Disentangling Loneliness, Depression, and Social Media Use: A Gaussian, Mixed, and Bayesian Network Approach in the...

Area of Science:

• Mental Health
• Computational Social Science
• Digital Psychiatry

Background:

• Loneliness, depression, and social media use (SMU) are complex, interconnected mental health issues.
• Previous research has not fully clarified the directional relationships and symptom-level interactions between these phenomena.
• Traditional analyses often oversimplify these constructs, missing nuanced dynamics.

Purpose of the Study:

• To investigate the intricate relationships between loneliness, depression, and SMU using advanced network modeling.
• To identify potential causal pathways and central symptoms within these interconnected mental health domains.
• To explore how sociodemographic factors moderate the network structure.

Main Methods:

• Employed Gaussian Graphical Models (GGMs), Moderated Mixed Graphical Models (MGMs), and Bayesian Network Analysis on a large European sample (N=25,646).
• Analyzed data from the EU Loneliness Survey, focusing on symptom-level interactions.
• Conducted moderation analyses to assess the influence of gender, religiosity, income, and education.

Main Results:

• Time spent on social media was identified as a key symptom linking loneliness and depression to SMU.
• Gender and religiosity significantly moderated network connections, with distinct patterns observed for women and religious individuals.
• Bayesian network analysis indicated a pathway from perceived lack of support to depression, followed by compensatory SMU.

Conclusions:

• Loneliness plays a central role in initiating depressive symptoms and maladaptive SMU.
• Sociodemographic factors like gender and religiosity significantly shape these mental health dynamics.
• Findings support the development of precise, symptom-focused digital mental health interventions.