Adding cognitive tasks to water-based exercise did not produce a clear improvement in overall cognition in a randomized trial of community-dwelling women aged 60 and older. In the main intention-to-treat analysis, total MoCA scores showed no significant effect of time, group, or the group-by-time comparison (p > 0.05). In plain terms, the study did not find a different pre-to-post cognitive change between dual-task exercise, conventional aquatic exercise, and a no-exercise control.
The study took place in Porto Alegre, Brazil. Of 186 interested people, 119 were excluded and 67 were randomized: 23 to dual-task water-based exercise, 23 to conventional water-based exercise, and 21 to control. Outcomes were measured at baseline and one week after the intervention by assessors who remained blinded to group assignment.
The added task was brief
Both exercise groups used the same session structure and intensity progression. The dual-task group added cognitive stimuli to the motor exercises: during a 30-minute main phase, participants completed 10 three-minute sets, with two minutes of movement alone and one minute combining movement with a cognitive task. That added up to 10 minutes of dual-task exposure; the conventional group did not receive the added cognitive stimuli.
The main outcome was the total score on the Montreal Cognitive Assessment, or MoCA, the study's measure of global cognition. Researchers also tested specific MoCA domains, verbal fluency, Trail Making, timed mobility, a six-minute walking test, 30-second sit-to-stand, handgrip, knee-extensor strength and muscle activity. The primary analysis used available-case intention-to-treat data without filling in missing post-intervention values. A separate per-protocol analysis required complete before-and-after assessments and at least 70% attendance.
The overall result was not the only result
Some secondary cognitive measures moved, but the pattern was narrow. Orientation showed a change over time (p = 0.006). Semantic verbal fluency increased in the dual-task group, fell in the conventional exercise group and stayed stable in control, producing a group-by-time result of p = 0.003. Phonemic fluency and Trail Making A and B did not show significant effects (p > 0.05). Because these were secondary tests, and because repeated testing can produce chance or practice-related findings, they do not overturn the main MoCA result.
Physical results were just as uneven. The standard Timed Up and Go test did not change significantly. The dual-task version showed a group-by-time difference (p = 0.032), but improvement occurred only in the control group. The six-minute walk produced a between-group difference (p = 0.020), yet showed neither an overall time effect nor a group-by-time interaction. That combination does not amount to a consistent mobility advantage for either exercise program.
Strength and muscle-activation measures offered no single direction either. Handgrip strength increased only in the conventional exercise group (p = 0.037). Maximum isometric knee-extensor strength increased in control while the intervention groups were maintained (p = 0.014). Activity in the rectus femoris increased in both exercise groups (p = 0.028), while lower-limb endurance and vastus lateralis activity showed no significant effects (p > 0.05).
Attendance complicates the picture
Attendance makes the findings harder to read as a test of the full exercise plan. Mean attendance was 55.2% in the dual-task group and 65.7% in the conventional group. No intervention-related adverse events were reported, but participants received a moderate share of the planned sessions. The result therefore speaks most directly to the dose people actually completed, not to the full planned dose.
The per-protocol check, which focused on participants with complete assessments and enough attendance, produced a different picture. Total MoCA improved in the conventional exercise group, while dual-task and control scores were maintained. The dual-task Timed Up and Go improved only in control (p = 0.005), the six-minute walk improved over time across groups, handgrip rose in conventional exercise, knee-extensor strength fell in that group, and rectus femoris activation rose in dual-task exercise. These estimates can be shaped by selection into the higher-adherence group, so the intention-to-treat analysis remained the main interpretation.
What the trial can and cannot say
The trial supports a limited conclusion: adding a brief cognitive-motor component to aquatic exercise did not show a reliable overall cognition benefit under the conditions tested. It does not establish long-term effects on dementia, falls, independence or quality of life, and its findings do not directly apply to people outside the eligible older-women population or to other aquatic exercise programs. Higher doses, stronger adherence and longer-term outcomes remain open questions.
Disclosures
Financial support was reported from CAPES Brazil, the UFRGS Pró-Reitoria de Extensão e Pesquisa and, in part, a CNPq/MCTI 2023 Emerging Groups call. The authors declared no commercial or financial conflicts and said raw supporting data would be made available without undue reservation.
Paper data and sources
Original title: Added effects of dual-task water-based exercise on global cognition and physical performance in older women: a three-arm randomized controlled trial
Authors: Miranda C, Borges L, e Silva A et al.
Journal/Repository: Frontiers in physiology
Status: Peer-reviewed
First online: 2026-10-10
DOI: Not available
Original paper · Full text