This research will aim to validate the capacity to measure cognitive impairment (and other neuropsychological dimensions) from five-minute speech samples by means of computer software. The software implements a measurement method (Gottschalk-Gleser) applied to the content and form analysis of five-minute samples of natural language. The speech samples are elicited by purposely ambiguous instructions to talk about any interesting or dramatic personal life experiences. The content analysis method focuses on each clause in a speech sample, parses these clauses, selects from a built-in dictionary those words or phrases that have been found relevant to specific content analysis scales, and derives a score for each scale. The software reads the speech sample from a computer file and compares the scores obtained with norms for adults and children. In addition to a Cognitive Impairment scale, the software can be applied to many other content analysis scales, to assess comorbid neuropsychiatric disorders. A group of 30 drug-abusing patients admitted to an inpatient service at the UCI Neuropsychiatric Center will be clinically evaluated. The scores from our computerized measure will be compare to scores based on the Minimental Status Examination and selected tests from the Halstead- Reitan Neuropsychological Test Battery, the Automated Neuropsychological Assessment Metric Battery (ANAM), the Word Association Test, the Stroop Color Test, the SMDT. Selected computerized tests will be adapted from the ANAM to supplement the Five-Minute Verbal Sample (computerized) Test.
