MU researcher testing limits of human perception capabilities
The ability to detect potentially harmful changes is of paramount importance in nuclear energy, air traffic control and security professions, among others. Being able to detect tiny abnormal events before they become huge problems is critical. A researcher in the MU College of Engineering is developing technology to measure how humans respond to those events in order to make these processes more efficient.
Jung Hyup Kim, assistant professor of industrial and manufacturing systems engineering at MU, and his student research team recently began using motion-capture and eye-tracking technology to investigate the relationship between human perception stress — measured by pupil diameter — and performance using human-in-the-loop simulation.
“Recently, advanced technology provides an almost unlimited ability to instantly process data. However, human ability to simultaneously understand this data is limited. Humans have limitations. If they’re overwhelmed, then detectability is lower, and they might miss critical signals from the system. ” Kim said. “The question is, how does human multitasking influence the detectability, and how do we overcome the information gap between data produced and information needed? We want to advance our understanding of human perceptual behavior and detection behavior through this experiment.”
To measure the behavior, volunteers wear an eye-tracking device and head motion tracking gear as they work on a simulation on two side-by-side monitors. Five feet behind each participant is a motion-capture device so the team can track not only where the eye’s focus is, but also the movement of the head.
Participants are tasked with watching the performance of gauges on two monitors, similar to console operators in oil and gas refineries, responding to process changes as they occur. Volunteers will take part in the experiment over the course of five days. The goal is to gauge not only detection and reaction speed, but also how those change over time as the eyes become fatigued, which is measured by pupil size. The simulation is adjustable to help eliminate another variable — skill level.
“Right now, we are finalizing our experimental setup to optimize the test environment.” IMSE graduate student and research team member Xiao Yang said.
Simultaneously, the team is calibrating the maximum amount of stimulus one person can handle during the experiment..
“There’s a ton of stuff going on and a ton of different things you have to hit,” IMSE junior and research team member Nolan Rackers said after taking part in a trial run with the simulator.
The goal of the initial phase of the research is to potentially uncover statistically significant relationship between pupil diameter and detectability. The hope is to have the pilot test finished and begin the research with student volunteers in April.
“If we can find a more clear model of the perception of behavior and detectability, we can expand our research to the real operators,” Kim said. “The thing is, many researchers already found there’s something good in eye-tracking data. So, we want to apply this science knowledge to the actual industry field.”