A – 120 The Loewenstein-Acevedo Scales for Semantic Interference and Learning and Plasma Phosphorylated Tau 217: Differentiating Cognitive Status in a Community-Based Sample

Abstract Objective The Loewenstein-Acevedo Scales for Semantic Interference and Learning (LASSI-L) detects early deficits through proactive semantic interference, failure to recover from interference, and semantic intrusion errors (SIEs). Plasma p-tau217 is a promising biomarker, but its relationship to cognitive markers like the LASSI-L remains underexplored. This study examined the relationship between LASSI-L performance and plasma p-tau217 in a community-based sample while comparing LASSI-L to a traditional memory measure (i.e., Benson Figure). Method Ninety-one older adults were deemed cognitively unimpaired (CU) p-tau217- (n = 46), amnestic mild cognitive impairment (aMCI) p-tau217- (n = 21), and aMCI p-tau217+ (n = 24). LASSI-L and Benson were not used for group classification to avoid circularity. Results Controlling for age, education, and MMSE, both aMCI groups performed significantly worse on the Benson than CU group (p=.004). Similarly, the CU group outperformed both aMCI groups on LASSI-L Cued B1 Recall (p = .035) and Cued B2 Recall (p = .007) with no significant differences between the two aMCI groups. However, SIEs were significantly elevated in the aMCI p-tau217+ group compared to both CU and aMCI p-tau217– groups for Cued B1 (p = .010) and Cued B2 intrusions (p = .002). Conclusion These findings demonstrate that elevated SIEs on the LASSI-L effectively differentiate CU from those with aMCI. While recall impairments were observed across aMCI groups, only the p-tau217+ group showed significantly higher intrusion errors, suggesting a potential cognitive signature of early tau-related pathology. The findings highlight the utility of the LASSI-L as a compelling alternative to traditional memory assessments for early detection.