Computer training helped thinking in mild memory problems. Dementia showed no clear gain.
HelpedThe short version
In mild memory problems, thinking improved more with training than in other groups.

- 17
- studies pooled
- 686
- older people with memory or thinking problems
- 1 tool
- tested
The review
Who took part
686 people with mild memory and thinking problems (17 studies) and 389 with dementia (12 studies). Average ages were 66 to 81, and about half of the first group and 6 in 10 of the second were women.
What they did
People did computer exercises for at least four hours, usually with a helper. Most of the other groups did other activities, such as paper tasks or education. The rest had no contact.
What happened
With mild memory problems, overall thinking and attention improved more with training, but planning and control skills did not. With dementia, the overall gain was slight, and without two studies there was no clear gain. Gains faded once training stopped, and too few studies followed people for long enough to know whether they last.
What the review found
- HelpedOther outcomes
In mild memory problems, thinking improved more with training than in other groups.
Pooled analysis of 17 studies with 686 people · vs other activities or no contact · ⚠ most studies had serious flaws
- HelpedFocus
In mild memory problems, attention improved more with training than in the other groups.
Pooled analysis of 6 studies · vs other activities or no contact · ⚠ most studies had serious flaws
- No differenceOther outcomes
Executive function, the skills for planning and control, did not improve more than in the other groups.
Pooled analysis of 13 studies · vs other activities or no contact · ⚠ most studies had serious flaws
- HelpedOther outcomes
In people with dementia, thinking improved a little more with training, but two studies did most of that.
Pooled analysis of 11 studies · vs other activities or no contact · ⚠ most studies had serious flaws
Source
Computerized Cognitive Training in Older Adults With Mild Cognitive Impairment or Dementia: A Systematic Review and Meta-Analysis (2017)
DOI: 10.1176/appi.ajp.2016.16030360 (opens in a new tab)

