The belief that healthy older brains are substantially smaller than younger brains has been deemed incorrect by Dutch researchers.
The authors suggest that previous findings may have overestimated atrophy and underestimated normal size for the older brain.
The new study tested participants in Holland's long-term Maastricht Aging Study who were free of neurological problems such as dementia, Parkinson's disease or stroke.
Once participants were deemed otherwise healthy, they took neuropsychological tests, including a screening test for dementia, at baseline and every three years afterward for nine years.
MRI scans were used to measure seven different parts of the brain, including the memory-laden hippocampus, the areas around it, and the frontal and cingulate areas of the cognitively critical cortex.
The participants were later divided into two groups: one group with 35 cognitively healthy people who stayed free of dementia (average starting age 69.1 years), and the other group with 30 people who showed substantial cognitive decline but were still dementia-free (average starting age 69.2 years).
In contrast to the 35 people who stayed healthy, the 30 people who declined cognitively over study-period showed a significant effect for age in the hippocampus and parahippocampal areas, and in the frontal and cingulate cortices.
In short, among the people whose cognition got worse, older participants had smaller brain areas than younger participants.
Thus, the seeming age-related atrophy in gray matter more likely reflected pathological changes in the brain that underlie significant cognitive decline than aging itself, wrote the authors.
As long as people stay cognitively healthy, the researchers believe that the gray matter of areas supporting cognition might not shrink much at all.
If future longitudinal studies find similar results, our conception of 'normal' brain aging may become more optimistic," said lead author Saartje Burgmans, who is due to receive her PhD later this year.
The study appears in journal Neuropsychology.