The objective of the present study was to provide normative data for Trail Making Test (TMT) time to completion and performance errors among cognitively normal older adults, and to examine TMT error rates in conjunction with time scores for pre-clinical and clinical Alzheimer's disease (AD) diagnostic decision-making. A sample of 526 individuals was classified into three diagnostic groups (normal controls, N=269; mild cognitive impairment, MCI, N=200; AD, N=57) by a multidisciplinary consensus conference. Results indicated that performance differed among the three groups for TMT A and B time scores as well as TMT B error rate. Diagnostic classification accuracy (i.e., sensitivity, specificity, and positive and negative predictive powers) is described for various combinations of the diagnostic groups. The findings show that TMT B time and errors are independently meaningful scores, and both therefore have clinical utility in assessing individuals referred for dementia evaluations.

While the FTT can be used to measure upper extremity motor ability, the GPT may be more strongly associated with general cognitive functioning in healthy adults. The FTT and GPT results presented will improve the utility of these tasks in clinical assessments of older adults.

The detection of suboptimal effort has become crucial in clinical neuropsychological practice in order to make accurate diagnoses, prognoses, and referrals. Symptom Validity Testing (SVT) has been the most commonly utilized model for assessing effort, and frequently includes recognition memory tasks. Some conflicting views on this model purport, however, that measures of effort gathered from a recognition memory paradigm do not necessarily extend to effort in other cognitive domains and other areas of performance. The present study sought to investigate whether performance on an SVT measure, which utilizes recognition memory, the TOMM, could predict performance on other measures that do not evaluate recognition memory or just memory per se in a group of mildly traumatic brain-injured litigants. Results indicated that poor performance on the TOMM was significantly correlated with poorer performance on the WAIS-R and the HRNB-A. Further, experimental exploration of these results indicated that the overall neuropsychological performance of litigants with suboptimal effort was poorer than what is generally expected from mild TBI individuals, and was also lower than the other mild TBI examinees in the study, who were not classified by the TOMM as exhibiting suboptimal effort. These findings support the proposition that poor effort as measured by recognition memory effort measures is not restricted to recognition and memory measures. In fact, in the present study it appears that a poor performance on the TOMM is predictive of a generalized poorer performance on standardized measures such as the WAIS-R and the HNRB-A.

It is well established that performance on the Wisconsin Card Sorting Test (WCST) tends to decline with advanced age, but the reason for this decline has not been established. The objective of the present study was to clarify this question using a qualitative approach to the task. The WCST was administered to 19 older adults and 25 younger participants. In addition to standard testing procedures, all participants were asked to verbalize their response strategy when placing each card. Results of this procedure implicate poor set shifting and set maintenance, consistent with reduced efficiency of feedback utilization, as the primary cause for age-related decline on the WCST.