JMIR Publications

ABSTRACT

Background:

Internalizing mental illnesses associated with psychological distress are often under-detected. Text-based detection using natural language processing methods are increasingly used to complement conventional detection efforts. However, these often rely on self-disclosure through autobiographical narratives, that may not always be possible, especially in collectivistic Japanese culture. As such, we propose the use of narrative writing as an alternative resource for mental illness detection in youths. Accordingly, this study investigates the textual characteristics of narratives that are written by youths with psychological distress.

Objective:

Our research focuses on the detection of psychopathological tendencies in written imaginative narratives. We apply NLP tools, such as stylometric measures and lexicon-based sentiment analysis.

Methods:

Using stylometric measures and sentiment analyses, we examined short narratives from 52 Japanese youths (M = 19.81, SD = 20.01) through crowdsourcing. Participants wrote a short narrative introduction to an imagined story, before completing a questionnaire on their psychological distress tendencies. Based on this score, participants were categorized into Higher distress and Lower distress groups. Written narratives were then analyzed using stylometric measures and sentiment analysis, and examined for between-group differences. While outside the scope of the current study, we also carried out a supplementary analysis of narratives written by adults using the same procedure.

Results:

Youths at higher tendencies towards psychological distress used significantly more positive (happiness-related) words, revealing differences in valence of the narrative content. No other significant differences were observed.

Conclusions:

Youths with tendencies towards mental illness were found to write more positive stories that contained more happiness-related terms. These results may potentially have more widespread implications on screening, particularly in cultures like Japan that are not accustomed to self-disclosure. Although the mechanisms behind these differences in frequency of happiness-related term usage is speculative, but this paves the way for online surveillance and detection efforts.