Exploring gait automaticity and prefrontal brain activity during single and dual-task walking in aging and Parkinson’s disease

Abstract Background Walking while performing a concurrent cognitive task leads to cognitive-motor interference, resulting in slower and more variable gait. This is particularly the case in Parkinson’s disease (PD), where dual-task situations exacerbate walking impairments, increasing fall risk and reducing quality of life. Cognitive-motor impairment has been linked to excessive attentional demands due to reduced locomotor automaticity. Neuroimaging studies suggest over-reliance on prefrontal resources potentially reflecting compensatory mechanisms. However, few studies link automaticity, performance, and cognitive capacity to prefrontal activity, particularly in PD. In older adults (OA) and people with PD, this study aims to: (1) describe dual-task effects and prefrontal cortical activity during walking with and without a dual-task, (2) determine the connection between prefrontal cortical activity and step time variability as a measure of gait automaticity, (3) explore associations between prefrontal cortical activity and other measures of gait automaticity and prioritization and (4) investigate executive function as a potential moderator in the compensatory relationship. Methods Data from 44 OA and 37 people with PD walking with and without an auditory Stroop task were analyzed. Gait variables were measured using inertial measurement units, and prefrontal activity was assessed with functional near-infrared spectroscopy (fNIRS). Executive function was determined with a trail making test. Data analysis involved linear regression models to explore relationships between prefrontal activity, gait automaticity, and executive function. Results Most participants had a cognitive priority trade-off when dual-tasking, and the OA group had more prefrontal activity compared to the PD group during single-task and dual-task walking. For PD there was a significant positive relationship between step time variability and prefrontal activity (β = 0.38, T = 6.26, p < 0.01), while OA had a relationship between age and prefrontal activity (β = 0.53, T = 2.33, p = 0.04). Secondary analyses showed relationships between prefrontal activity and dual-task cost of gait speed (β = 0.25, T = 2.90, p = 0.02) and Stroop response time (β = 0.27, T = 3.10, p = 0.01) in PD, but not in OA. No moderation effects were detected in the relationship between gait automaticity and prefrontal activity. Conclusions In PD, loss of gait automaticity is linked to increased prefrontal activity, suggesting compensatory mechanisms. In OA, prefrontal activity during walking seems to be primarily age-related.