Computational Analysis of Conversation Dynamics through Participant Responsivity

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hughes, Margaret, Roy, Brandon, Poole-Dayan, Elinor, Roy, Deb, Kabbara, Jad
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915591358513152
author Hughes, Margaret
Roy, Brandon
Poole-Dayan, Elinor
Roy, Deb
Kabbara, Jad
author_facet Hughes, Margaret
Roy, Brandon
Poole-Dayan, Elinor
Roy, Deb
Kabbara, Jad
contents Growing literature explores toxicity and polarization in discourse, with comparatively less work on characterizing what makes dialogue prosocial and constructive. We explore conversational discourse and investigate a method for characterizing its quality built upon the notion of ``responsivity'' -- whether one person's conversational turn is responding to a preceding turn. We develop and evaluate methods for quantifying responsivity -- first through semantic similarity of speaker turns, and second by leveraging state-of-the-art large language models (LLMs) to identify the relation between two speaker turns. We evaluate both methods against a ground truth set of human-annotated conversations. Furthermore, selecting the better performing LLM-based approach, we characterize the nature of the response -- whether it responded to that preceding turn in a substantive way or not. We view these responsivity links as a fundamental aspect of dialogue but note that conversations can exhibit significantly different responsivity structures. Accordingly, we then develop conversation-level derived metrics to address various aspects of conversational discourse. We use these derived metrics to explore other conversations and show that they support meaningful characterizations and differentiations across a diverse collection of conversations.
format Preprint
id arxiv_https___arxiv_org_abs_2509_16464
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Computational Analysis of Conversation Dynamics through Participant Responsivity
Hughes, Margaret
Roy, Brandon
Poole-Dayan, Elinor
Roy, Deb
Kabbara, Jad
Computation and Language
Computers and Society
Growing literature explores toxicity and polarization in discourse, with comparatively less work on characterizing what makes dialogue prosocial and constructive. We explore conversational discourse and investigate a method for characterizing its quality built upon the notion of ``responsivity'' -- whether one person's conversational turn is responding to a preceding turn. We develop and evaluate methods for quantifying responsivity -- first through semantic similarity of speaker turns, and second by leveraging state-of-the-art large language models (LLMs) to identify the relation between two speaker turns. We evaluate both methods against a ground truth set of human-annotated conversations. Furthermore, selecting the better performing LLM-based approach, we characterize the nature of the response -- whether it responded to that preceding turn in a substantive way or not. We view these responsivity links as a fundamental aspect of dialogue but note that conversations can exhibit significantly different responsivity structures. Accordingly, we then develop conversation-level derived metrics to address various aspects of conversational discourse. We use these derived metrics to explore other conversations and show that they support meaningful characterizations and differentiations across a diverse collection of conversations.
title Computational Analysis of Conversation Dynamics through Participant Responsivity
topic Computation and Language
Computers and Society
url https://arxiv.org/abs/2509.16464