Tom Griffiths' Computational Cognitive Science Lab. Studying the computational problems human minds have to solve.
Princeton, NJ
Joined August 2020
- Tweets223
- Following171
- Followers6.8K
- Likes112
Pinned Tweet
Excited to announce a new book telling the story of mathematical approaches to studying the mind, from the origins of cognitive science to modern AI! The Laws of Thought will be published in February, and is available for pre-order now.
Griffiths Computational Cognitive Science Lab retweeted
Princeton is seeking an AI Postdoctoral Research Fellow in Natural and Artificial Minds. An amazing program! Deadline is Feb. 28
apply.interfolio.com/194534
Griffiths Computational Cognitive Science Lab retweeted
🧵[1/N] Are large language models more creative than humans at combining ideas? 🤔
Combinatorial creativity involves combining familiar ideas in unfamiliar ways, and is central to scientific discovery and technological innovation.
In fact, breakthrough scientific work is often associated with more distant combinations of methods, techniques, and concepts.
Here, we introduce Kombine, a new benchmark of combinatorial creativity for LLMs.
TLDR:
- LLMs are more combinatorially creative than the average human.
- However, conceptual blending is the hardest task for LLMs, and even frontier models struggle.
- LLMs also tend to homogenize, inventing concepts with the same properties across separate models.
- Moreover, humans and LLMs combine in distinct regions of semantic space.
- As a result, we anticipate greater conceptual diversity can be unlocked through human-AI collaboration. 🗣️
Thread below🧵⬇️
Griffiths Computational Cognitive Science Lab retweeted
Have we built machines that think like humans? 🧠🤖
In a new preprint, we propose CogGym, which compares AI and human judgments across 258 commonsense-reasoning experiments from 100 cognitive-science papers.
Paper: arxiv.org/abs/2609.21259
Platform: coggym.org
ALT CogGym workflow diagram: research experiments are standardized in EML by an LLM agent with human review, then rendered for human participants and computational models. Replicated and original human data are compared with model responses in a shared analysis.
Griffiths Computational Cognitive Science Lab retweeted
LATTE is accepted to NeurIPS 2026! ☕️
Adding agents is easy. Coordinating is hard. LLM teams often duplicate work, overwrite each other, and stall: classic distributed-systems failures.
We borrowed the fix: a shared task graph that agents build, claim, and revise as they go 👇
Griffiths Computational Cognitive Science Lab retweeted
Can we make an 🤖 AI system 🤖 safer 🛡️ by adding more agents? Not necessarily ❌
In our new paper, we ask a simple question: as AI teams get bigger, what happens when more of the agents are trying to mislead the rest? 😈
We find that:
- Defection rates increase ~linearly with the proportion of deceivers 📈
- You can’t just outnumber the attackers by scaling up: larger teams weren’t more resistant when the fraction of deceptive agents stayed the same
- Multi-agent systems are vulnerable even when deceivers remain a minority! Within humans, misleading participants had much weaker effects until they formed a majority
- Surprisingly, allowing deceivers to coordinate 🤝 results in less defection
For agent swarms, the key question isn’t just “how many agents do I have?”, but instead, “who makes up the swarm?”
More agents is not necessarily better. If the adversary scales with you, they can keep up
You have to spend effort making your agents accurate 🎯. The adversary just needs to dump enough plausible slop 🗑️ to sway the group
Thanks so much to @cocosci_lab and @karthik_r_n for advising! Had a lot of fun co-leading this project with @Jasin_Cekinmez and @Michel_Liao!
📄 Check out the paper! arxiv.org/abs/2609.30028
Griffiths Computational Cognitive Science Lab retweeted
LLMs have positional biases — but only sometimes.
Same evidence, fed bit-by-bit: no bias.
All at once: bias, in the opposite direction from humans.
In court, that's guilty vs not guilty.
📄 arxiv.org/pdf/2608.12387
w/ @addisonwu_ & @cocosci_lab
Accepted @ EMNLP 2026 Findings🇭🇺!
Griffiths Computational Cognitive Science Lab retweeted
We’re excited to announce DiG-bench, a new benchmark for discovery!
Over the last few weeks we’ve been testing frontier AI models on our novel discovery games and seeing how they score. Each game is a text-based environment, so they probe discovery capabilities in the natural domain of language models, rather than requiring additional, potentially confounding, visual understanding.
TL;DR frontier models have improved a lot over the last few months. But they are still stumped by some surprisingly simple problems, even in their native text domain.
With
@cocosci_lab (@Princeton) @MITCoCoSci (@MIT) @SchmidhuberAI (@KAUST_News) @misovalko (@Inria) @tri_dao (@PrincetonCS) @RMBattleday @zebkDotCom @FraserGreenlee @akaijsa @ClareMaguire @TimMuller1 @kubicek_ales @physicscat0x7d @SukritSumant @thoughtchannel_
(1/5)
Griffiths Computational Cognitive Science Lab retweeted
What comes to mind when you hear “bolt”?
🔩, ⚡, or 🏃?
We found AI settles on fewer meanings than people do, especially when generating an image compared to a sentence.
Oral 🗣️ @COLM SciFM—presenting with @addisonwu_, @RajaMarjieh, @cocosci_lab!
arxiv.org/abs/2608.00410
Exciting new paper on how people choose what to do when playing games they have never seen before, a very different challenge from learning how to make better moves in a familiar game.
Games have long served as a testing ground for understanding problem-solving and strategic reasoning in humans and machines, across many fields – from AI, to cognitive science, economics, and more. Yet, in life people are regularly faced not just with repeated instances of the same problem or “system of rule and reward”, but entirely new problems, new systems, new games.
And when encountering new games, people not only need to decide what actions to take but whether a game is even worth playing at all!
How can people think reasonably about new games for the first time, so flexibly and quickly?
We take steps towards this question in our new work, out now in @Nature!
Apple: podcasts.apple.com/us/podcas…
Spotify: open.spotify.com/show/7rhwBG…
If you are heading to the Cognitive Science Society conference next week, you might enjoy the most recent episodes of the Cognition Project podcast: Donald Norman, Eleanor Rosch, and George Lakoff talk about the first time the conference was held (among many other things!)
Large language models can memorize patterns they see in text, but that backfires when a task deviates from a common pattern. We demonstrate this phenomenon using riddles: when something looks like a riddle but has a simple answer AI systems make surprising mistakes.
New paper! w/@cocosci_lab 🧵Can large language models reason flexibly, or have they learned what reasoning looks like? We introduce a new paradigm to test this question—the riddle riddle—and find that humans and LLMs show opposite patterns of performance. 📜
New preprint explores how using AI to discover psychological theories and propose experiments to test them can create an automated cognitive scientist
1/ 🚨 New preprint: "Closing the Loop to Discover Psychological Theories with an Automated Cognitive Scientist"
Co-led w/ Younes Strittmatter
Co-mentored by @suyoghc and @cocosci_lab
In collaboration w/ @kachergis, @norijacoby, @nathanieldaw, & @cpilab
Introducing AutoCog — a fully autonomous AI system that runs the entire scientific discovery cycle in cognitive science to surface novel theories of human behavior 🧵
#CognitiveScience #AI4Science #LLMs
ALT AUTOCOG runs as a closed discovery loop iterated over Ncycles: Experimental Design → Behavioral Data Collection → Analysis & Arbitration → Theory Revision , after which a revised theory re-enters the loop.
Griffiths Computational Cognitive Science Lab retweeted
Are there limits to what AI will contribute to human mathematics? In a new paper with @LakeBrenden @cocosci_lab we probe this question with a simpler one: Can a language model discover 0? The resounding answer (so far) is no.
arxiv.org/abs/2606.17289
Probabilistic models give us a way to characterize optimal inductive inference. Using the output of those models to fine-tune large language models improves the inductive reasoning capabilities of AI systems.
1/7
🚨 New preprint : Using Probabilistic Programs to Train Inductive Reasoning in Large Language Models
👥 in collaboration with/ @akjagadish, @lakebrenden, and @cocosci_lab
Most LLM reasoning post-training focuses on deductive reasoning problems: tasks with single verifiable answers.
However, much of real-world reasoning is inductive: sparse evidence, latent causes, uncertain beliefs.
Examples: forecasting, diagnosis, scientific inference, or everyday judgment.
Current LLMs perform poorly on inductive reasoning. How do we remedy that?
Griffiths Computational Cognitive Science Lab retweeted
Just published in @PNASNews, we resolve a 50-year-old riddle from Richard Feynman's handwritten notes, prove and generalize it, and run a large-scale human study to reveal near-optimal heuristics in sequential decision problems:
doi.org/10.1073/pnas.2509612…
Griffiths Computational Cognitive Science Lab retweeted
Excited to share that we will be presenting this as an oral at ICML 2026! 🥳
Griffiths Computational Cognitive Science Lab retweeted
Excited to share our new paper using cognitive science to distinguish AI agents and humans!
We administered CogCAPTCHA30, a set of 30 cognitive tasks, to frontier VLMs (GPT-5, Sonnet 4.5, Gemini 2.5 Pro) and humans. We found that processes differ between AI agents and humans - even when the final output is identical.
Link: arxiv.org/abs/2605.06524
This work was led by @milenamr7 and co-authored with @cocosci_lab, and @mayankagrawal
Griffiths Computational Cognitive Science Lab retweeted
🚨 What if your LLM team could dynamically reorganize itself mid-task? Introducing LATTE: a framework that gives teams a ~caffeine~ boost via an evolving task graph they build and adapt in real time, achieving SOTA performance with fewer tokens and conflicts. New paper 🧵👇
Griffiths Computational Cognitive Science Lab retweeted
A computational theory of Aha! moments! New paper explaining why Aha! moments occur and why they feel so good. TLDR: They are a form of meta-cognitive prediction errors i.e. they occur when we surprise ourselves about our own abilities!
@cocosci_lab
sciencedirect.com/science/ar…
