The Victoria Symptom Validity Test, development of a new clinical measure of response bias
Abstract (Summary)This dissertation desmies the development of the Victoria Symptom Validity Test (VSVT). The VSVT was designed to assist in screening for non-optimal performance during neuropsychologicaI evaluation due to rnaiingering, psychiatnc disturbance, or other environmental or dispositional factors. Specifically, the VSVT is a test of recognition memory that uses the forced-choice paradigm for detecting biased or random respondhg. Response latency is also recorded. Results from pilot and follow-up normative studies with experimental and dinical populations are presented. The VSVT was found to have exceiient divergent and adequate convergent validity in samples of compensation-seekiog and noncompensation-seeking patients. Classifications of experimentalparticipants using below chance peflormance as a cutoffwere consistent with the majority of previous studies in finding 100%specificitybut poor sensitivity . A new system wherein a third category is added for questionable(at chance) performance showed greatiy inaeased sensitiviîy, with no decrement in specificity. Although scores in the questionable range are not unequivocal indicators of malingering, findings suggest good utility for screening or corroborative purposes. A Bayesian diagnostic probability matrix that takes base-rates into account was also provided as a more flexible alternative to absolute atoff scores. Response latency as a measure of symptom validity was shown to have adequate sensitivity for screening, but less acceptable discriminant validity and Iower specificity. Limitations of experimentalresults and clinical applications of symptom validity tests are dkcussed.
Source Type:Master's Thesis
Date of Publication:01/01/1997