Can brief brain or body training strengthen older adults' white matter?
Ten weeks of computer brain training or exercise did not change white matter integrity in older adults at risk of dementia, even though a cognitively active lifestyle was linked to better cognitive change.
Source
No Evidence That Short-Term Cognitive or Physical Training Programs or Lifestyles Are Related to Changes in White Matter Integrity in Older Adults at Risk of Dementia
Study at a glance
- Design
- Other — Non-randomised controlled trial with matched/minimisation allocation to computerised auditory and working-memory training, multimodal physical training or a wait-list control, with DTI and cognitive testing at pretest, posttest and 3-month follow-up, plus observational analyses of self-reported lifestyle.
- N
- N=39 · 47 were allocated (13 cognitive training, 14 physical training, 20 controls); 39 were included in the analysis; training-performance analyses used only 9 people with usable data.
- Population
- Community-dwelling older adults in Germany at risk of dementia (memory complaints), heavily screened
- Outcome
- Fractional anisotropy composite of the genu of the corpus callosum, fornix and hippocampal cingulum; global cognition, episodic memory and executive function composites
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
Older adults with memory complaints were allocated (without randomisation, but matched on age, education, sex and cognitive screening score) to 50 hours of home-based auditory and working-memory training, a multimodal exercise programme, or a wait-list control. White matter integrity was measured with diffusion tensor imaging as fractional anisotropy in three tracts affected by aging and Alzheimer's disease, and cognition with a test battery, before training, right after it and three months later. Self-reported cognitive and physical lifestyles were also related to changes in brain and cognition.
What they found
Neither training programme changed the white matter composite compared with controls at posttest or follow-up; effect sizes were near zero and their 95% confidence intervals excluded medium effects. Lifestyles did not predict white matter change either. A more cognitively varied lifestyle did predict improvements in global cognition and episodic memory (standardised betas of at least 0.40), and this link held after accounting for white matter change. At baseline, people with better cognitive-training skill had higher white matter integrity (r = 0.68), and fitter people had higher fornix integrity (r = 0.35).
The limits
What it doesn't show
Allocation was not randomised, participants could not be blinded, and the analysed sample was small (39 people, with training-skill analyses on only 9), so small effects could easily be missed. Only three pre-chosen tracts were examined, so changes elsewhere, for instance in motor tracts that exercise is known to affect, would be invisible. The training lasted only ten weeks, so the study says nothing about longer-term programmes, and the lifestyle findings are observational and cannot show cause. Baseline correlations between fitness or skill and white matter do not show that training would change the tracts.
Key terms
- White matter integrity
- An index of the microstructural condition of nerve-fibre bundles that connect brain regions, often reduced in aging and Alzheimer's disease.
- Diffusion tensor imaging (DTI)
- An MRI technique that measures how water diffuses in tissue, revealing the orientation and organisation of white matter fibres.
- Fractional anisotropy (FA)
- A DTI measure from 0 to 1 of how directional water diffusion is; higher values usually indicate more coherent, intact fibre tracts.
- Fornix
- A white matter tract carrying output from the hippocampus, important for episodic memory and vulnerable in Alzheimer's disease.
- Minimisation allocation
- A non-random method of assigning participants to groups so as to keep groups balanced on key characteristics.
- Hedges' g
- A standardised effect size similar to Cohen's d with a correction for small samples.
Flashcards
0 of 11 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
What was the main white matter finding for the two training programmes?
Common questions
Does this prove that brain training and exercise never affect white matter?
No. It shows no detectable change in three tracts after ten weeks in a small at-risk sample; longer programmes, other tracts or other populations might show effects.
Why is it informative that the confidence intervals excluded medium effects?
A non-significant result alone could reflect low power; showing that plausible effect sizes are all small makes a meaningful short-term benefit unlikely in this setting.
If white matter didn't change, why did cognitively active people improve cognitively?
The link between cognitive lifestyle and cognitive gains was not explained by white matter change, suggesting other brain mechanisms, though the lifestyle data are self-reported and observational.
More on Neurocognitive aging