Previous research has been equivocal about whether cognitive training helps cognitively healthy older adults. One recent review concluded that cognitive training could help slow age-related decline in a range of cognitive tasks; another found no evidence that such training helps slow or prevent the development of Alzheimer’s in healthy older adults. Most of the studies reviewed looked at single-domain training only: memory, reasoning, processing speed, reading, solving arithmetic problems, or strategy training (1). As we know from other studies, training in specific tasks is undeniably helpful for improving your performance at those specific tasks. However, there is little evidence for wider transfer. There have been few studies employing multi-domain training, although two such have found positive benefits.

In a new Chinese study, 270 healthy older adults (65-75) were randomly assigned to one of three groups. In the two experimental groups, participants were given one-hour training sessions twice a week for 12 weeks. Training took place in small groups of around 15. The first 15 minutes of each hour involved a lecture focusing on diseases common in older adults. The next 30 minutes were spent in instruction in one specific technique and how to use it in real life. The last 15 minutes were used to consolidate the skills by solving real-life problems.

One group were trained using a multi-domain approach, involving memory, reasoning, problem solving, map reading, handicrafts, health education and exercise. The other group trained on reasoning only (involving the towers of Hanoi, numerical reasoning, Raven Progressive Matrices, and verbal reasoning). Homework was assigned. Six months after training, three booster sessions (a month apart) were offered to 60% of the participants. The third group (the control) was put on a waiting list. All three groups attended a lecture on aspects of healthy living every two months.

All participants were given cognitive tests before training and after training, and again after 6 months, and after one year. Cognitive function was assessed using the Stroop Test, the Trail Making test, the Visual Reasoning test, and the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS, Form A).

Both the multi-domain and single-domain cognitive training produced significant improvement in cognitive scores (the former in RBANS, visual reasoning, and immediate and delayed memory; the latter in RBANS, visual reasoning, word interference, and visuospatial/constructional score), although single-domain training produced less durable benefits (after a year, the multi-domain group still showed the benefit in RBANS, delayed memory and visual reasoning, while the single-domain group only showed benefits in word interference). Booster training also produced benefits, consolidating training in reasoning, visuospatial/constructional abilities and faster processing.

Reasoning ability seemed particularly responsive to training. Although it would be reasonable to assume that single-domain training, which focused on reasoning, would produce greater improvement than multi-domain training in this specific area, there was in fact no difference between the two groups right after training or at six months. And at 12 months, the multi-domain group was clearly superior.

In sum, the study provides evidence that cognitive training helps prevent cognitive decline in healthy older people, that specific training can generalize to other tasks, but that programs that involve several cognitive domains produce more lasting benefits.