How Differing Audiences Were Associated with User Emotional Expression on a Well-Being App.

Area of Science:

• Digital health
• Psychology
• Natural Language Processing

Background:

• Mobile applications for emotional well-being have rapidly increased.
• Limited research exists on user interaction and therapeutic alliance formation within these apps, particularly through writing.
• Writing is a key practice for emotional processing and expression, with linguistic features reflecting affect.

Purpose of the Study:

• To investigate how audience influences users' emotional expression in writing within a well-being app.
• To analyze the linguistic characteristics of user-generated text in relation to audience.
• To understand the role of writing in developing a therapeutic alliance in digital mental health tools.

Main Methods:

• Utilized IBM Watson's Natural Language Processing (NLP) software to analyze user-generated text.
• Examined differences in emotional expression based on whether users wrote to an explicit audience.
• Applied narrative theory and literacy theories to interpret linguistic features of writing.

Main Results:

• Audience significantly impacted how users expressed emotions in their writing.
• Writing to an explicit audience resulted in longer narratives.
• Users writing to an audience exhibited less sadness, anger, and disgust, and more joy.

Conclusions:

• The intended audience is a critical factor in shaping emotional expression within well-being apps.
• Findings offer valuable insights for the design and development of more effective digital mental health interventions.
• Understanding audience-specific writing patterns can enhance the therapeutic potential of well-being applications.