Evaluating the Personality Assessment Inventory (PAI) Cognitive Bias Scale (CBS): Considering Alternative Scoring Methods for Military Populations

Invalid response detection remains a concern across populations, including military personnel (Fortney et al., 2017; Morris et al., 2022; Tolin et al., 2010). However, research on response styles in AD and Veteran populations are comparatively limited, restricting the generalizability of findings from other groups (Ingram & Ternes, 2016). The Minnesota Multiphasic Personality Inventory (MMPI) and Personality Assessment Inventory (PAI) are widely used multiscale inventories featuring embedded validity scales based on similar detection frameworks. Despite this shared foundation, small discrepancies in effect size estimates persist, generally favoring MMPI measures (e.g., RBS vs. CBS; Tylicki et al., 2021). Differences in response category structures may underlie variations in the diagnostic utility of MMPI and PAI instruments. A study of 559 AD military personnel evaluated five alternative scoring techniques for the PAI’s cognitive bias scale, examining post hoc dichotomization procedures. Although each method addressed error variance differently, diagnostic accuracy and between-group effect sizes were largely comparable across original and alternative techniques. Factors such as item-level construct fidelity, poor item performance, and limitations in embedded Symptom Validity Tests (SVTs) predicting Performance Validity Test (PVT) outcomes may hinder scale refinement. These limitations, along with directions for future research, are discussed.