Intersubject correlation as a predictor of attention: a systematic review
Qing Liu, Yuhang Lin, Wenjuan Zhang
Abstract
This meta-analysis examines the challenge of capturing brain activity in real-world and laboratory settings by integrating naturalistic neuroimaging and experimental data with behavioral measures to explore the predictive role of intersubject correlation (ISC) in attention. Using databases such as Web of Science and PubMed, we conducted a comprehensive search from January 2000 to July 2024. Our meta-analysis of 14 studies and 27 effect sizes reveals a significant positive correlation between ISC and attention (r = 0.65, p < 0.001), demonstrating that ISC serves as a reliable neural marker for attentional engagement under various experimental conditions. By incorporating naturalistic stimuli such as video clips and controlled laboratory tasks, we provide insights into the application of ISC to predict attention in ecologically-valid contexts. Moreover, our addition of behavioral data further enhances the understanding of how neural synchronization reflects attentional states. Our results underscore the potential of utilizing ISC to develop personalized assessments and interventions in educational and cognitive settings.