arXiv subjects

Nori Jacoby

Publications and source records attributed to Nori Jacoby.

At least 19 recordsLinked to original sources

arXiv2026

Sparks of In Silico Cognitive Science: Theories from Simulated Data Can Generalize to Humans

Behavioral foundation models have been proposed as stand-ins for human participants across settings, but it is unclear whether theories discovered on them generalize to humans or merely characterize the simulator. We ran the Automated Cognitive Scientist (\textsc{AutoCog}), a closed-loop discovery system in which LLM agents design theory-discriminating experiments, collect responses, arbitrate between competing theories, and synthesize successors, entirely on behavior simulated by Centaur, a foundation model of human behavior. In a multi-attribute decision-making setting, the theories \textsc{AutoCog} found on Centaur generalized to human data: they outperformed canonical theories on ten held-out experiments and were rivaled only by theories found by running the same loop on people. We argue that this succeeds despite the simulator's inevitable imperfections because a discovery loop that arbitrates between competing theories demands less of its simulator than estimation does. The simulator only needs to capture the regularities that distinguish the theories, and not necessarily reproduce behavior precisely. Imperfect simulators can therefore widen the search over theories, with human data then testing whether the surfaced theories generalize.

cs.AI↗

arXiv2026

Closing the Loop to Discover Psychological Theories with an Automated Cognitive Scientist

Across the sciences, autonomous systems are increasingly being used in closed-loop discovery, proposing new theories and designing and running experiments to test them. This approach is yet to be applied in the field of cognitive science, where the central bottleneck is theory-building: the creative step of turning the accumulated failures of existing models into better ones. Theory generation has remained manual even as data collection, modeling, and experiment design have been automated. We present the Automated Cognitive Scientist (AutoCog), a fully autonomous agentic-AI system that closes this loop. Large-language-model agents advocate competing theories, each expressed as an executable cognitive model, design experiments that best discriminate them, collect behavioral data from participants recruited online, score theories against collected data based on their generative performance, diagnose why they fail, and synthesize a better successor. Repeating this cycle allows them to search the space of theories, models, and experiments. In the domain of decision-making, AutoCog recovered known decision-making strategies from simulated behavior, including unconventional ones, showing that its discoveries are ultimately driven by the data rather than strictly bound by the priors of the underlying language models. When run with human participants, it produced theories that outperformed the established theories it was seeded with and generalized to held-out studies in two different experimental settings. It also surfaced a novel theory of multi-cue decision-making in which choices show diminishing sensitivity to feature values. The distinctive predictions of this theory were confirmed in a preregistered study with new participants. AutoCog demonstrates how an automated discovery system can be used to turn cognitive theory-building into an explicit, executable, and cumulative science.

q-bio.NC↗

arXiv2026

An Experimental Method to Study Opinion Diffusion in Human-AI Hybrid Societies

As artificial intelligence increasingly mediates public discourse, it becomes important to understand how human-AI collectives shape opinion formation, deliberation, and democratic outcomes. We present a novel experimental method for studying opinion dynamics in hybrid human-AI social networks. Participants, human or AI, were embedded in $5\times5$ grid lattice networks and iteratively asked to select and revise statements on a given polarizing topic over eight rounds. We compared three conditions: human-only, AI-only, and hybrid networks with equal proportions of human and AI participants. Hybrid human-AI networks achieved the lowest final polarization while, in contrast, human-only networks exhibited higher polarization with lower neighbor agreement. We also ran additional experiments varying Large Language Model (LLM) prompt framing to explore whether instruction design might influence convergence patterns. Although these early findings are preliminary and cannot yet support broad generalizations, they highlight the potential value of experimental social networks for understanding opinion dynamics in human-AI hybrid societies.

cs.SI↗

arXiv2026

DUET: Optimize Token-Budget Allocation for Reinforcement Learning with Verifiable Rewards

Reinforcement learning with verifiable rewards (RLVR) generates hundreds of thousands of tokens per training step, with rollout generation dominating the computational cost. The overall token budget can be controlled along two main dimensions: (i) deciding which prompts to allocate rollouts to, and (ii) deciding how long each rollout should be. Prior work has generally controlled only one of these dimensions at a time. We show that jointly tuning both decisions under a shared compute budget improves both reasoning quality and wall-clock training time. We instantiate this view as \textbf{DU}al-controlled tok\textbf{E}n alloca\textbf{T}ion (DUET), a computationally efficient layer over GRPO that uses a lightweight pre-rollout surrogate of prompt informativeness to set how many rollouts each prompt receives, and a marker-gated abort rule with importance reweighting to set when to stop them. On Qwen3-1.7B trained on MATH, DUET outperforms full-budget GRPO and the other three budget-aware baseline methods. DUET's advantage further generalizes to other benchmarks across math and coding, and is on par with the best baseline on the scientific Q\&A domain, while also achieving a $1.62\times$ wall-clock speedup. More notably, using only 50\% of the token budget, DUET still outperforms all baseline methods at their full budget, achieving an even higher $2.51\times$ speedup over full-budget GRPO. We verify the high performance of DUET on other backbone LLMs, including Qwen3-4B and Llama-3.2-3B-Instruct. Notably, the gap between DUET and the strongest baseline \emph{widens} as the budget tightens, contrary to the usual pattern in which efficient methods trade off quality as compute decreases. More broadly, these results suggest that DUET budget-aware control strategies are valuable not only for accelerating training, but also for improving the quality of the learning signal.

cs.LG↗

arXiv2026

Do Melody and Rhythm Coevolve?

Music comprises two core structural components, melody and rhythm, that vary widely across cultures. Whether these components coevolve in a coupled way or follow independent trajectories remains unclear. We introduce a novel computational pipeline to extract vocal melodic pitch-interval and percussive inter-onset timing distributions from 27,628 popular songs across 59 countries, enabling large-scale cross-cultural comparison that bypasses traditional music annotations. Musical similarities between countries aligned with geographic and linguistic relationships, validating our approach. Substantial variation emerged in both melodic and rhythmic structures across countries, yet the diversity of the two components was not significantly correlated, challenging assumptions of coupled evolution. Only rhythmic diversity was significantly associated with ethnic and linguistic heterogeneity, while melodic diversity showed no such association. These findings suggest that melody and rhythm constitute partially independent systems shaped by distinct cultural and evolutionary pressures, rather than components of a single monolithic musical style.

cs.SD↗

arXiv2026

Active Inference with People: a general approach to real-time adaptive experiments

Adaptive experiments optimize their design throughout data collection, which can bring substantial benefits compared to conventional experimental settings. Potential applications include, among others, computerized adaptive testing (when selecting informative tasks in ability measurements), adaptive treatment assignment (when searching for experimental conditions maximizing certain outcomes), and active learning (when choosing optimal training data for machine learning algorithms). However, implementing these techniques in real time poses substantial computational and technical challenges. In this paper, we introduce a practical and unified approach to real-time adaptive experiments that can encompass these scenarios across textual, visual, and audio tasks. Our strategy combines active inference, a Bayesian framework inspired by cognitive neuroscience, with Pyro, a probabilistic programming library, and PsyNet, a modular Python package for large-scale online behavioral experiments. Active inference provides a task-agnostic optimization objective and efficient inference strategies; probabilistic programming makes the computations practical, reducing implementation costs; and PsyNet makes the resulting procedure deployable with humans in real time across diverse behavioral paradigms. We illustrate this approach through two concrete examples: (1) an adaptive testing experiment estimating participants' ability by selecting optimal challenges, reducing the number of trials required by 30--40\%; and (2) an adaptive treatment assignment strategy that identifies the optimal treatment up to three times as accurately as a fixed design. We provide instructions to facilitate adoption of the workflow.

stat.ME↗

arXiv2026

Why Human Guidance Matters in Collaborative Vibe Coding

Writing code has been one of the most transformative ways for human societies to translate abstract ideas into tangible technologies. Modern AI is changing this process by enabling experts and non-experts alike to generate code without actually writing it, instead using natural language instructions or "vibe coding". While increasingly popular, the impact of vibe coding on productivity and collaboration, and the role of humans in this process, remains unclear. Here, we introduce a controlled experimental framework for studying collaborative vibe coding and use it to compare human-led, AI-led, and hybrid groups. Across 20 experiments involving 737 human participants, we show that people provide uniquely effective high-level instructions for vibe coding, whereas AI-provided instructions often result in performance collapse. We further demonstrate that hybrid systems perform best when humans lead by providing instructions while evaluation is delegated to AI. Although AI systems can rapidly optimize performance for specific tasks, our work highlights the importance of human guidance in shaping future hybrid societies.

cs.HC↗

arXiv2026

Popularity Feedback Constrains Innovation in Cultural Markets

Real-world creative processes ranging from art to science rely on social feedback-loops between selection and creation. Yet, the effects of popularity feedback on collective creativity remain poorly understood. We investigate how popularity ratings influence cultural dynamics in a large-scale online experiment where participants ($N = 1\,008$) iteratively \textit{select} images from evolving markets and \textit{produce} their own modifications. Results show that exposing the popularity of images reduces cultural diversity and slows innovation, delaying aesthetic improvements. Popularity feedback is associated with changes to both selection and creative stages. During selection, popularity information triggers cumulative advantage, with participants preferentially building upon popular images, reducing diversity. During creation, participants make less disruptive changes, and are more likely to expand existing visual patterns. Feedback loops in cultural markets thus not only shape selection, but also, directly or indirectly, the form and direction of cultural innovation.

cs.SI↗

arXiv2026

Human-AI Synergy Supports Collective Creative Search

Generative AI is increasingly transforming creativity into a hybrid human-artificial process, but its impact on the quality and diversity of creative output remains unclear. We study collective creativity using a controlled word-guessing task that balances open-endedness with an objective measure of task performance. Participants attempt to infer a hidden target word, scored based on the semantic similarity of their guesses to the target, while also observing the best guess from previous players. We compare performance and outcome diversity across human-only, AI-only, and hybrid human-AI groups. Hybrid groups achieve the highest performance while preserving high diversity of guesses. Within hybrid groups, both humans and AI agents systematically adjust their strategies relative to single-agent conditions, suggesting higher-order interaction effects, whereby agents adapt to each other's presence. Although some performance benefits can be reproduced through collaboration between heterogeneous AI systems, human-AI collaboration remains superior, underscoring complementary roles in collective creativity.

cs.SI↗

arXiv2025

Using LLMs to Advance the Cognitive Science of Collectives

LLMs are already transforming the study of individual cognition, but their application to studying collective cognition has been underexplored. We lay out how LLMs may be able to address the complexity that has hindered the study of collectives and raise possible risks that warrant new methods.

q-bio.NC↗

arXiv2025

GlobalMood: A cross-cultural benchmark for music emotion recognition

Human annotations of mood in music are essential for music generation and recommender systems. However, existing datasets predominantly focus on Western songs with terms derived from English, which may limit generalizability across diverse linguistic and cultural backgrounds. We introduce 'GlobalMood', a novel cross-cultural benchmark dataset comprising 1,180 songs sampled from 59 countries, with large-scale annotations collected from 2,519 individuals across five culturally and linguistically distinct locations: U.S., France, Mexico, S. Korea, and Egypt. Rather than imposing predefined emotion and mood categories, we implement a bottom-up, participant-driven approach to organically elicit culturally specific music-related emotion terms. We then recruit another pool of human participants to collect 988,925 ratings for these culture-specific descriptors. Our analysis confirms the presence of a valence-arousal structure shared across cultures, yet also reveals significant divergences in how certain emotion terms (despite being dictionary equivalents) are perceived cross-culturally. State-of-the-art multimodal models benefit substantially from fine-tuning on our cross-culturally balanced dataset, particularly in non-English contexts. Broadly, our findings inform the ongoing debate on the universality versus cultural specificity of emotional descriptors, and our methodology can contribute to other multimodal and cross-lingual research.

cs.IR↗

arXiv2025

The Dynamics of Collective Creativity in Human-AI Hybrid Societies

Generative AI is shaping an increasingly hybrid society, where ideas and cultural artefacs are created both by humans and intelligent machines. Human creativity is influenced in complex, nonlinear ways by the actions of AI-driven agents within their social networks, but these influences are difficult to measure using traditional methods. This study examines how human-AI interactions shape the evolution of collective creation within large-scale social network experiments, where human and AI participants collectively create stories. Participants (either humans or AI) joined 5x5 grid-based networks in which stories were selected, modified, and shared over many iterations. Initially, AI-only networks showed greater creativity (rated by a separate group of human raters) and collective diversity of stories than human-only and human-AI networks. However, over time, hybrid human-AI networks became more diverse in their creations than AI-only networks. In part, this is because AI agents retained little from the original stories, while human-only networks preserved continuity. These findings highlight the value of experimental social networks in understanding human-AI hybrid societies.

cs.SI↗

arXiv2025

Reasoning within and between collective action problems

Understanding cooperation in social systems is challenging because the ever-changing rules that govern societies interact with individual actions, resulting in intricate collective outcomes. In virtual-world experiments, we allowed people to make changes in the systems that they are making decisions within and investigated how they weigh the influence of different rules in decision-making. When choosing between worlds differing in more than one rule, a naive heuristics model predicted participants decisions as well, and in some cases better, than game earnings (utility) or by the subjective quality of single rules. In contrast, when a subset of engaged participants made instantaneous (within-world) decisions, their behavior aligned very closely with objective utility and not with the heuristics model. Findings suggest that, whereas choices between rules may deviate from rational benchmarks, the frequency of real time cooperation decisions to provide feedback can be a reliable indicator of the objective utility of these rules.

q-bio.NC↗

arXiv2025

How constraints on editing affects cultural evolution

When is it beneficial to constrain creativity? Creativity thrives with freedom, but when people collaborate to create artifacts, there is tension between giving individuals freedom to revise, and protecting prior achievements. To test how imposing constraints may affect collective creativity, we performed cultural evolution experiments where participants collaborated to create melodies and images in chains. With melodies, we found that limiting step size (number of musical notes that can be changed) improved pleasantness ratings for created tunes. Similar results were observed in cohorts of musicians, and with different selection regimes. In contrast, limiting step size in creating images consistently reduced pleasantness. These conflicting findings suggest that in domains such as music, where artifacts can be easily damaged, and where evolutionary outcomes are hard to foresee, collective creativity may benefit from imposing small step sizes. We discuss parallels with search algorithms and the evolution of conservative birdsong cultures.

q-bio.NC↗

arXiv2025

Visual and Auditory Aesthetic Preferences Across Cultures

Research on how humans perceive aesthetics in shapes, colours, and music has predominantly focused on Western populations, limiting our understanding of how cultural environments shape aesthetic preferences. We present a large-scale cross-cultural study examining aesthetic preferences across five distinct modalities extensively explored in the literature: shape, curvature, colour, musical harmony and melody. We gather 401,403 preference judgements from 4,835 participants across 10 countries, systematically sampling two-dimensional parameter spaces for each modality. The findings reveal both universal patterns and cultural variations. Preferences for shape and curvature cross-culturally demonstrate a consistent preference for symmetrical forms. While colour preferences are categorically consistent, ratio-like preferences vary across cultures. Musical harmony shows strong agreement in interval relationships despite differing regions of preference within the broad frequency spectrum, while melody shows the highest cross-cultural variation. These results suggest that aesthetic preferences emerge from an interplay between shared perceptual mechanisms and cultural learning.

cs.MM↗

arXiv2025

Characterizing the Interaction of Cultural Evolution Mechanisms in Experimental Social Networks

Understanding how cognitive and social mechanisms shape the evolution of complex artifacts such as songs is central to cultural evolution research. Social network topology (what artifacts are available?), selection (which are chosen?), and reproduction (how are they copied?) have all been proposed as key influencing factors. However, prior research has rarely studied them together due to methodological challenges. We address this gap through a controlled naturalistic paradigm whereby participants (N=2,404) are placed in networks and are asked to iteratively choose and sing back melodies from their neighbors. We show that this setting yields melodies that are more complex and more pleasant than those found in the more-studied linear transmission setting, and exhibits robust differences across topologies. Crucially, these differences are diminished when selection or reproduction bias are eliminated, suggesting an interaction between mechanisms. These findings shed light on the interplay of mechanisms underlying the evolution of cultural artifacts.

cs.SI↗

arXiv2025

Are Expressions for Music Emotions the Same Across Cultures?

Music evokes profound emotions, yet the universality of emotional descriptors across languages remains debated. A key challenge in cross-cultural research on music emotion is biased stimulus selection and manual curation of taxonomies, predominantly relying on Western music and languages. To address this, we propose a balanced experimental design with nine online experiments in Brazil, the US, and South Korea, involving N=672 participants. First, we sample a balanced set of popular music from these countries. Using an open-ended tagging pipeline, we then gather emotion terms to create culture-specific taxonomies. Finally, using these bottom-up taxonomies, participants rate emotions of each song. This allows us to map emotional similarities within and across cultures. Results show consistency in high arousal, high valence emotions but greater variability in others. Notably, machine translations were often inadequate to capture music-specific meanings. These findings together highlight the need for a domain-sensitive, open-ended, bottom-up emotion elicitation approach to reduce cultural biases in emotion research.

cs.CL↗

arXiv2025

Institutional Preferences in the Laboratory

Getting a group to adopt cooperative norms is an enduring challenge. But in real-world settings, individuals don't just passively accept static environments, they act both within and upon the social systems that structure their interactions. Should we expect the dynamism of player-driven changes to the "rules of the game" to hinder cooperation -- because of the substantial added complexity -- or help it, as prosocial agents tweak their environment toward non-zero-sum games? We introduce a laboratory setting to test whether groups can guide themselves to cooperative outcomes by manipulating the environmental parameters that shape their emergent cooperation process. We test for cooperation in a set of economic games that impose different social dilemmas. These games vary independently in the institutional features of stability, efficiency, and fairness. By offering agency over behavior along with second-order agency over the rules of the game, we understand emergent cooperation in naturalistic settings in which the rules of the game are themselves dynamic and subject to choice. The literature on transfer learning in games suggests that interactions between features are important and might aid or hinder the transfer of cooperative learning to new settings.

cs.SI↗