The words people use in their daily lives can reveal important aspects of their social and psychological worlds. With advances in computer technology, text analysis allows researchers to reliably and quickly assess features of what people say as well as subtleties in their linguistic styles. Following a brief review of several text analysis programs, we summarize some of the evidence that links natural word use to personality, social and situational fluctuations, and psychological interventions. Of particular interest are findings that point to the psychological value of studying particles—parts of speech that include pronouns, articles, prepositions, conjunctives, and auxiliary verbs. Particles, which serve as the glue that holds nouns and regular verbs together, can serve as markers of emotional state, social identity, and cognitive styles.

Psychological aspects of natural language. use: our words, our selves.

Major depression constitutes a serious challenge in personal and public health. Tens of millions of people each year suffer from depression and only a fraction receives adequate treatment. We explore the potential to use social media to detect and diagnose major depressive disorder in individuals. We first employ crowdsourcing to compile a set of Twitter users who report being diagnosed with clinical depression, based on a standard psychometric instrument. Through their social media postings over a year preceding the onset of depression, we measure behavioral attributes relating to social engagement, emotion, language and linguistic styles, ego network, and mentions of antidepressant medications. We leverage these behavioral cues, to build a statistical classifier that provides estimates of the risk of depression, before the reported onset. We find that social media contains useful signals for characterizing the onset of depression in individuals, as measured through decrease in social activity, raised negative affect, highly clustered egonetworks, heightened relational and medicinal concerns, and greater expression of religious involvement. We believe our findings and methods may be useful in developing tools for identifying the onset of major depression, for use by healthcare agencies; or on behalf of individuals, enabling those suffering from depression to be more proactive about their mental health.

/pdf/predicting-depression-via-social-media-45oic93snq.pdf

Predicting Depression via Social Media

Depression is a serious and widespread public health challenge. We examine the potential for leveraging social media postings as a new type of lens in understanding depression in populations. Information gleaned from social media bears potential to complement traditional survey techniques in its ability to provide finer grained measurements over time while radically expanding population sample sizes. We present work on using a crowdsourcing methodology to build a large corpus of postings on Twitter that have been shared by individuals diagnosed with clinical depression. Next, we develop a probabilistic model trained on this corpus to determine if posts could indicate depression. The model leverages signals of social activity, emotion, and language manifested on Twitter. Using the model, we introduce a social media depression index that may serve to characterize levels of depression in populations. Geographical, demographic and seasonal patterns of depression given by the measure confirm psychiatric findings and correlate highly with depression statistics reported by the Centers for Disease Control and Prevention (CDC).

/pdf/social-media-as-a-measurement-tool-of-depression-in-tnimi00o6y.pdf

Social media as a measurement tool of depression in populations

Sensors in everyday devices, such as our phones, wearables, and computers, leave a stream of digital traces. Personal sensing refers to collecting and analyzing data from sensors embedded in the context of daily life with the aim of identifying human behaviors, thoughts, feelings, and traits. This article provides a critical review of personal sensing research related to mental health, focused principally on smartphones, but also including studies of wearables, social media, and computers. We provide a layered, hierarchical model for translating raw sensor data into markers of behaviors and states related to mental health. Also discussed are research methods as well as challenges, including privacy and problems of dimensionality. Although personal sensing is still in its infancy, it holds great promise as a method for conducting mental health research and as a clinical tool for monitoring at-risk populations and providing the foundation for the next generation of mobile health (or mHealth) interventions.

Personal Sensing: Understanding Mental Health Using Ubiquitous Sensors and Machine Learning.

Objective: Somatization has often been viewed as a defense against awareness of emotional distress or as a masked version ofdepression. This report examines whether community residents with high levels offunctional somatic symptoms also report overt psychological distress and whether somatization is associated with any specific psychiatric disorder. Method: Analyses used data from the community sample of the National Institute of Mental Health Epidemiologic Catchment Area (ECA) study, a population-based survey ofpsychiatric morbidity among more than 1 8,000 residents of five U.S. communities. Results: Increasing number of somatization symptoms was strongly associated with overt expression of psychological distress and psychiatric symptoms. Among ECA respondents with five or more current functional somatic symptoms, 63% reported current psychological symptoms and 50% met criteria for a current psychiatric diagnosis (compared to 7% and 6%, respectively, among those with no current somatization symptoms). Somatization symptoms showed strongest associations with anxiety and depressive symptoms, intermediate association with symptoms ofpsychotic disorders, and weakest associations with symptoms ofsubstance abuse and antisocial personality. Conclusions: ECA study respondents with high levels ofsomatization symptoms typically reported overt psychological distress, especially anxiety and depression. Patterns of response do not support a dissociation between physical and emotional symptoms. Functional somatic symptoms appear to be common expressions of distress instead of defenses against awareness. (Am J Psychiatry 1991; 148:1494-1500)

Somatization and psychiatric disorder in the NIMH Epidemiologic Catchment Area study.

In order to assess which of three current models is most useful in understanding paranoia, the authors applied computer speech content analysis to 55 patients--24 of whom were in four groups expressing paranoid delusions and 31 of whom were in four groups not expressing such delusions. The results delineated a semantic or verbal profile of paranoid self-presentation. This self-presentation is more identifiable than the effects of any other patient characteristic, even if the delusion is not discussed by the patient. The strength of the statistical evidence supports the model of paranoid delusions as a separate disease rather than as a subtype of schizophrenia or as a trait that exists on a spectrum from normality to pathology.

The language of paranoia.

DSM-III has established diagnostic criteria that separate somatization disorder from other overlapping symptom configurations. Nevertheless, information regarding the experience of somatization disorder is far from complete. Terms such as "masked depression" or "alexithymia" imply that a disturbance of affect is a central but guarded issue for at least some somatizing patients. Through content analysis of speech, the authors investigated the self-experience of somatization disorder in relation to affective disorder. Rather than defended depression, a distinctive characteristic found in the language of patients with somatization disorder reflects a confused, negative self-identity.

Thomas E. Oxman

Papers

The language of paranoia.

Linguistic dimensions of affect and thought in somatization disorder.