27 May 2026
A research study led by Dr Chenwei Li from the International Business School Suzhou (IBSS), Xi’an Jiaotong-Liverpool University, has been published in the leading international journal Decision Support Systems. The paper, entitled “Which Nonverbal Cues Engage You? The Role of Speaker's Acoustic Characteristics and Presence in User-Generated Videos”, provides empirical insights into how vocal and visual nonverbal signals shape audience engagement in digital short‑form content.
Against the backdrop of a rapidly expanding global video ecosystem and intensifying competition for viewer attention, creators of user‑generated video (UGV) face a critical challenge: how to capture, engage, and retain audiences effectively. To address this issue, Dr Li’s study combines large‑scale, real‑world data from Bilibili — including 40,742 daily observations across 1,188 videos — with controlled online experiments to identify the causal effects of speakers’ nonverbal cues on audience engagement. The research demonstrates that speakers’ acoustic features and visual presence serve as powerful nonverbal drivers that enhance engagement through emotional arousal, offering evidence‑based strategies for content creation and platform optimisation.
The study identifies four critical vocal cues that exert distinct curvilinear effects on audience engagement. Empirical results confirm a U‑shaped relationship between pitch/intensity and viewer engagement, and an inverted U‑shaped relationship between jitter/shimmer and viewer engagement, with emotional arousal acting as a significant mediating mechanism. Speaker presence — whether the creator appears on camera — strengthens the U‑shaped effect of intensity and the inverted U‑shaped effect of shimmer, as visual cues amplify emotional contagion between speakers and viewers. Notably, pitch and jitter represent subtler acoustic signals, and their effects are not significantly moderated by visual speaker presence.
These findings offer clear, actionable implications for creators and platforms. Content creators can strategically use more extreme pitch or intensity to emphasise emotion, maintain moderate levels of jitter and shimmer to avoid both monotony and over‑expression, and combine high‑impact vocal delivery with on‑camera presence to maximise audience engagement. For digital platforms, the research supports the design of AI‑driven, real‑time feedback tools for vocal performance, as well as features that encourage the integration of visual and vocal signals. In an era where nonverbal communication often outweighs verbal content in driving engagement, mastering these cues has become essential for UGC creators seeking to build and sustain a competitive advantage.
This study makes significant theoretical and practical contributions to the fields of digital media, human‑computer interaction, and content analytics. By examining the nonlinear and context-dependent effects of nonverbal cues, it advances understanding of audience behaviour in interactive video environments. For practitioners, the evidence‑based framework enables a more strategic approach to content creation, moving beyond trial‑and‑error, while also informing platform design and governance to support higher‑quality user‑generated content.
About the Author
Dr Chenwei Li is an associate professor at International Business School Suzhou (IBSS), Xi’an Jiaotong-Liverpool University. Her research focuses on social media and consumer behaviour, human-computer interaction, information security and privacy. Her work has been published in leading journals, including Decision Support Systems (DSS), International Journal of Electronic Commerce (IJEC), International Journal of Information Management (IJIM), Information Processing and Management (IPM) among others as well as conference proceedings such as ICIS, PACIS, WeB, CSWIM, and CNAIS. She serves as an editorial board member of the Pacific Asia Journal of the Association for Information Systems (PAJAIS) and as a guest editor for Internet Research (IR).
Journal Introduction
Decision Support Systems is a world‑renowned, peer‑reviewed journal published by Elsevier. It is ranked JCR Q1 in multiple disciplines, including Computer Science, Information Systems, Artificial Intelligence, and Operations Research & Management Science, with a recent impact factor of 6.8 and a five‑year impact factor of 8.2. The journal is listed in the Senior Scholars’ List of Premier Journals by the Association for Information Systems (AIS) and holds the highest A* rating from the Australian Business Deans Council (ABDC). Decision Support Systems publishes high‑quality, innovative research on decision support systems, business intelligence, human‑computer interaction, data analytics, and digital platform governance, making it one of the most influential academic outlets in information systems and management science worldwide.
By Linlin Xie
Editor by Thomas Durham
27 May 2026