Objective: The assessment of memory in healthy older adults requires novel approaches to show sensitivity to established biomarkers. We hypothesized that healthy adults at risk for Alzheimer's disease (AD) as indicated by abnormal cerebrospinal fluid (CSF) biomarkers would show reduced primacy effects during list learning. Method: Participants were age, gender, and education-matched healthy adults (N = 34) who were part of the control cohort evaluated at an Alzheimer's Disease Center. Half of the subjects, however, demonstrated at least one CSF biomarker consistent with AD. There were no differences in the number of ApoE ɛ4 carriers between groups. CSF total tau, phospho-tau, and Aβ42 were determined using x-MAP technology. Participants were administered the 10-item CERAD list learning (CWL) test. CWL data collected closest in time to CSF measurement (Range 0–14 months) were recoded to capture temporal retrieval information across 3 learning trials. We tested our hypothesis with a retention index that differentially weighted recall from primacy regions. Results: There was no main effect of biomarker group on total recall, nor was there an interaction between group and learning trial, (p > .10). For primacy weighted retrieval, however, we observed a trend level interaction between group and list recall, F(1, 32) = 3.05, p < .10, suggesting different retrieval patterns in the two groups. Paired t-tests revealed a shallower primacy-weighted learning curve across trials in the pathological CSF group (p < .05). Conclusion(s): This result extends our previous findings of an association of probability of first recall with risk of AD in cognitively normal older adults, adding to the growing literature suggesting detectable cognitive and structural brain changes in this population.3