A heart rate variability-driven framework for depression screening leveraging emotion-elicited autonomic divergence

Area of Science:

• Psychophysiology
• Affective Computing
• Clinical Psychology

Background:

• Depression is characterized by emotional dysregulation, including increased sadness sensitivity and decreased happiness responsiveness.
• Understanding physiological differences in emotional responses is crucial for developing objective depression detection methods.

Purpose of the Study:

• To analyze physiological response disparities in emotional states between individuals with depression (IWD) and healthy controls (HC).
• To establish empirical foundations for optimizing affective computing-based depression screening using heart rate variability (HRV).

Main Methods:

• Employed structured emotion induction protocols and dual-phase statistical identification using Mann-Whitney U tests.
• Compared HRV features between emotional states and resting conditions to validate emotion elicitation.
• Developed machine learning models using HRV features and intergroup differential response patterns for IWD/HC classification.

Main Results:

• Identified heightened sadness reactivity and attenuated happiness responsiveness in IWD compared to HC.
• Machine learning models achieved 76.8% accuracy for sadness-specific conditions, outperforming other emotion-induced models and resting-state baselines.
• Consistent directional patterns and response magnitudes were observed across groups, with critical divergences in specific emotional responses.

Conclusions:

• Systematic investigation of HRV patterns during discrete emotion induction confirms clinically significant differential responsiveness in IWD.
• Findings empirically validate heightened sadness susceptibility in individuals with depression.
• Results provide guidance for refining affective computing algorithms for depression screening and understanding disorder-specific physiological responses.