefs and Mental States
This study investigates whether large language models like GPT-3 comprehend human beliefs by testing them with a linguistic false belief task. Results show models differentiate mental states but still lag behind human understanding, suggesting multiple mechanisms contribute to understanding beliefs.
August 13, 2023
137,169 views
Why Artificial Intelligence Does Not Truly Understand Human Beliefs
Artificial intelligences like Chat GPT and Gemini generate responses based on statistical probabilities rather than genuine understanding or awareness. They do not possess true intelligence or reasoning capabilities, highlighting their limitations in comprehending human beliefs and mental states.
May 23, 2025
122 views
-
RT @mustafaergen_: Reading List for the Interested: https://t.co/o572chfhuD https://t.co/D58GFGCJOm https://t.co/f57yvMAHTG https://t.co/o…
view full postFebruary 9, 2025
1
-
Mustafa Ergen
@mustafaergen_ (Twitter)Reading List for the Interested: https://t.co/o572chfhuD https://t.co/D58GFGCJOm https://t.co/f57yvMAHTG https://t.co/oytipstWr3 https://t.co/maH2nNdkRT https://t.co/iUkRMpOYdb https://t.co/xOPpGQ5yhl https://t.co/0EZBtfBCLk https://t.co/SUHWDy3K9m https://t.co/aPbcxjgiBw
view full postFebruary 6, 2025
1
1
-
liwei
@sakura_liwei (Twitter)【人工智能】万字通俗讲解大语言模型内部运行原理 | LLM | 词向量 | Transformer | 注意力机制 | 前馈网络 | 反向传播 ... https://t.co/wHCuI8rTOl via @YouTube
view full postJune 5, 2024
-
H|ψ⟩ Quantum Finance
@haiyunsky (Twitter)【人工智能】万字通俗讲解大语言模型内部运行原理 | LLM | 词向量 | Transformer | 注意力机制 | 前馈网络 | 反向传播 ... https://t.co/RbI0hHJaW8 via @YouTube
view full postMarch 7, 2024
-
Yunus Şahin
@herrbokologist (Twitter)RT @GoksunTilbe: Does GPT-3 have a theory of mind? Check out the blog published in CogSci Blogs based on this paper: https://t.co/79HXlCK…
view full postFebruary 27, 2024
8
-
Nazli Altin, PhD
@NzlAltin (Twitter)RT @GoksunTilbe: Does GPT-3 have a theory of mind? Check out the blog published in CogSci Blogs based on this paper: https://t.co/79HXlCK…
view full postFebruary 27, 2024
8
-
Ferah Şenaydın
@Ferahsenaydin (Twitter)RT @GoksunTilbe: Does GPT-3 have a theory of mind? Check out the blog published in CogSci Blogs based on this paper: https://t.co/79HXlCK…
view full postFebruary 27, 2024
8
-
Eren Günseli
@eren_gunseli (Twitter)RT @GoksunTilbe: Does GPT-3 have a theory of mind? Check out the blog published in CogSci Blogs based on this paper: https://t.co/79HXlCK…
view full postFebruary 27, 2024
8
-
Jian Wang
@jwang3417 (Twitter)【人工智能】万字通俗讲解大语言模型内部运行原理 | LLM | 词向量 | Transformer | 注意力机制 | 前馈网络 | 反向传播 ... https://t.co/P4oX8T1HsI via @YouTube
view full postJanuary 11, 2024
-
Sean Trott
@Sean_Trott (Twitter)@ericzelikman @noahdgoodman @mcxfrank One more ToM paper in case you're interested: https://t.co/vRrFblYtNg
view full postDecember 27, 2023
2
-
Tech Robot
@_tech_robot_ (Twitter)-Do Large Language Models Know What Humans Know? | #AIMadeSimple #AI https://t.co/8HJBY1liNh
view full postNovember 26, 2023
-
Alp Aytuglu
@alp_aytuglu (Twitter)RT @GoksunTilbe: Does GPT-3 have a theory of mind? Check out the blog published in CogSci Blogs based on this paper: https://t.co/79HXlCK…
view full postOctober 5, 2023
8
-
Abdullah Topraksoy
@abd_topraksoy (Twitter)RT @GoksunTilbe: Does GPT-3 have a theory of mind? Check out the blog published in CogSci Blogs based on this paper: https://t.co/79HXlCK…
view full postOctober 4, 2023
8
-
Dr Astrid Zeman
@astrid_zeman (Twitter)RT @GoksunTilbe: Does GPT-3 have a theory of mind? Check out the blog published in CogSci Blogs based on this paper: https://t.co/79HXlCK…
view full postOctober 3, 2023
8
-
CogSci Society
@cogsci_soc (Twitter)RT @GoksunTilbe: Does GPT-3 have a theory of mind? Check out the blog published in CogSci Blogs based on this paper: https://t.co/79HXlCK…
view full postOctober 3, 2023
8
-
Tilbe Göksun
@GoksunTilbe (Twitter)Does GPT-3 have a theory of mind? Check out the blog published in CogSci Blogs based on this paper: https://t.co/79HXlCKVSl https://t.co/kWeXlzLEpY
view full postOctober 3, 2023
33
8
-
Kindergarten Head
@Amormaid (Twitter)LLM https://t.co/L7xJs6VMAq
view full postSeptember 14, 2023
-
Daryl Costello
@DarylC71 (Twitter)Do Large Language Models Know What Humans Know? - Trott - 2023 - Cognitive Science - Wiley Online Library https://t.co/LwTqQJIv0y
view full postAugust 18, 2023
-
alexandroslee創造價值夥伴VCP李東昇
@alexandroslee (Twitter)【人工智能】萬字通俗講解大語言模型內部運行原理 | LLM | 詞向量 | Transformer | 注意力機制 | 前饋網絡 | 反向傳播 | 心智理論 語言大模型內部究竟是如何工作的？本文用最少的數學知識和術語進行解釋。 https://t.co/7G3xa8Twfp
view full postAugust 13, 2023
-
Cameron Jones
@camrobjones (Twitter)@Starcourse @binarybits @JohnRentoul @tylercowen Agree, and we make a similar point here (https://t.co/L0PowtiMBS) and here (https://t.co/Euc3FaFLPP). Other great work testing the robustness of LLM ToM!https://t.co/ojzH0Wpi6j @NatalieShapira @MaartenSap ; https://t.co/UmlHiRapT7 @gandhikanishk @jphilippfranken @tobigerstenberg
view full postJuly 30, 2023
-
Uğurcan Vurgun, PhD, MA
@ugurcan_vurgun (Twitter)Do Large Language Models Know What Humans Know? by Sean Trott, Cameron Jones, Tyler Chang, James Michaelov, Benjamin Bergen https://t.co/NAOBmDIYWe
view full postJuly 24, 2023
2
-
Sean Trott
@Sean_Trott (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 15, 2023
21
-
Philosophy of Futures Studies
@philofutures (Twitter)"Do Large Language Models Know What Humans Know?" | @Sean_Trott, @camrobjones, Tyler Chang, @jamichaelov, Benjamin Bergen | #AI #Philosophy #FuturesStudies #ArtificialIntelligence #Cognition (1/n) https://t.co/tT2PgqzY4s https://t.co/CpPFYgm7bJ
view full postJuly 10, 2023
2
-
Chris ☕
@mixingchris.bsky.social (Bluesky)Do large language models know that humans know? https://onlinelibrary.wiley.com/doi/epdf/10.1111/cogs.13309
view full postJuly 10, 2023
-
Chris
@MixingChris (Twitter)Do large language models know that humans know? https://t.co/Xn6uShnnFI
view full postJuly 10, 2023
-
SHIRÅSUNA, Ⅿ.
@mitoop_quiz (Twitter)【ちょっと関連?】 偶然にもつい先日、こんな論文も出ていました https://t.co/dsk4vKbB72 (こちらについても、活動報告で少し触れています)
view full postJuly 7, 2023
-
Petra B. Schumacher
@pb_schumacher (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 6, 2023
21
-
Ryo Maie
@MaieRyo (Twitter)RT @mrkm_a: "while statistical learning from language exposure may in part explain how humans develop the ability to reason about the menta…
view full postJuly 6, 2023
3
-
Adam Safron
@adamsafron (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 5, 2023
21
-
nasakti
@Homodigitalist (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 5, 2023
21
-
Jesus Ramirez-Bermudez
@JRBneuropsiq (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 5, 2023
21
-
sisyfos
@zKsisyfos (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 5, 2023
21
-
Shun
@Spectrum_cj (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 5, 2023
21
-
Mr. Flood
@hongyifan20 (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 5, 2023
21
-
cecilia böcking
@cecilia_bocking (Twitter)RT @mrkm_a: "while statistical learning from language exposure may in part explain how humans develop the ability to reason about the menta…
view full postJuly 5, 2023
3
-
Dr Shiyu He
@ShiyuHe1 (Twitter)RT @mrkm_a: "while statistical learning from language exposure may in part explain how humans develop the ability to reason about the menta…
view full postJuly 5, 2023
3
-
Akira Murakami
@mrkm_a (Twitter)"while statistical learning from language exposure may in part explain how humans develop the ability to reason about the mental states of others, other mechanisms are also responsible." Trott et al. (2023). Do Large Language Models Know What Humans Know? https://t.co/5h8FNTRImk
view full postJuly 5, 2023
8
3
-
Diverse Intelligences Summer Institute #DISI
@DivIntelligence (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 5, 2023
21
-
Pelin Soydan
@pelinatte (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 5, 2023
21
-
Ana Chkhaidze
@chkhaidze_ (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 5, 2023
21
-
Eric Leonardis @ NeurIPS 2025
@EricLeonardis (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 5, 2023
21
-
ɐɯıɾɐW ıɥso⅄
@yoshi_mjm (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 4, 2023
21
-
Daniel Horschler
@DanielHorschler (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 4, 2023
21
-
James Michaelov @ NeurIPS2025
@jamichaelov (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 4, 2023
21
-
Nithin George
@george_nit (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 4, 2023
21
-
Felix Binder
@flxbinder (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 4, 2023
21
-
The Great Indian Drishti
@GrtIndicDrishti (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 4, 2023
21
-
Martin Zettersten
@m_zettersten (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 4, 2023
21
-
CogSci Society
@cogsci_soc (Twitter)RT @camrobjones: Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cogniti…
view full postJuly 4, 2023
21
-
Cameron Jones
@camrobjones (Twitter)Really excited that our paper comparing human and GPT-3 performance at the False Belief Task has been published in Cognitive Science https://t.co/L0PowtiMBS @cogsci_soc @Sean_Trott @jamichaelov @tylerachang
view full postJuly 4, 2023
101
21
-
Design, User Experience, and Usability: Marcelo M. Soares ...
LNCS 13309, Universal Access in Human-Computer Interaction: User and Context ... doi.org/10.1111/1468-2427.00075 39. Statista.: Leading European city ...
view full postDecember 15, 2025
News
-
HCI International 2022 - Late Breaking Papers: Design, User ...
LNCS 13309, Universal Access in Human-Computer Interaction: User and Context ... 10.1111/j.1479-8425.2004.00137.x 17. Gil, S., Teissèdre, F., Chambres, P ...
view full postDecember 15, 2025
News
-
Antimicrobial Resistance: Factors To Findings: Vijay Soni Ajay ...
https://doi.org/10.1111/j.1365-2958.1993.tb01111.x. 37. Eldholm V ... 2022;12(1):13309. 10. Murray CJ, et al. Global burden of bacterial antimicrobial ...
view full postDecember 15, 2025
News
-
Design, User Experience, and Usability: Design For Emotion, Well ...
LNCS 13309, Universal Access in Human-Computer Interaction: User and Context ... 10.1111/j.1471-1842.2010.00880.x. Prensky, M.: Digital natives, digital ...
view full postDecember 15, 2025
News
-
Lies, Damned Lies, and Distributional Language Statistics ...
Cog- Patrik Bartak, and Samuel F. Brown. 2023b. nitive Science, 47(7):e13309. Tall Tales at Different Scales: Evaluating Scaling Trends For Deception in ...
view full postDecember 15, 2025
News
-
Beyond The Conservatory Model | PDF | Curriculum | Brand
com/atar-arad-orchestra/, accessed September 3, 2018. Beyond Beauty, Brilliance, and Expression 17 3. Like cogs ... doi: 10.1111/j.1465-3435.2006.00255.x ...
view full postDecember 15, 2025
News
-
Data Scopus2020 | PDF | Toxicology | Pharmacology
... cog/ta/2020/00000025/00000001/art00005;jsessionid=nae demand in Indonesia: A and ... 10.1111/aji.13296 checkpoints CTLA-4, Sri Ratna Dwiningsih, Arif ...
view full postDecember 15, 2025
News
-
Indigenous Peoples, Marine Space and Resources, and ...
Constellations Volume 15 issue 4 2008 [doi 10.1111%2Fj.1467-8675.2008.00510.x] Jürgen Habermas -- The Constitutionalization of International Law and the ...
view full postDecember 15, 2025
News
-
Oksitosin | PDF | Pituitary Gland | Hypothalamus
... 10.1111/j.1471- 0528.1999.tb08112.x. 31. Arletti R, Benelli A, Bertolini A ... cog- 917. Scheele D, Striepens N, Güntürkün O, Deutschländer S, Maier W ...
view full postDecember 15, 2025
News
-
Cengizhan & Ucar- 2023 - The Effect of Mindfulness‐Based Sexual ...
Mindfulness and cog- Psychiatr Care 2018;54(3):416-421. doi:10.1111/ppc ... 2018;97(4):380-387. Laeger. 2019;181(18):V12180853. doi:10.1111/aogs.13309 ...
view full postDecember 15, 2025
News
-
ICPCG-Chapter-8-2024 | PDF | Glycated Hemoglobin | Diabetes
Current glycemic target rec- Optimal glycemia is associated with improved cog- ... doi.org/10.1111/pedi.13309 practice consensus guidelines 2022: manag- 69 ...
view full postDecember 15, 2025
News
-
2022 Book HCIInternational2022Posters | PDF | Computer Graphics ...
LNCS 13309, Universal Access in Human-Computer Interaction: User and Context ... 2(4), 716–724. https://doi.org/10.1111/j.1756-8765.2010.01115.x 4 ...
view full postDecember 15, 2025
News
-
Language Model | PDF | Language Mechanics | Syntax
Cognitive (EMNLP-IJCNLP), pages 5831–5837. Science, 47(7):e13309. https://doi.org https://doi.org/10.18653/v1/D19-1592 /10.1111/cogs.13309, PubMed: 37401923 ...
view full postDecember 15, 2025
News
-
Are Language Models More Like Libraries or Like Librarians | PDF ...
mans know? Cognitive Science, 47(7):e13309. doi.org/10.1177/17456916231201401, https://doi.org/10.1111/cogs.13309 ...
view full postDecember 15, 2025
News
-
human positron emission tomography study of intranasal insulin in ...
doi: 10.1111/acel.12498. 8. Guo Z, Chen Y, Mao YF, et al. Long-term ... doi: 10.3233/jad-2008-13309. 19. Reger MA, Watson GS, Green PS ...
view full postMay 27, 2025
News
-
Re-evaluating TOM Evaluation in Large Language Models | PDF ...
Do Large Language Models Know What Humans Know? Cognitive Science, 47(7):e13309, July 2023. ISSN 0364-0213. doi: 10.1111/cogs.13309. URL https ...
view full postFebruary 28, 2025
News
-
At Boys Town, we focus on six areas of research, including hearing ...
... cogs.advising@ubc.ca :) The UBC Social Health Lab: Smartphones and ... com/doi/10.1111/aogs.13309 Kudos Mr Agata!! Power on making your ...
view full postJuly 19, 2024
Facebook
-
Artificial Intelligence Design | Medium
Do Large Language Models Know What Humans Know? https://onlinelibrary.wiley.com/doi/pdf/10.1111/cogs.13309 [9] Scammers use AI to mimic ...
view full postDecember 5, 2023
News
-
通俗解构语言大模型的工作原理|上下文|向量|字母|工作原理|神经网络 ...
com/doi/full/10.1111/cogs.13309）的那样，成功的表现可能归因于任务中的混淆因素——一种“聪明汉斯（clever Hans，指一匹名为汉斯的马看似能完成一些 ...
view full postAugust 1, 2023
News
Abstract Synopsis
- The study explores whether large language models like GPT-3 understand human beliefs by testing them with a linguistic version of the False Belief Task, comparing their responses to those of humans.
- Results show that GPT-3 can differentiate between different mental states and perform above chance, but humans still outperform the model, indicating that language exposure alone isn't enough to fully explain humans' ability to understand others' beliefs.
- This suggests that humans likely develop their understanding of mental states through multiple mechanisms, not just exposure to language, highlighting the role of innate or other experiential factors beyond statistical learning from language data.]
Gulenber
@Gulenber5 (Twitter)