Detecting response bias on the mindstreams battery

Omer Hegedish, Glen M. Doniger, Avraham Schweiger

Research output: Contribution to journalArticlepeer-review

Abstract

The most effective tests for detection of negative response bias (NRB) are based on the forced choice paradigm, but patients coached prior to their administration may easily discern their purpose. Algorithms for NRB detection based upon parameters from standard neuropsychological tests may be more difficult for patients to discern and thus more effective in detecting NRB. The current study aimed to uncover abnormal performance patterns associated with NRB on the MindStreams computerized battery. Simulators, cognitively healthy controls, and genuinely cognitively impaired patients participated in the study. A repeated measures analysis was applied to delineate anomalous patterns on Verbal Memory, Non-Verbal Memory, Staged Information Processing Speed, and Stroop tests. The variables derived from this analysis were entered into a multivariate discriminant function analysis (DFA). The DFA model, based on accuracy, reaction time (RT) and accuracy/RT measures, differentiated efficiently between simulators and the two control groups. Various cut-offs were provided based on this model. It is concluded that computerized cognitive testing may be valuable for malingering assessment using detection strategies based on performance patterns.

Original languageEnglish
Pages (from-to)262-281
Number of pages20
JournalPsychiatry, Psychology and Law
Volume19
Issue number2
DOIs
StatePublished - Apr 2012
Externally publishedYes

Keywords

  • attention
  • CVA
  • executive functions
  • information processing
  • malingering
  • MindStreams
  • neuropsychological assessment
  • NeuroTrax
  • response bias
  • TBI

ASJC Scopus subject areas

  • Pathology and Forensic Medicine
  • Psychology (miscellaneous)
  • Psychiatry and Mental health
  • Law

Fingerprint

Dive into the research topics of 'Detecting response bias on the mindstreams battery'. Together they form a unique fingerprint.

Cite this