A study combining walking tests with interviews found one clear change when people performed a cognitive task while walking: they completed the task in less time than in the comparison condition. After the researchers corrected for the number of tests they ran, completion time was the only measure that differed significantly from zero. The result suggests a measurable difference in this experimental setup, but it does not by itself explain why the time changed.
The median change was a 20.60% lower completion-time cost in the dual-task condition. The reported matched-pairs rank-biserial effect was r = −0.78, and the Holm-adjusted p value was 0.0009. These figures describe a relative performance change within the study, not an absolute clinical outcome.
What the researchers tested
The research question was whether individual experiential characteristics and baseline capacities were related to cognitive-motor dual-task performance. To examine that question, the researchers combined treadmill walking and RPMT measurements in single-task and dual-task conditions with independent explicitation interviews, then integrated the quantitative and qualitative findings.
The analysis began with 44 participants at the first session. Ten did not return for the second session, and one was excluded because alcohol had been consumed within the preceding 24 hours. The final analytical sample therefore comprised 33 people who completed both sessions.
No a priori power calculation was performed, and the convenience sample was not prospectively sized for a prespecified effect or a multivariable model predicting profile membership. The design was therefore exploratory rather than a formal test of a planned effect or profile-prediction model.
The researchers tested eight cognitive and gait dual-task-cost measures against zero, using either a two-sided one-sample t-test or a Wilcoxon signed-rank test, depending on the data's compatibility with normality. They also reported effect sizes and 95% confidence intervals, and used Holm correction to account for multiple testing.
Most measures did not show a corrected difference
The completion-time result stood apart from the rest. None of the remaining cognitive or gait dual-task costs differed significantly from zero after Holm correction. In other words, the study did not produce a corrected statistical signal across the other measures it examined.
The researchers also compared participants who reported giving random responses with those who did not. Before correction, the random-response group had a lower DRS measure, with medians of −4.44% and +6.92%, and a higher DH measure, at 5.71% versus −8.55%. Neither comparison remained significant after correction: the adjusted p values were 0.176 for DRS and 0.297 for DH.
The team sorted the participants into four exploratory performance profiles. Twelve were classified as showing mutual interference, nine as a gait-priority trade-off, seven as a cognitive-priority trade-off, and five as mutual facilitation. Those counts show that all four profiles appeared in the sample, but they do not establish that any interview characteristic was associated with a particular profile.
Interviews added different ways of experiencing the task
The interviews were used to explore how people experienced the dual task, rather than to prove why their measured performance looked different. The qualitative material was independently double-coded, with disagreements discussed and resolved by consensus. The analysis identified distinct experiential strategies within each quantitative profile.
The authors interpreted distress and cognitive overload as features of mutual interference. Analytical or automatized strategies were described alongside cognitive-priority trade-offs, while relaxed immersion and positive affect were interpreted as compatible with mutual facilitation. Gait-priority cases were described in terms of bodily immersion and sensory detachment. These interpretations were hypothesis-generating, not causal explanations.
Why the result needs careful reading
The single-task and dual-task sessions were not counterbalanced or procedurally equivalent. The single-task condition was seated and paper-based, while the dual-task condition used a screen during treadmill walking with oral responses, and the same RPMT items were repeated. The comparison therefore cannot isolate concurrent dual-task effects from differences between the sessions and task formats.
The authors describe the work as methodological and hypothesis-generating rather than confirmatory. The convenience sample and exploratory profile counts limit what can be inferred from the findings, and the study does not support strong causal conclusions. The interview patterns therefore point to questions for future research rather than settling causal relationships.
The authors reported financial support for the work or its publication, with salary support from all organizations listed in the affiliations. They also refer readers to an OSF repository containing radar plots and verbatim explicitation-interview excerpts.
Paper data and sources
Original title: Between gait and cognition: an exploratory mixed-methods study of cognitive—motor dual-task profiles and lived experience
Authors: Buisseret F, Dierick F, Houdart E et al.
Journal/Repository: Frontiers in psychology
Status: Peer-reviewed
First online: 2026-10-07
DOI: Not available
Original paper · Full text