Masters Capstone Project: “Vista” Trace Visualization Tool

Speaker
Marianne Berkovich, Elaine Kwong, and Andrew Yang
Master Students, Human-Computer Interaction Institute, Carnegie Mellon University

When
-

Where
Newell-Simon Hall 1305 (Michael Mauldin Auditorium)

Description

Eleven Masters students in Carnegie Mellon’s Human-Computer Interaction Institute worked with the Apex development team at NASA Ames Research Center’s Cognitive Modeling Group for nine months under the charge to “make a human-performance modeler’s life easier.” Human performance modeling is a method for predicting how people interact with new technology before it is built; a powerful method that has not gotten wide dissemination in real-world design because it is too hard to learn and too time consuming to perform. This collaboration between NASA and CMU’s HCI Masters program aimed to bring human performance modeling to real world design by providing tools that are easier to learn and use than previous modeling environments. To that end, this student project used HCI methods like contextual inquiry and iterative design and user testing to create and improve tools for modelers to learn, understand, produce, and share models in the Apex modeling environment. At the end of the spring semester, the team split into two groups and each group pursued the creation and refinement of a prototype of the selected design ideas. The NASA-West team developed the trace visualization tool design, which we dubbed “Vista.” Trace visualization is needed because every model run produces a trace, that is an unwieldy text file of each event that occurred during the run. The main challenges when working with this trace are dealing with a huge amount of raw data that is produced with every run, anticipating which information will be needed beforehand and filtering for it, locating information about a single task that may be spread over the entire trace. We will be showing how Vista evolved through iterations of user testing to address these challenges and make it easier to answer specific questions about the model as well exploring the results of the model.

Speaker's Bio

The CMU “NASA-West” MHCI project team consisted of five Masters in Human Computer Interaction students with backgrounds in Computer Science, Design, English, and Civil Engineering. The culmination of the Masters project allowed us to spend the past summer at Moffett Field, California working at NASA Ames Research Center to develop a prototype for a trace visualization tool.