Uncovering the Most Important Factors for Predicting Sexual Desire Using Explainable Machine Learning

Area of Science:

• Psychology
• Human Sexuality
• Machine Learning Applications

Background:

• Low sexual desire is a prevalent issue, affecting 34% of women and 15% of men.
• Understanding factors influencing sexual desire is crucial for enhancing overall quality of life.

Purpose of the Study:

• To identify key individual and relational predictors of dyadic (partnered) and solitary sexual desire.
• To utilize advanced statistical methods for analyzing a large number of potential predictors.

Main Methods:

• Employed a random forest machine learning algorithm to predict sexual desire.
• Analyzed data from two online samples (N=1,846), including couples.
• Utilized Shapley value technique to interpret predictor variable effects.

Main Results:

• Machine learning models explained approximately 40% of the variance in sexual desire.
• Sexual satisfaction and romantic love were significant predictors of dyadic sexual desire.
• Masturbation frequency and attitudes toward sexuality predicted solitary sexual desire.
• Gender was not a significant predictor of sexual desire.

Conclusions:

• Relationship satisfaction and romantic love are vital for partnered sexual desire.
• Attitudes toward sexuality and masturbation are important for solitary sexual desire.
• Findings support addressing relationship dynamics and sexual attitudes in therapy for low desire.