Linguistic Markers Indicating Therapeutic Outcomes of Social Media Disclosures of Schizophrenia

Abstract

Self-disclosure of stigmatized conditions is known to yield therapeutic benefits. Social media sites are emerging as promising platforms enabling disclosure around a variety of stigmatized concerns, including mental illness. What kind of behavioral changes precede and follow such disclosures? Do the therapeutic benefits of "opening up" manifest in these changes? In this paper, we address these questions by focusing on disclosures of schizophrenia diagnoses made on Twitter. We adopt a clinically grounded quantitative approach to first identify temporal phases around disclosure during which symptoms of schizophrenia are likely to be significant. Then, to quantify behaviors before and after disclosures, we define linguistic measures drawing from literature on psycholinguistics and the socio-cognitive model of schizophrenia. Along with significant linguistic differences before and after disclosures, we find indications of therapeutic outcomes following disclosures, including improved readability and coherence in language, future orientation, lower self preoccupation, and reduced discussion of symptoms and stigma perceptions. We discuss the implications of social media as a new therapeutic tool in supporting disclosures of stigmatized conditions.

Hateful People or Hateful Bots?

“…The topic of understanding online human behavior has been of a great interest to CSCW/HCI researchers in various contexts such as mental health [19,23], political polarization [11,35], and abusive social behaviors [13,50]. Our findings challenge the assumption often made by such studies that online social media content is always created by humans.…”

Variability in Language used on Social Media prior to Hospital Visits

“…We use three sets of linguistic features for measuring change in language: a) open-vocabulary topics 28 , b) dictionary-based psycholinguistic features 29 www.nature.com/scientificreports www.nature.com/scientificreports/ (positive or negative affectivity), arousal (how calming or exciting the post is) 30 , and extent of anxious, extraverted, and depressed language 31 by applying previously developed statistical models, and meta features such as posting statistics (average number of 1-grams) and time of posts. These have been shown to be predictive of several health outcomes such as depression 4 , schizophrenia 32 , attention deficit hyperactivity disorder (ADHD) 33 , and general well-being 34 . As researchers have used various types of biomarkers for diseases 35,36 , our aim was to identify the language markers 37 among these features that are predictive and are differentially expressed prior to hospital visits.…”

Social Media and Mental Health: Benefits, Risks, and Opportunities for Research and Practice

“…Several studies have also demonstrated that when compared with a control group, Twitter users with a selfdisclosed diagnosis of schizophrenia show unique online communication patterns (Birnbaum et al 2017a), including more frequent discussion of tobacco use (Hswen et al 2017), symptoms of depression and anxiety (Hswen et al 2018b), and suicide (Hswen et al 2018a). Another study found that online disclosures about mental illness appeared beneficial as reflected by fewer posts about symptoms following self-disclosure (Ernala et al 2017). Each of these examples offers early insights into the potential to leverage widely available online data for better understanding the onset and course of mental illness.…”