The application of a computerized measurement of the content analysis of natural language to the assessment of the effects of psychoactive drugs

• Published in: Methods and findings in experimental and clinical pharmacology Vol. 21; no. 2; p. 133
• Main Author: Gottschalk, L A
• Format: Journal Article
• Language: English
• Published: Spain 01.03.1999
• Subjects: Artificial Intelligence; Female; Humans; Male; Psychopharmacology; Psychotropic Drugs - pharmacokinetics; Psychotropic Drugs - pharmacology; Verbal Behavior - drug effects
• ISSN: 0379-0355
• DOI: 10.1358/mf.1999.21.2.529240

This article aims to answer the question whether developing technology is now capable of measuring objectively and accurately the effects of psychoactive pharmacological agents by means of computerized assessments of psychological states and traits through the analysis of content and form of people's speech. At the present time, psychopharmacological researchers involved in clinical trials rely on DSM-IV criteria and standardized self-report measures and observer rating scales to assess psychoactive drug effects. Attention is drawn to the potentially unrecognized measurement errors and relatively low interrater reliability by these methods--for example, all raters are not free of observer bias and every subject administered a drug is not equally and accurately well-informed about the self. The computerized content analysis methods tend to avoid these biases and measurement errors. A review is provided describing the Gottschalk-Gleser method of measuring psychobiological dimensions from the form and content of short (usually five-minute) speech samples of verbal behavior, generally elicited by standardized and purposely ambiguous instructions to talk about any interesting or dramatic personal life experiences. Norms have been obtained by this method of speech elicitation, adjusted for age, sex and educational level. Sections are provided covering cross-cultural and language validation research on the Gottschalk-Gleser content analysis method, the influence of medical or psychiatric illness as well as psychoactive drugs on verbal content analysis-derived scores, and the research carried out for more than 20 years computerizing this content analysis procedure through the development of artificial intelligence software enabling these measurements to be done from typescripts of speech samples on computer diskettes. A brief review deals with the general applications of this method of measurement to basic and clinical psychiatry, psychosomatic medicine, neuropsychology, the diagnostic process in a psychiatry outpatient clinic, children's mental health problems, dream research, and assessment of mental processes during PET scanning of the brain. This is followed by a review of the applications of this method of content analysis of speech to neuropsychopharmacological testing of antianxiety, antidepressive, antipsychotic, and other psychoactive drugs, as well as to pharmacokinetic variables and clinical response. This method is now in the process of being used in clinical trials in psychopharmacology and is recommended for more extensive use in this research area.