Tech Report CS-08-10
Effects of Illumination, Texture, and Motion on Task Performance in Streamtube Visualization of Deffusion Tensor MRI
Devon Penney, Jian Chen, Member, IEEE, and David H. Laidlaw, Senior Member, IEEE
We present results from a user study comparing user task performance on streamtube visualizations generated from diffusion tensor magnetic resonance imaging (DTI) datasets. The independent variables include illumination model (global illumination or OpenGL local illumination), texture (with and without), motion (with and without), and tasks. The three spatial analysis tasks are: 1) a depth judgment task: determining which of two marked tubes is closer to the user's viewpoint, 2) a visual tracing task: marking the endpoint of a tube, 3) a contact judgment task: analyzing tube-sphere penetration. Our results indicate that global illumination did not improve task completion time for the depth judgment task or the tracing task, contrary to the results of a previous study. Global illumination improved accuracy of participants answers over local OpenGL-style rendering for the contact judgment task only. Motion contributes to spatial understanding for the visual tracing and contact judgment tasks but at the cost of longer task completion time. A high-frequency texture pattern with more than one directional component led to longer task completion time and higher error rates. These results suggest that black-box visualization techniques are unlikely to be better than task-specific techniques tailored to the data being presented. Lighting design noticeably alters performance, indicating that designers need to pay specific attention to such issues.
(complete text in pdf)