Spatial alignment of audio–haptic directional feedback shapes non-visual hand guidance

Non-visual wearable guidance relies on continuous directional cues delivered through sound or touch. Whether combining modalities improves performance, and how directional information should be organised across them, remain unclear. We compared four wrist-worn feedback conditions during a reaching task: Aligned (audio and vibration conveying the same direction), Axis-split (audio guiding left–right, vibration guiding depth), and two single-modality conditions. Endpoint accuracy was similar across conditions. However, movement behaviour differed: aligned feedback produced faster and more direct movements, whereas axis-split required more corrective actions and was less efficient than either modality alone. The alignment advantage was strongest at cardinal target angles, where the axis-split mapping produced conflicting single-channel guidance while aligned guidance maintained redundant two-channel coverage. This advantage was universal across participants and persistent across experimental cycles. These findings demonstrate that how directional information is structured across modalities matters more than how many modalities are used. The results inform the design and evaluation of multimodal wearable systems.