Bridging Online Behavior and Clinical Insight: A Longitudinal LLM-based Study of Suicidality on YouTube Reveals Novel Digital Markers

2025-06-11

AI Summary

This study investigates the digital behavioral representations of suicide attempt survivors on YouTube and their associations with clinical cognition. Leveraging longitudinal channel data from 181 suicide attempt survivors and 134 matched controls, we integrate BERTopic-based LLM thematic modeling, temporal statistical analysis, clinical narrative coding, and a human–AI collaborative annotation framework. We identify five significantly associated themes, two of which exhibit strong temporal precursory signals (p < .01) preceding suicide attempts. Notably, we report the first evidence that platform-specific “YouTube Engagement” metrics—such as view duration, interaction frequency, and comment sentiment intensity—significantly increase prior to suicide attempts. Furthermore, we distinguish two distinct psychological shifts—“helping motivation” and “self-recovery orientation”—demonstrating that digital behaviors reflect underlying cognitive and affective transitions. Finally, we propose a tripartite paradigm—computationally driven discovery, expert-informed interpretation, and clinically validated evaluation—to bridge the gap between bottom-up data patterns and top-down theoretical frameworks.

Technology Category

• Application Domains: Humanities & Computational Social Science
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Data Mining & Knowledge Management: Mining of Visual, Multimedia & Multimodal Data

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Web Mining and Content Analysis: Topic discovery and tracking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Suicide remains a leading cause of death in Western countries, underscoring the need for new research approaches. As social media becomes central to daily life, digital footprints offer valuable insight into suicidal behavior. Focusing on individuals who attempted suicide while uploading videos to their channels, we investigate: How do suicidal behaviors manifest on YouTube, and how do they differ from expert knowledge? We applied complementary approaches: computational bottom-up, hybrid, and expert-driven top-down, on a novel longitudinal dataset of 181 YouTube channels from individuals with life-threatening attempts, alongside 134 control channels. In the bottom-up approach, we applied LLM-based topic modeling to identify behavioral indicators. Of 166 topics, five were associated with suicide-attempt, with two also showing temporal attempt-related changes ($p<.01$) - Mental Health Struggles ($+0.08$)* and YouTube Engagement ($+0.1$)*. In the hybrid approach, a clinical expert reviewed LLM-derived topics and flagged 19 as suicide-related. However, none showed significant attempt-related temporal effects beyond those identified bottom-up. Notably, YouTube Engagement, a platform-specific indicator, was not flagged by the expert, underscoring the value of bottom-up discovery. In the top-down approach, psychological assessment of suicide attempt narratives revealed that the only significant difference between individuals who attempted before and those attempted during their upload period was the motivation to share this experience: the former aimed to Help Others ($eta=-1.69$, $p<.01$), while the latter framed it as part of their Personal Recovery ($eta=1.08$, $p<.01$). By integrating these approaches, we offer a nuanced understanding of suicidality, bridging digital behavior and clinical insights. * Within-group changes in relation to the suicide attempt.

Problem

• Identify digital markers of suicidality on YouTube
• Compare suicidal behaviors with expert clinical knowledge
• Analyze longitudinal data to understand suicide-related content

Innovation

• LLM-based topic modeling for behavioral indicators
• Hybrid approach combining expert and computational analysis
• Longitudinal dataset integrating digital and clinical insights