Longitudinal Stability of Cognitive Self-Appraisal Calibration in MIDUS

Adults differ in how they rate their own memory relative to cognitive-test performance, but it is unclear whether this subjective-objective mismatch is transient or moderately stable. In MIDUS longitudinal completers from the national sample, we indexed cognitive self-appraisal calibration as the age-peer-relative discrepancy between standardized subjective memory appraisal and poorer objective cognitive performance at MIDUS 2 (2004-06) and MIDUS 3 cognitive testing (2013-17). Calibration showed moderate M2-to-M3 stability: ICC = .601 (bootstrap 95% CI [.570, .629]; N = 2,125) for the BTACT composite-based index and ICC = .444 (bootstrap 95% CI [.410, .478]) for an episodic-memory-matched sensitivity index. Stability was only slightly diminished after adjusting for age, education, depression, neuroticism, sense of control, mastery, and perceived constraints (fully adjusted composite ICC = .562, bootstrap 95% CI [.530, .593], complete-covariate N = 2,092). A descriptive four-cell profile model showed 54.3% same-profile persistence (unweighted Cohen's kappa = .389, bootstrap 95% CI [.361, .417]), with off-diagonal transitions concentrated in neighboring profiles. These findings provide preliminary measurement evidence that cognitive self-appraisal calibration is moderately stable and not fully accounted for by measured mood, personality, or control-belief alternatives. This is a preprint. It has not been peer-reviewed.