As people grow older, their subjective and objective sleep patterns change: sleep is often experienced as less deep, more broken, less refreshing – and these alterations find their objective correlate in polygraphic sleep recordings. Reductions in high amplitude slow wave sleep, rapid eye movement (REM) sleep and sleep maintenance are the best documented of these. Besides, there are changes in the EEG pattern during sleep (fewer and slower sleep spindels, fewer K-complexes and other phasic events). Daytime EEG recordings in the elderly are characterized by slowing of the dominant alpha rhythm, diffuse or localized slow waves and reduced reactivity to stimuli. Only few studies, however, have addressed the question of how daytime EEG alterations are related to changes of the sleep polygram, and how these electrophysiological parameters relate to measures of mental performance which also undergo changes with aging. A review of published results and data from our own studies suggest that, within the non-pathological range, few correlations exist between polygraphic sleep, daytime EEG and mental performance data if age as an independent factor is kept constant. The only relations that were significant in some of the studies had opposite directions in different subjects’ samples. Thus, until more is known, these 3 areas of assessment should be studied and conceptualized separately. Our lack of understanding in this field is further illustrated by results of drug studies: compounds with confirmed effects on mental performance and mood in young subjects, such as amphetamine, fail to be useful stimulants or antidepressants in the elderly, and drugs like co-dergocrine mesylate (Hydergine ®) which are of use in mentally deteriorating old persons have no effects on vigilance and mental performance in young, healthy subjects. Therefore, extrapolations from one level of assessment to another and from experiments in young subjects to studies in the elderly appear unwarranted at the present