B-179: Examination of Updated Stroop Color and Word Test Norms in a Diverse Veteran Sample Without Cognitive Disorders

Abstract

Objective: Given the need to validate updated tests/norms before adopting them clinically (Bush et al., 2018), we examined the application of updated Stroop Color and Word Test (Stroop) norms in a diverse older adult sample without neurocognitive disorders. Stroop Normative Update (Golden et al., 2022) was compared to previous norms (Golden & Freshwater, 2002; Ivnik et al., 1996; Lucas et al., 2005). Method/Data Selection: Eighty-five veterans (M_age=67.71) who passed performance validity testing and had no neurocognitive disorder were included. The sample was 61.2% White/Non-Latino, 94.1% Male, and 74.1% Monolingual-English; 63% had a primary psychiatric diagnosis. Repeated-measures analyses of variance (ANOVAs), paired-samples t-tests, and Wilcoxon signed-rank tests examined T-scores across norms. Results/Data Synthesis: Repeated-measures ANOVA found significant differences in Color-Word (F[1.80, 151.11]=573.30, p<.001) and Interference (F[1,84]=185.58, p<.001) T-scores across norms. Color-Word T-scores were lowest for Golden-2002 (M=44.55, SD=7.55), followed by MOANS (M=46.09, SD=9.05), Golden-2022 (M=46.84, SD=7.78), and MOAANS (M=56.21, SD=7.55). Effect sizes were small for MOANS/Golden-2022 contrast (η2=.05) and large for other contrasts (η2=.62+). Interference scores were lower with Golden-2022 (M=44.75, SD=6.79) compared to Golden-2002 norms (M=45.51, SD=6.37; Cohen’s D=1.48). Findings were similar across subgroup analyses. Golden-2002 Word and Color T-scores were ~1 SD below expected means (T-score=50), whereas Golden-2022 Color scores were closer to expectations. Conclusions: Updated norms for Color-Word and Interference were consistent across race/ethnic groups, and yielded scores closer to expected means than prior norms. Compared with older adult norms (MOANS/MOAANS), the updated norms were less likely to overcorrect Color-Word. Findings support using updated Stroop norms, though Word T-scores may be lower in clinical-psychiatric samples.