A novel method for objectively classifying sequential emotion within dreams: a proof-of-concept pilot study.

Area of Science:

• Psychology
• Neuroscience
• Computer Science

Background:

• Traditional dream emotion analysis relies on subjective human ratings, which have limitations.
• Objective quantification of sequential dream emotions is needed to overcome subjectivity.
• Natural language processing offers potential for automated emotion analysis.

Purpose of the Study:

• To develop and validate a novel method for objective, sequential emotion classification in dream reports using sentiment analysis.
• To demonstrate the visualization of emotion fluctuations within dreams.
• To introduce novel "emotion indicators" derived from these fluctuations as proxies for emotion regulation and assess their correlation with sleep, affective, and personality variables.

Main Methods:

• Utilized sentiment analysis (Valence Aware Dictionary for sEntiment Reasoning - VADER) on 14 dream reports collected during polysomnography.
• Developed 6 novel "emotion indicators" based on visualized emotion fluctuations (peaks, troughs, area under the curve, trendline).
• Correlated these indicators with pre/post-sleep affective data and personality metrics.

Main Results:

• Sentiment analysis successfully generated continuous valence ratings, enabling visualization of emotion dynamics in dreams.
• Six novel objective dream emotion indicators were defined: Peak/Trough counts, Positive/Negative/Overall Emotion Intensity, and Emotion Gradient.
• Five out of six emotion indicators showed significant correlations with sleep, affective, or personality variables.

Conclusions:

• The developed sentiment analysis method provides an objective approach to quantifying sequential emotions in dreams.
• Novel "emotion indicators" show potential as proxies for in-dream emotion regulation processes.
• This study lays the methodological groundwork for future research on objective dream emotion analysis.