One might argue that the decrease in the number of brain cells as a function of age could be the source of the functional age deficits in memory performance. However, this possibility seems less likely since the actual loss of neurons up to advanced age is relatively small. There are no good estimates of the loss of synapses. Golgi staining of cortical neurons would indicate that there is a loss with higher age. So far, however, the most convincing data of marked loss with age appear at the biochemical level. Most human data fail to demonstrate a decrease in cholinergic and serotonergic activity as a function of normal aging, although there is a loss of corresponding receptors. In AD/SDAT, however, there is a marked damage to these systems. Conceivably, acetylcholine may be providing informational rather than tone setting or balancing influence on memory function. This may explain the failure of cholinomimetic drugs to improve memory in AD/SDAT due to their inability to supply the informational properties of normal neuronal transmission. The catecholamines, noradrenaline and dopamine are both lost in normal aging and to a much higher degree in AD/SDAT. Animal data show that noradrenaline deficiency results in scattered attention. Such a pattern might also exist in the intact aged and through guidance by means of instructions, contextual cues, and a richer TBR information, the elderly are being forced to attend. This may promote and supersede the normal functions of the noradrenaline system by directions from external rather than internal influences, conceivably by potentiating the remaining noradrenaline neurons. The cortical motor areas are relatively spared from neuro-degenerative changes in normal aging and in AD/SDAT and this might provide a neuroanatomical basis for the elderly's and mildly to moderately demented patients' success in memory performance when motor action is involved. The role of dopamine in motor function and its stability with age in hippocampus may also provide a neurochemical basis for the preservation of memory when the subjects are allowed to act physically during encoding.