Detecting Fatigue and Assessing Cognitive Performance in Video-Based Meetings: Integration of Lab, Field, and Machine Learning Approaches

Abstract
The increased use of videoconferencing has led to the development of a new mental health
concern known as Zoom Fatigue. This phenomenon is characterized by emotional exhaustion
and impaired well-being resulting from prolonged gazing at computer screens and long hours of
back-to-back video meetings. The purpose of this research is to examine the impact of Zoom
Fatigue on cognitive performance across gender and develop an automated method to detect it.
The research will accomplish three specific aims: (1) To examine participants’ behavior and
cognitive performance during a two-hour live video meeting, analyzing the relationship between
cognitive performance, fatigue, working memory capacity, and gender. Data collected from
communication transcripts, chat messages, meeting behaviors, and nonverbal cues will be
subsequently utilized to train machine learning algorithms to detect the presence of Zoom
Fatigue. (2) To investigate the relationship between self-reported and physiological measures
(heart rate, skin temperature, pulse rate and blood pressure) of Zoom Fatigue in males and
females over time using a wearable device. The physiological data will also be used to improve
the accuracy of predicting Zoom Fatigue in the next aim of creating a computational tool using
statistical and machine learning methods. (3) To develop a preliminary statistical and machine
learning-based computational tool to monitor and detect predictors of Zoom Fatigue. The
proposed study is innovative in its focus on a new mental health concern, its interdisciplinary
approach, and its comprehensive data sources. Combining these elements, this research will
generate not only knowledge about the cognitive processes associated with Zoom Fatigue and its
impact on anxiety but also develop an innovative tool to help identify and prevent the onset of
the issue. Additionally, this proposal will provide research experiences to diverse students,
including underrepresented minority and female students, allowing them to acquire skills such as
data analysis used in modern scientific investigations. Ultimately, the results will inform the
development of tools and strategies to mitigate the negative impact of Zoom Fatigue and enhance
the overall experience of videoconferencing, thereby promoting mental well-being.

Grant Number: 1R15MH135412-01A1
NIH Institute/Center: NIH
Principal Investigator: Jonali Baruah