Objective: Paired Associates Learning (PAL) test assesses brain functions in those brain regions affected earliest by Alzheimer's dementia. The aim of the present study was to assess the usability of our implementation of the PAL test for screening mild cognitive impairment.
Methodology: Based on Petersen criteria, 14 out of the 63 subjects were diagnosed with mild cognitive impairment. Visuospatial learning was assessed by our implementation of PAL test. The ability of the PAL test to differentiate between study groups was compared to the Addenbrook Cognitive Examination (ACE) and to the Mini Mental State Examination (MMSE). Linear logistic regression was used for statistical analysis, and the results are presented as Receiver Operating Characteristics (ROC) curves. All analyses were performed by SAS 9.2.
Results: All the results of neuropsychological tests differed significantly between the study groups. However, considerable difference could be detected between the tests regarding specificity and sensitivity. The PAL test reached the sensitivity of the ACE, while its specificity was slightly under the ACE.
Discussion: The PAL test developed in the framework of the present study is found to be able to differentiate between MCI and healthy controls. It outperformed the MMSE in terms of sensitivity and specificity, while it needs comparable time to perform. Its sensitivity, the important parameter for screening, is comparable to ACE, while it needs significantly shorter time and less assistance.