Rising early warning signals in affect associated with future changes in depression: a dynamical systems approach

Area of Science:

• Mental Health Research
• Psychopathology Prediction
• Dynamical Systems Theory

Background:

• Predicting future depressive episodes is crucial in mental health.
• Dynamical systems theory suggests early warning signals (EWSs) may precede disorder severity changes.
• This study examines EWSs in major depressive disorder (MDD).

Purpose of the Study:

• Investigate if rising EWSs are associated with future changes in MDD severity.
• Analyze time-series data for predictive signals of depression.
• Determine the predictive power of specific EWSs.

Main Methods:

• 31 MDD patients participated over 8 weeks.
• Daily smartphone surveys collected positive and negative affect data.
• A rolling window approach analyzed auto-correlation, temporal variance, and network connectivity.

Main Results:

• Rising auto-correlation was significantly linked to worsening depression symptoms (r = 0.41, p = 0.02).
• No significant association was found between rising temporal standard deviation and depression changes (r = -0.23, p = 0.23).
• Rising network connectivity also did not predict changes in depression symptoms (r = -0.12, p = 0.59).

Conclusions:

• Rises in auto-correlation emerged as the sole EWS predicting future worsening depression.
• This study rigorously examined EWSs in a larger MDD cohort.
• Findings highlight auto-correlation's potential as a predictive marker for depression.