Dual-Task Training for Older Adults: What the Trials Show

Training the brain and body at the same time appears to improve both more than training either one alone—at least in several randomized trials of older adults. The best evidence comes from a 24-week tai chi trial in people with mild cognitive impairment (MCI) or memory concerns. But the result does not mean every dual-task program works, or that hearing loss has already been shown to blunt the benefit. Those questions remain unsettled.

Why walking while thinking is a meaningful test

Walking through daily life rarely means walking through an empty hallway. You may be crossing a street, answering a question, scanning for a curb, and ignoring a rogue shopping cart. Those tasks compete for attention. When cognitive capacity declines, gait often slows or becomes less stable under that pressure.

A 2024 GAIT cohort study of 711 adults across normal cognition, MCI, and dementia found that gait worsened as cognitive impairment progressed. Walking while naming animals best differentiated the cognitive groups among the tested gait conditions. That is an association, not proof that dual-task walking causes cognitive decline. It does, however, support dual-task gait as a useful outcome for trials that aim to improve cognitive-motor function.

The biological rationale is plausible. Functional near-infrared spectroscopy, or fNIRS, measures changes in cortical blood oxygenation during activity. A 2024 validation study found that fNIRS can track executive demand during walking in younger adults, older adults, and people with Parkinson's disease. Other work reports greater prefrontal activation during dual-task than single-task walking. More activation does not automatically mean better brain health; it may reflect compensation for a task that has become less automatic. Still, the pattern fits a limited-attention model: walking and thinking draw on overlapping executive resources.

The strongest MCI trial used cognitively enhanced tai chi

The most persuasive trial in this group is a 2023 Annals of Internal Medicine randomized controlled trial led by Fuzhong Li and colleagues. It enrolled 318 community-dwelling older adults with MCI or self-reported memory concerns and assigned them to one of three programs: cognitively enhanced Tai Ji Quan, standard Tai Ji Quan, or stretching.

Participants trained by videoconference twice a week for 60 minutes over 24 weeks. The cognitively enhanced program paired tai chi with added cognitive challenges, whereas standard Tai Ji Quan matched the motor format without that augmentation. The co-primary outcomes were Montreal Cognitive Assessment (MoCA) scores and dual-task walking cost.

The enhanced program outperformed both standard tai chi and stretching on global cognition and dual-task walking cost. Its advantage over standard tai chi matters because the groups shared the movement practice; the difference points toward the cognitive component rather than exercise time alone. Benefits reportedly persisted through the 48-week follow-up among a large majority of enrolled participants.

The publication supports the direction of benefit, but the available report does not establish reader-ready effect sizes, confidence intervals, or the percentage change in walking cost. Those numbers should not be improvised. The study also lacked a no-exercise group, so it cannot quantify the benefit over doing nothing. It tells us that this specific combined program performed better than two active comparators, not the size of its absolute effect in the community.

Smaller trials point in the same direction

A 2023 randomized trial by Wiśniowska et al. tested the basic idea more cleanly, though in a smaller and healthier group. Fifty-eight older adults completed four weeks of cognitive-motor dual-task training, cognitive-only training, or motor-only training. They trained three times weekly for 30 minutes; the average age was 71.2 years.

The dual-task group reduced its interference effect more than either single-task group, based on planning latency and error rate. That result fits the larger tai chi study, but it cannot establish a minimum effective dose for MCI. Each group had only about 19 to 20 completers, follow-up lasted four weeks, and the participants were cognitively healthy.

A separate randomized comparison of progressive single- and dual-task training in healthy older adults also supports dual-task practice for gait and cognition, although the available published details do not allow a reliable numerical synthesis. The convergence is encouraging; it is not a license to treat every puzzle-plus-balance class as interchangeable.

Delivery may be flexible. In a 2022 randomized trial, 50 adults with MCI received exergaming-based tai chi, traditional tai chi, or no intervention. The active groups completed 36 sessions—three 50-minute sessions weekly for 12 weeks. The exergaming study reported improvements in cognitive function and dual-task walking, suggesting that technology-supported tai chi can be feasible. It does not prove that every digital version matches the enhanced tai chi protocol used in the larger Annals trial.

Durability looks possible, with an important caveat

A 2025 two-year follow-up trial provides the longest follow-up in this evidence base for tai chi and cognition. It studied adults aged 60 or older with both MCI and type 2 diabetes. The result adds useful long-term context, but diabetes changes the interpretation: glycemic control, vascular risk, and medications can each affect cognitive trajectories. It cannot tell us that the same two-year outcome would occur in an MCI population without diabetes.

Hearing loss is a credible concern, not a settled modifier

Hearing has become the most tempting wrinkle in this story. Effortful listening can demand executive attention, so untreated hearing loss could plausibly leave fewer resources for a person who is simultaneously walking and performing a cognitive task. Observational findings cited in a 2026 SYNERGIC trial sub-analysis link hearing loss with higher dual-task costs in gait speed and cadence among older adults.

But the key clinical claim has not been confirmed: SYNERGIC was designed to test whether audiometric hearing ability predicts baseline dual-task performance and modifies response to combined exercise and cognitive training in adults aged 60 to 85 with MCI. The available findings do not establish a hearing-loss-by-training interaction, an effect size, or that hearing correction restores training gains.

Likewise, a 2023 multicenter randomized trial tested at-home computerized executive-function training in normal-hearing older adults and hearing-aid users, but the available evidence does not provide outcome estimates. Addressing hearing difficulty is sensible clinical care, especially before a program that relies on spoken instruction. It should not be sold as a proven way to amplify dual-task training until the trials report that result.

Physical dual-tasking may not be the only route

The tai chi findings do not prove that physical-cognitive simultaneity is uniquely necessary. A 2024 randomized trial of transcranial direct current stimulation (tDCS) plus interactive computerized cognitive training in older adults with MCI found cognitive and gait improvements without a physical dual-task component. Because the intervention combined stimulation and cognitive training, it cannot isolate the contribution of tDCS. It does show that a non-physical approach can affect cognitive-gait outcomes.

Virtual reality is another possible delivery route, but the evidence is uneven. A 2025 systematic review and meta-analysis found substantial variation in VR dual-task results across protocols, tasks, outcomes, and populations. Small studies and publication bias may also inflate pooled estimates. A headset is a platform, not a mechanism.

What to do with this evidence

For an older adult who can safely exercise, a structured program that pairs movement with a real cognitive demand has better trial support than practicing cognition or movement alone. The best-validated MCI template is cognitively enhanced Tai Ji Quan delivered twice weekly for 60 minutes over 24 weeks. That does not make it the only reasonable option, and it does not establish the shortest effective program.

Start with safety: a clinician or physical therapist should help tailor training for people with falls, severe frailty, major mobility limits, or neurologic disease. If hearing is impaired, make sure instructions and communication are accessible and pursue appropriate hearing evaluation. The rationale is strong; the claim that it improves training response still awaits data.

The evidence supports practicing the problem you want to solve: moving safely while attention is divided. It does not yet justify promises about dementia prevention, universal benefit, or a hearing-aid multiplier effect.