Guiding sightless people one dimension at a time: how the strategy used in spontaneous voice instructions benefits sonification-based guidance

Dataset for the paper "Guiding sightless people one dimension at a time: how the strategy used in spontaneous voice instructions benefits sonification-based guidance", submitted to PLOS One. Open the README file for more information about the data. Abstract of the paper : Sensory substitution devices (SSDs) aim to assist visually impaired people by converting spatial information into non-visual modalities, such as sound. Among existing guidance methods, voice instructions are the most intuitive, while sonification offers continuous and precise feedback. In this study, we explore a sequential sonification approach (which conveys instructions one dimension at a time, one after the other), inspired by voice guidance, and compare its effectiveness against parallel sonification (which conveys all spatial dimensions simultaneously) and voice guidance provided by a naive human guide. Two experiments were conducted using an invisible target-reaching task in a 3D virtual space. The first experiment was carried out on 20 sighted participants (eyes closed) and one visually impaired participant. This experiment compared sequential and parallel sonification, demonstrating that sequential sonification enables faster target acquisition and is preferred by participants due to its lower cognitive load. The second experiment involved 10 participants and 10 guides. It compared sequential sonification with voice guidance, showing that while voice instructions are natural and do not require a long learning phase, they suffer from latency and discontinuity, making them less efficient than sonification. Furthermore, our analysis of naive guides’ voice commands assessed that they indeed naturally provide instructions one spatial dimension at a time, reinforcing the relevance of sequential sonification. These findings highlight the advantages of our sequential sonification, which combines the precision of non-spatialized sonification, the intuitiveness of voice guidance, and the natural cues of spatialized sonification (stereo). By balancing efficiency and cognitive ease, this approach represents a promising solution for auditory guidance systems designed for visually impaired users.