AI & ML
An Affordable Vibrotactile Surface for Tactile Figure Recognition: Exploration Strategies of Blind and Sighted Users
arXiv cs.HC0 views
arXiv:2611.22614v1 Announce Type: new Abstract: Refreshable tactile displays can make graphics accessible to blind and low-vision people, but high-resolution actuator arrays remain expensive. We study whether a low-cost vibrotactile surface can convey simple figures, and how they are read. Nine participants (four blind; five sighted, tested blindfolded then open-eyed) identified seven figures by touch, with video and timing analysis of their exploration. Blind and blindfolded-sighted readers achieved comparable accuracy, but blind participants were about three times faster, using a continuous, rhythmic "gliding" strategy-reading through motion, whereas sighted participants searched in halting, pause-heavy passes; among the sighted, more pausing accompanied errors. Recognition failures were concentrated in specific figures (a diagonal was misread even with vision), locating the bottleneck in representation rather than resolution. We contribute an empirical characterization of tactile reading strategies and design implications for low-cost refreshable tactile displays and affordable tactile representation.
Source:https://arxiv.org/abs/2611.22614PreviousControlling Proactive Multimodal AI: Separating Perception from ParticipationNextRAW: Refusal Aversion in the Wild, A Causal Measurement Method for Deployed LLMs
Related posts
How OpenAI scales PostgreSQL to 800 million users
0dev.to Database
Spec-Driven PBT: Deriving Semantic Invariants from Code Specifications for Agentic Property-based Testing
0arXiv cs.SE
A Grey Literature Review of Graduate Students' Research Practices and Skills in the LLM Era
0arXiv cs.SE
ReBoT-MiCS: Reliability-Aware Bounding Time for Enhancing Timing Behavior of Embedded Mixed-Criticality Systems
0arXiv cs.SE
VoxelSBFL: Voxel-Level Spectrum-Based Failure Prediction for Vision-Language-Action Models
0arXiv cs.SE
Discontinuity-Aware Solvers for Vapor-Compression Cycles
0arXiv cs.NA