Reading Between the Lines: Exploring Lexical and Emotional Content in Baseline Trauma Impact Statements

The words a person uses while describing salient life events/experiences can provide a wealth of information about their cognitive and emotional functioning. Linguistic analyses can thus play an important role in clinical settings by providing insights into patients' mental health. The current paper applies such techniques to the study of posttraumatic stress disorder (PTSD). It uses a combination of computational and human rating methods to analyze the lexical and emotional content of trauma appraisal essays (i.e., impact statements) written by a treatment-seeking clinical sample (N=94). More frequent use of first-person plural pronouns (e.g., 'we', 'our') was associated with lower self-reported PTSD and depressive severity, higher ratings of expressed anger were associated with higher clinician-rated PTSD severity, and higher ratings of expressed vulnerable emotions were associated with higher self-reported PTSD and depressive severity. Text-mining approaches also provided insights into common themes across participants' trauma impact statements (e.g., safety, trust, and self-esteem).