An Outdoor Dual-Task Study on Cognitive-Motor Interference during Exoskeleton-Assisted Walking

Norman Riedel^1*Giorgos Marinou²Katja Mombaur^3,4Barbara Deml¹

• ¹Institute of Human and Industrial Engineering, Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany
• ²Institute of Computer Engineering, Heidelberg University, Heidelberg, Baden-Württemberg, Germany
• ³Institute for Anthropomatics and Robotics, BioRobotics Lab, Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany
• ⁴Systems Design Engineering and the Department of Mechanical and Mechatronics Engineering, University of Waterloo, Waterloo, Ontario, Canada

Controlling a powered lower-limb exoskeleton may increase the need demand forof cognitive resources due to mechanical constraints and assisting torques that disrupt natural movement. This study employed a dual-task paradigm to investigate cognitive-motor interferences and short-term familiarization effects in an outdoor walking experiment with twenty healthy adults using a powered lower-limb exoskeleton. In contrast to unassisted walking, performing a serial subtraction task during exoskeleton-assisted walking led to a slight increase in gait velocity and a reduction in stride time variability. This suggests that externalizing attention may facilitate the coordination with external rhythmic cues provided by the exoskeleton. Concurrently, cognitive performance, measured by correct response rates, declined during exoskeleton-assisted walking, indicating a posture-first strategy. Short-term familiarization during exoskeleton-assisted walking reduced perceived workload and improved cognitive performance, yet cognitive performance remained lower than in both the seated control condition and unassisted walking. This suggests that walking with the exoskeleton continues to require significant attentional resources. These findings emphasize the necessity of evaluating the cognitive fit of exoskeletons to ensure safe human-exoskeleton interaction.