Our hypothesis is that there is 'wear and tear' in the brain, which is the basis of the process of aging, but that stimulation of brain function may slow down brain aging and diminish the risk for neurodegenerative diseases such as Alzheimer's disease (AD), possibly by activating repair mechanisms. Evidence supporting this hypothesis is presented in this review. During normal aging and in AD, cell loss is not as prominent a phenomenon as is often presumed. In fact, unaltered neuronal numbers have been reported in many brain areas in AD, e.g. in the nucleus basalis of Meynert (NBM) where the number of large neurons decreases while that of small neurons increases. Decreased neuronal activity is an essential characteristic of AD, and a substantial decrease of cerebral glucose metabolism may even precede cognitive impairments. Some hypothalamic neurons remain intact and active during the process of aging, others become even hyperactive, which may lead to disorders. Arginine vasopressin (AVP) levels were found to be higher in the elderly than in young subjects. There is an age-related, sex-dependent activation of the AVP neurons in the supraoptic nucleus (SON) and in the paraventricular nucleus (PVN), which may be the basis of analogous changes in the prevalence of hypertension and hyponatraemia in the elderly. No significant functional loss of magnocellular hypothalamic neurosecretory neurons were found in the SON or PVN in AD. The activity of the corticotrophin releasing hormone (CRH) neurons in the hypothalamic PVN is the basis for the activity of the hypothalamo-pituitary-adrenal (HPA) axis, which is activated during aging in a sex-dependent way, and even more activated in AD. The activated HPA axis is a risk for depression. Environmental stimulation increases brain reserve. An increase in time spent on intellectual activities was associated with a significant decrease in probability to get AD, and occupation has even a stronger indication of diminished risk for dementia. A series of observations showed that a dysfunctional clock may underlie the disordered rhythms in AD. Additional bright light improved the rest-activity rhythms, while giving bright light and/or melatonin to AD patients ameliorated the progression of cognitive and noncognitive symptoms. This implies that neurons affected by AD can still be reactivated if the right stimuli are applied. Unknown diffusible factors from the neural stem cells improve the survival of aged and degenerating neurons in postmortem human brain slice cultures. Gene therapy with nerve growth factor aimed at the NBM showed metabolic activation of various brain regions. A microarray study of the prefrontal cortex in the course of AD revealed an increased expression of genes related to synaptic activity and changes in plasticity during the very early pre-symptomatic stages, which is proposed to represent a coping mechanism against increased soluble β-amyloid levels. In brief, these examples of the 'use it or lose it' principle during the course of aging or AD now provide novel targets for the development of therapeutic strategies aiming at the prevention and treatment of AD.