Experienced researcher investigating human perception and user performance using head- and eye-tracking, statistical modeling, and machine learning.
While many fields of research stand to benefit from first-person data that is representative of everyday experience, little of this data is widely available. I work as part of an NSF-funded project to collect 240 hours of egocentric video, head movement, and eye movement data from a diverse group of participants across three locations in the United States. I play a role in collecting data for this project at the University of Nevada, Reno, but also work towards development of data analysis libraries in Python.
As the head moves during everyday life, sense organs like the eyes, vestibular organs, and ears also move. As a result, any stimulation these organs receive is systemically modulated. Head movements are not made at random, however, and are highly regular: subsequent stimulation of the sense organs in the head is also regular. It is likely that the nervous system uses these regularities in some manner to constrain nervous system processing, but without accurate characterization of these regularities it is difficult to model these processes. Using a lightweight tracking camera worn on participants' heads, I characterize natural head movements during five hour-long recording sessions while participants complete unconstrained, everyday behavior.
Virtual reality has exceptional potential as an aid for training people to work in dangerous environments or with novel equipment. We designed and evaluated a head-mounted display capable of delivering visual stimuli to a participant while they are submerged underwater with SCUBA equipment. We test usability by training participants to complete a jetpack locomotion task analogous to one astronauts working for NASA train for at the Neutral Buoyancy Lab in Houston, Texas. Participants are able to learn and complete the task relatively quickly, and do not experience clinically significant amounts of simulator sickness while using the device.