@GreenLab15

Green Lab @Georgetown — Investigating the Cognitive Science of Creativity

 Washington, D.C.
cng.georgetown.edu/home
Joined February 2020
  • Tweets20
  • Following43
  • Followers24
2Media
More evidence of LLM homogeneity on creativity tasks. Even across different LLM models/families, their outputs are more similar to each other than human responses are, lacking semantic diversity. @yoed_kenett arxiv.org/abs/2501.19361
6
32
945
I am delighted to announce that my book, The Creativity Choice, is now available for preorder. The book asks how to transform your creative ideas into something tangible. And it provides what you need to know to make it so #Creativity #CreativityUnleashed #bookstoread
1
9
31
869
🚨 Publication alert! Student #creativity contributes to #educational performance, beyond GPA and #personality Our (@JamesKaufman @dhcropley @_sarahrezaei Tim Patston) research with @ACARAeduau #NAPLAN data is now out! Full text: shorturl.at/ZS24X
1
9
1
19
728
Five years in the making: our meta-analysis of 10 fMRI datasets across Asia, Europe & N America (N=2,433). We find creative ability emerges from dynamic switching between default and executive brain networks, with optimal creativity at a "sweet spot." nature.com/articles/s42003-0…

Dynamic switching between brain networks predicts creative ability

Communications Biology - Robust evidence that creativity is tied to the capacity to dynamically switch between brain networks supporting spontaneous and controlled cognition.

nature.com
5
48
5
133
11,134
New study finds enhanced creativity in autistic adults is linked to co-occurring ADHD rather than autism itself (N=352). psycnet.apa.org/fulltext/202…
3
14
60
4,289
The Society for the Neuroscience of Creativity (@tsfnc) is coming to Europe (Paris) for the first time! Abstracts are due Feb. 1. We welcome neuroscience, behavioral, and computational work. tsfnc.org/sfnc2025-creativit…
1
13
2
41
3,719
New study found that creativity predicts standardized educational outcomes beyond GPA and conscientiousness. @JamesKaufman sciencedirect.com/science/ar…
6
9
699
.@Moshe_Glickman and @affectivebrain reveal a human-AI feedback loop, where AI amplifies subtle human biases, which are then further internalized by humans. This cycle increases human bias over time across domains. nature.com/articles/s41562-0…

How human–AI feedback loops alter human perceptual, emotional and social judgements

Nature Human Behaviour - Glickman and Sharot reveal a human–AI feedback loop, where AI amplifies subtle human biases, which are then further internalized by humans. This cycle, observed...

nature.com
22
2
81
5,403
Meta-analysis (n=52 studies, 195 effect sizes) confirms the incubation effect: taking breaks away from a problem leads to more creative solutions (d=0.39) psycnet.apa.org/record/2025-…

Exploration of impact of incubation on creativity: A meta-analysis.

The present study employed a multivariate meta-analysis technique to investigate the influence of incubation on creative tasks, examining the role of moderators in their relationship. Analyses with...

psycnet.apa.org
1
28
3
118
7,491
Study by @will_orwig et al. finds people who are more verbally creative can produce more creative generative AI prompts and Dalle images—the creatively rich get richer with AI? tandfonline.com/doi/abs/10.1…

Using AI to Generate Visual Art: Do Individual Differences in Creativity Predict AI-Assisted Art...

As artificial intelligence (AI) advances in the realm of generative art, a critical question emerges: does human creativity matter? That is, do more-creative people produce more-creative AI-assiste...

tandfonline.com
1
7
32
2,357
A comprehensive review of research on creativity in AI---from artistic creation to scientific discovery---grounded in creativity theories. arxiv.org/abs/2410.17218

Creativity in AI: Progresses and Challenges

Creativity is the ability to produce novel, useful, and surprising ideas, and has been widely studied as a crucial aspect of human cognition. Machine creativity on the other hand has been a...

arxiv.org
7
31
2,024
Preprint 🚨: Each marginal human-written essay adds 2-8x more unique semantic content than LLM essays, revealing systematic homogenization in LLM outputs. Analysis of 2,200 college essays using BERT embedding distances. osf.io/preprints/psyarxiv/8p… #NLP #LLM
1
5
6
668
We introduce 'diversity growth rate' - a novel metric quantifying marginal contribution of each additional text to collective semantic diversity. Method: Bootstrap sampling sentences, computing BERT embeddings, measuring cosine distances between word pairs.
1
57
Key finding: While parameter-modified GPT-4 achieved higher individual semantic diversity (DSI) vs humans (d=2.66), its diversity growth rate remained significantly lower (b=-0.25, p<.001). Chain-of-thought prompting showed similar patterns.
47
New paper on human-AI interaction. We hire 18 writers to edit quirks in AI writing & see if #AI can mimic this process to improve its own writing Verdict: Writer-edited > AI-edited > AI-generated In other words:🚨Edits enhance alignment in writing🚨 🔗arxiv.org/pdf/2409.14509
4
41
6
167
31,348
🚨 Calling all Creativity Researchers! We are seeking existing data for an international, community-driven project on automated creativity scoring.
1
5
2
16
5,651
The Laboratory for Relational Cognition at Georgetown, directed by Adam Green, is seeking a postdoctoral scientists to lead a collaborative project to implement "closed-loop" TMS-EEG brain stimulation experiments in creativity and executive function!
1
3
4
804
For more information on the lab, see our website. With any questions, please contact Maggie Shideler (ms5324@georgetown.edu).
106
Green Lab retweeted
CLS
@ChengleiSi9 Sep 2024
Automating AI research is exciting! But can LLMs actually produce novel, expert-level research ideas? After a year-long study, we obtained the first statistically significant conclusion: LLM-generated ideas are more novel than ideas written by expert human researchers.
94
743
276
3,568
1,124,762
Ever watched a cup teeter on the edge of a table? Our research shows that young children, adults, and machines (DNNs) judge an object's physical stability by geometric cues, especially the geometric centroids. Full-text: rdcu.be/dwpwz. @LourencoLab @vayzenberg90
Lourenco Lab@LourencoLab19 Jan 2024
Led by @Yaxin_Liu, out today in Scientific Reports--Using psychophysics and computational modeling, we show that adults and children rely on object geometry when judging stability. Mental stimulation (àla a physics engine) appears unnecessary rdcu.be/dwpwz @vayzenberg90
2
13
1,520
Load more