The aim of this study is to provide normative data for a phonological alternating task (FAT), a semantic alternating task (SAT) and an excluded letter task (ELT). The tasks were administered to 146 Flemish-speaking, cognitively healthy elderly. Data from 102 were used and were classified according to the significant variables. Subsequently, these tasks were administered to seven patients diagnosed with mild neurocognitive impairment (mild cognitive impairment, MCI) and seven patients with onset dementia of the Alzheimer type (DAT). Results of the standard study show that the level of education is a significant variable for all complex VFT and age for the SAT and the ELT, while age related deterioration is highest for the ELT. The error rate is highest for the ELT and lowest for the SAT. Analysis of the time duration shows that data should be collected for at least 2 min. The patients scored significantly lower than the normgroup of healthy adults. The error rate is highest for the SAT and lowest for the ELT.