Proposed new words and collective identity in human and language model agent groups

Abstract Groups develop distinctive words even when ordinary expressions are sufficient. Such words can symbolize collective identity, and observational studies link language use to community attachment. We examined identity change in temporary groups of one human and four disclosed large language model (LLM) agents, comparing discussions with and without a neologism proposal across recruitment waves. We recruited 100 Japanese participants in two sequential waves; an agent proposed a new word only in the first wave. Mean collective identity change did not differ significantly between waves. Within the proposal wave, favorable ratings of the proposed word were associated with change in identity, but this association did not survive correction for multiple testing (adjusted $$p=0.229$$ ). The three-way interaction among recruitment wave, message count, and team impression had a nominal $$p=0.027$$ but did not survive correction (adjusted $$p=0.082$$ ). In a sensitivity analysis, the interaction was not significant after adjustment for midpoint identity (nominal $$p=0.078$$ ; adjusted $$p=0.624$$ ). The nonsignificant between-wave comparison and exploratory associations that did not survive correction, together with wave confounding and post-discussion evaluations, limit causal inference and motivate randomized studies with repeated measures to test whether proposed group language affects identification.

Authors

Institutions

Publication Details

Journal
Scientific Reports
Published
2026-11-16
DOI
https://doi.org/10.1038/s41598-026-80001-6
Primary Topic
Social and Intergroup Psychology
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Proposed new words and collective identity in human and language model agent groups

Kazuhiro Ito, Eiji Aramaki, Shoko Wakamiya, Hiroyuki Nagai
1 links
Scientific Reports
Social and Intergroup Psychology
article

Proposed new words and collective identity in human and language model agent groups

Kazuhiro Ito, Eiji Aramaki, Shoko Wakamiya, Hiroyuki Nagai
article en

Abstract

Abstract Groups develop distinctive words even when ordinary expressions are sufficient. Such words can symbolize collective identity, and observational studies link language use to community attachment. We examined identity change in temporary groups of one human and four disclosed large language model (LLM) agents, comparing discussions with and without a neologism proposal across recruitment waves. We recruited 100 Japanese participants in two sequential waves; an agent proposed a new word only in the first wave. Mean collective identity change did not differ significantly between waves. Within the proposal wave, favorable ratings of the proposed word were associated with change in identity, but this association did not survive correction for multiple testing (adjusted $$p=0.229$$ ). The three-way interaction among recruitment wave, message count, and team impression had a nominal $$p=0.027$$ but did not survive correction (adjusted $$p=0.082$$ ). In a sensitivity analysis, the interaction was not significant after adjustment for midpoint identity (nominal $$p=0.078$$ ; adjusted $$p=0.624$$ ). The nonsignificant between-wave comparison and exploratory associations that did not survive correction, together with wave confounding and post-discussion evaluations, limit causal inference and motivate randomized studies with repeated measures to test whether proposed group language affects identification.

Scientific Reports
The University of Tokyo (JP), Nara Institute of Science and Technology (JP)
Japan Science and Technology Agency
Social and Intergroup Psychology
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon