Objective: Embedded measures of effort are commonly administered and interpreted although their use in older populations is under-researched. This study looks at the CVLT's Forced Choice Index (FC), the traditional Reliable Digit Span (RDS), and two previously proposed WAIS-IV RDS versions (Reese, Suhr, & Riddle, 2011) in older individuals. 2. Method: 221 patients from a memory disorder clinic who had obtained a consensus neurological and neuropsychological diagnosis of normal aging, Mild Cognitive Impairment [MCI], or early Alzheimer's (AD) dementia and had been administered both the WAIS-IV Digit Span Subtest and the CVLT-II were included. FC and the RDS versions were compared across the three groups. 3. Results: The CVLT FC results are generally consistent with the prior smaller study of Clark, et. al., (2012). Significant group differences (p < .001) of AD < MCI < normal emerge with only 44% of those with AD, 69% of those with MCI, and 100% of normals achieving perfect FC scores. All three RDS versions show significant group differences when comparing either MCI or normals to AD but cannot separate MCI from normals. Depending on the RDS version, only between 41-62% of those with AD and 66–78% of those with MCI or judged to be normal exceeded the proposed cutoffs. Conclusion: The CVLT FC shows a pattern of poorer performance as MCI appears and transitions to AD. The three versions of the RDS examined separate AD from the similar performing MCI and normals. All three groups have a high rate of failure to reach proposed cutoffs and should be used cautiously.