Uncovering psychopaths: an automated lingustic approach
The identification of psychopaths, especially those who commit violent crimes, is important to society because of their increased risk of criminal recidivism rates. In this paper, we examine communication patterns that may be unique to individuals high in psychopathy. Several linguistic features in the narratives of prisoners convicted of murder identified by the use of automated Natural Language Processing (NLP) suggest that spontaneously produced psychopathic speech differs from non-psychopathic speech. Psychopaths produced more speech disfluencies, past tense verbs, and food/drink related words in their narratives compared to non-psychopaths. Surprisingly, no overall differences in affective word use were detected.
Jeff Hancock, Geri Gay, Michael Shapiro
psychopaths; psychopathy; natural language processing; language generation; detection; linguistic analysis; thesis; automation
dissertation or thesis