Embedded dyads: How the personal network relates to linguistic accommodation
- 1Department of Sociology, University of Geneva, Geneva, Switzerland
- 2LIVES Centre, Swiss Centre of Expertise for Life Course Research, University of Geneva, Geneva, Switzerland
- 3Department of English, North Carolina State University, Raleigh, United States of America
Abstract
Language serves as a key medium through which individuals express and negotiate social belonging. By observing individuals’ behavior within an interaction, we can access the relational stance they adopt and infer the social relationships that link the two interactants. This process is particularly studied within the framework of the Communication Accommodation Theory (CAT). According to CAT, speakers tend to align linguistically with their interlocutors to signal social proximity. Yet, most analyses remain focused on the psychological and dyadic levels, without considering the broader social influences that may shape linguistic choices. In this paper, we argue that considering the relational structure surrounding both interactants is crucial. In particular, we hypothesize that the position of the interlocutor within the speaker’s personal network is a key factor driving linguistic behavior. To address this, we conducted an exploratory empirical investigation with 21 young adults who recorded themselves in an average of three interactions each (resulting in a total of 71 discussions). Their personal networks and linguistic data were both collected. Linguistic similarity was operationalized via the Manhattan distance of intensifier frequency between the respondent (ego) and the interlocutor (alter). We used Generalized Estimating Equation models to test our hypotheses. Results indicate that speakers tend to align linguistically with their interlocutors when the latter occupy a central position within a compositionally homogeneous cluster—that is, when the alter is embedded in a structurally dense subgroup composed of individuals originating from the same social domain.
Related
- Content-Sensitive Linguistic Representations in the Human Multiple-Demand Network
- Bedtime rumination mediates the burnout-insomnia link and relates to shift adaptation in emergency department staff
- Diet Quality from Midlife to Later Life Relates to Late-Life Brain Health and Verbal Memory in the SG70 Cohort
- A cross-kingdom interactome predicted by AlphaFold3 reveals a DNF2-centered interface required for symbiotic accommodation
- Sensory receptor expansion and neural accommodation in butterfly color vision
BibTeX
@article{fernandez2026embedded,
title = {{Embedded dyads: How the personal network relates to linguistic accommodation}},
author = {Guillaume P. Fernandez and Eric D. Widmer and Robin Dodsworth},
journal = {PLOS One},
year = {2026},
month = sep,
doi = {10.1371/journal.pone.0357457},
url = {https://doi.org/10.1371/journal.pone.0357457},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Embedded dyads: How the personal network relates to linguistic accommodation
AU - Guillaume P. Fernandez
AU - Eric D. Widmer
AU - Robin Dodsworth
PY - 2026
DA - 2026/09/21
JO - PLOS One
DO - 10.1371/journal.pone.0357457
UR - https://doi.org/10.1371/journal.pone.0357457
AB - Language serves as a key medium through which individuals express and negotiate social belonging. By observing individuals’ behavior within an interaction, we can access the relational stance they adopt and infer the social relationships that link the two interactants. This process is particularly studied within the framework of the Communication Accommodation Theory (CAT). According to CAT, speakers tend to align linguistically with their interlocutors to signal social proximity. Yet, most analyses remain focused on the psychological and dyadic levels, without considering the broader social influences that may shape linguistic choices. In this paper, we argue that considering the relational structure surrounding both interactants is crucial. In particular, we hypothesize that the position of the interlocutor within the speaker’s personal network is a key factor driving linguistic behavior. To address this, we conducted an exploratory empirical investigation with 21 young adults who recorded themselves in an average of three interactions each (resulting in a total of 71 discussions). Their personal networks and linguistic data were both collected. Linguistic similarity was operationalized via the Manhattan distance of intensifier frequency between the respondent (ego) and the interlocutor (alter). We used Generalized Estimating Equation models to test our hypotheses. Results indicate that speakers tend to align linguistically with their interlocutors when the latter occupy a central position within a compositionally homogeneous cluster—that is, when the alter is embedded in a structurally dense subgroup composed of individuals originating from the same social domain.
SN - 1932-6203
ER -