Fengyuan Shirley Yang

Hi! I am a sophomore at Yuanpei College, Peking University. I am broadly interested in cognitive science(s), which I pursue through the Tong Class, an AI-centered honors program, and the Thought and Society program, an interdisciplinary program integrating humanistic interpretation and social-scientific methods to study classical intellectual traditions and contemporary social issues.

I am deeply grateful to be advised by Prof. Yixin Zhu and to be part of the PKU Cognitive Reasoning (CoRe) Lab. I have also greatly benefited from the guidance of Prof. Yanchao Bi, Dr. Qian Wang, Dr. Huichao Yang, and Prof. Chi Zhang.

I am thrilled to join the Comparative Cognition Lab (CCL) at UC San Diego in Summer 2026, where I will have the opportunity to be advised by Prof. Federico Rossano.

I explore how language and cognitive tools shape—and are shaped by—how people think and communicate. I am particularly interested in the trade-offs and biases these systems introduce, and how they influence human reasoning and social interaction in real-world contexts. I aim to combine computational modeling with experimental and observational approaches to study behavior in ecologically valid contexts.

Publications

CogSci 2026

Rational Communication Shapes Morphological Composition

Fengyuan Yang, Yongqian Peng, Yuxi Ma, Chenheng Xu, and Yixin Zhu

In Proceedings of the 48th Annual Meeting of the Cognitive Science Society, 2026

Abs PDF Website

Human languages expand vocabularies by combining existing morphemes rather than inventing arbitrary forms. Communicative efficiency shapes lexical systems at multiple levels (Gibson et al., 2019), yet morphological composition—combining morphemes through compounding or affixation—has rarely been modeled as a historically situated speaker choice among competing morpheme sequences, leaving unanswered why a language settles on one morpheme combination over other plausible alternatives. We ask whether a trade-off between listener recoverability and speaker production cost can predict attested compositions over contemporaneously available alternatives. Here we show, within the Rational Speech Act (RSA) framework (Frank & Goodman, 2012; Goodman & Frank, 2016) using a time-indexed lexicon constructed from Corpus of Historical American English (COHA) and Corpus of Contemporary American English (COCA), that across 4323 naturally occurring English compounds and derivations spanning 1820–2019, attested compositions are systematically ranked above unattested alternatives generated from contemporaneously available morphemes. Models integrating semantic informativeness with production cost outperform semantic-only and cost-only baselines on Mean Reciprocal Rank (MRR) and top-k accuracy (Acc@k), with the advantage of the Pragmatic Speaker model (𝑆1) over the semantic-only baseline growing as the candidate set expands, where meaning alone leaves morphological choice underdetermined. These findings suggest that lexicalization reflects a communicative trade-off between expressiveness and efficiency, extending rational accounts of communication from utterance-level choice to the internal structure of words.
CHI 2026

NarrativeLoom: Enhancing Creative Storytelling through Multi-Persona Collaborative Improvisation

Yuxi Ma^*, Yongqian Peng^*, Fengyuan Yang, Siyu Zha, Chi Zhang, Zixia Jia, Zilong Zheng, and Yixin Zhu

In Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems, 2026

Abs PDF Video Website

Large Language Models show promise for AI-assisted storytelling, yet current tools often generate predictable, unoriginal narratives. To address this limitation, we present NarrativeLoom, a multi-persona co-creative system grounded in Campbell’s Blind Variation and Selective Retention theory. NarrativeLoom deploys specialized AI personas to generate diverse narrative options (blind variation), while users act as creative directors to select and refine them (selective retention). We designed a controlled study with 50 participants and found that stories co-authored with NarrativeLoom were not only perceived by users as more novel and diverse but were also objectively rated by experts as significantly better across all Torrance Test creativity dimensions: fluency, flexibility, originality, and elaboration. Stories are significantly longer with richer settings and more dialogue. Writing expertise emerged as a moderator: novices benefited more from structured scaffolding. This demonstrates the value of theory-informed co-creative systems and the importance of adapting them to varying user expertise.