Computer Use Could Help Predict Early-Stage Alzheimer's (thestack.com)
An anonymous reader cites an article on The Stack: Infrequent use of a computer in later life could be an early sign of reduced cognitive ability, according to research from Oregon Health and Science University. A study, which involved 27 'cognitively-healthy' adults aged 65 or older, used MRI scans to measure the volume of the hippocampus -- a small area of the brain with a key role in memory function used to pick up early biomarkers of dementia and Alzheimer's. Data was also collected on computer use among participants via mouse movement detection software. Results showed an additional hour of computer use each day was linked to a 0.025% larger hippocampal volume(PDF), thus indicating that lower computer usage could help predict cognitive decline.
and her computer usage hasn't declined at all. She actually ends up spending MORE time on it since using apps like QuickBooks takes her longer to use.
Remember, if you run an experiment like this one a hundred times, you'll get this result once on average by pure statistical chance. The other 99 don't get published. There were a couple other things they measured (education and MMSE) that could have been interesting enough to publish, too, so knock the odds down to 1/33. Or, since they tried a couple different methods to normalize the data, odds are up to 1/20 that this study would have produced numbers this significant on a single variable by pure chance.