Background 
The advancement of the information age has led to the widespread use of the internet, accompanied by numerous internet-related issues that often correlate with various physical and mental health conditions, particularly among college freshmen. 

We examined the relationship between internet addiction (IA) and insomnia among these students, using emotion regulation (ER) and anxiety as mediators and gender as a moderating variable.

Methods 
This cross-sectional study included 7,353 freshmen from a university in Jingzhou City, Hubei Province, China. Data were collected through an online self-administered questionnaire, including the Internet Addiction Test (IAT), the Emotion Regulation subscale (ER), the Generalized Anxiety Disorder 7-item scale (GAD-7), and the Insomnia Severity Index (ISI). Data analysis was conducted using SPSS 21.0 and PROCESS version 4.1 to test the hypothesized relationships among variables.

Results

• In our survey, correlation analysis showed that ER was significantly negatively correlated with IA, anxiety, and insomnia.
• IA was significantly positively correlated with anxiety and insomnia (all p<0.01). 
  The mediating effect analysis indicated that IA was a significant positive predictor of insomnia.
• ER and anxiety played a chain - mediating role in the development of insomnia (β=0.039, 95% confidence interval=0.035–0.043).
• The moderating effect analysis showed that the interaction term of IA and gender had a significant negative predictive effect on ER (β=-0.014, 95% confidence interval [-0.027, -0.001]) and insomnia (β=-0.022, 95% confidence interval [-0.036, -0.007]).
• Males (direct effect: β=0.048, 95% confidence interval=[0.037, 0.059]) had a stronger predictive ability for the level of insomnia than females (direct effect: β=0.026, 95% confidence interval=[0.014, 0.037]).
• Females (indirect effect 1: β=0.015, 95% confidence interval=[0.010, 0.020]; indirect effect 2: β=0.041, 95% confidence interval=[0.037, 0.045]) had a stronger predictive ability for the level of insomnia through the level of IA than males (indirect effect 1: β=0.014, 95% confidence interval=[0.009, 0.018]; indirect effect 2: β=0.037, 95% confidence interval=[0.033, 0.041]).

Conclusion 
IA can exacerbate insomnia in college freshmen by compromising their ER, subsequently triggering anxiety symptoms. The process differs by gender, suggesting tailored strategies for each. These findings may play crucial roles in promoting the physical and mental well-being of college freshmen.