Subjective Cognitive Decline (SCD), the perception of deteriorating cognition in the absence of apparent impairment on objective testing, has gained momentum in recent literature as a risk marker for AD. Traditional neuropsychological assessments, while typically inclusive of a word list learning task, often do not include a comparable figure learning task. Growing evidence suggests that nonverbal assessments may be particularly sensitive to the earliest cognitive changes associated with Alzheimer’s disease. The Biber Figure Learning Test (BFLT), a visuospatial analogue to verbal list learning tasks, has been shown to associate with brain-based biomarkers of Alzheimer’s disease (AD; hippocampal volume, amyloid load). This study investigates the utility of the BFLT in capturing SCD above and beyond other cognitive measures sensitive to AD progression.

50 community-dwelling, cognitively normal individuals (78% White, 16% Black, 6% Other; 92% Non-Hispanic; 64% Female; Education M=17.1, SD=2.1; Age M=72.7, SD=6.2) participated in a study of SCD. All participants performed >-1.5 SD on clinical neuropsychological testing including a word list learning task. SCD was assessed using a 20-item scale querying individuals’ perception of difficulty across a range of memory and non-memory abilities in relation to others of the same age. Participants completed the BFLT, Loewenstein-Acevedo Scales of Semantic Interference and Learning (LASSI-L), Short-Term Memory Binding (STMB), and Face-Name Associative Memory Exam (FNAME), previously established as being sensitive to pre-clinical AD, were examined as predictors of SCD. A multiple regression adjusted for demographics (age, gender, education) was used to investigate the extent to which BFLT Trial 1 (T1) predicted SCD above and beyond these other cognitive measures sensitive to AD progression. Trial 1 of the BFLT was used based on a separate abstract examining which BFLT score was most highly associated with SCD (Kann et al., pending acceptance).

Adjusting for demographics, the present model accounts for 42% of the variance in SCD, while Biber T1 alone accounts for 20% and is the only significant individual predictor of SCD (β=-0.55, p=0.004). In contrast, other variables in the model independently accounted for less than 1% to 4% each (age β=-0.23, p=0.15; gender β=-0.15, p=0.34; education β=0.06, p=0.66; LASSI-L β=-0.11, p=0.55; STMB β=-0.03, p=0.85; FNAME β=-0.10, p=0.64).

The present study demonstrates the usefulness of the first learning trial of the BFLT as an independent predictor of SCD above and beyond other verbal and nonverbal measures sensitive to AD pathology. It also highlights the value of including even one trial of figure learning (< 5 minutes) in both clinical and research assessments seeking to capture cognitive changes which may be the earliest indicators of a neurodegenerative process. Ongoing longitudinal research is examining the predictive utility of the BFLT for future cognitive decline and transition to Mild Cognitive Impairment. Further research should explore the association between Biber T1, specifically, and neuropathological biomarkers of AD to further establish its utility as a portent of AD.