This was really interesting. It explored the effects of time-on-task fatigue on hazard recognition and visual attention/gaze characteristics during a simulated excavator activity with experienced operators.
Time-on-task fatigue describes the effects (typically negative) from prolonged and often monotonous tasks, like cognitive performance decrements and/or drowsiness.
12 participants were included in the study in a lab environment. An excavator simulator on a construction site which introduces hazards like people walking behind the excavator (which are visible in the rearview and side mirrors).
Eye movement (which recorded gaze and pupil characteristics) was tracked using eye tracking glasses. Fatigue was measured using a subjective scale. When hazards appeared in the mirrors, the subjects were required to hit a button as quickly as possible, and the latency of hitting the button and misses or false alarms were also tracked.
Results:
As expected, as operators’ experienced higher subjective fatigue as time-on-task increased and this was matched by increases in hazard miss rate (not observing hazards), false alarm rate (incorrectly observing a hazard), and slowed reaction time. Participants’ rate of missing peripheral hazards increases by over 40% after a continuous 60 min block of the task.
The authors state that the findings “showed that mental fatigue can heavily impair construction equipment operators’ ability to detect hazards even in early phases of excavation tasks (nearly 30% of hazard miss rate for the 36 min task” (p10).
It’s noted that while the time to recognise a hazard was monitored, they didn’t track the time to respond to the hazard. Thus, “hazards can be worse in actual construction sites because the reaction time may be underestimated compared with real operation tasks” (p10).
It’s said that “the results showed that participants need a considerable amount of time to identify hazards. They increasingly missed hazards and made considerable false identification of non-hazards as mental fatigue increased” (p9).
Interestingly, the eye tracking showed one mechanism on how operators missed or misinterpreted site hazards. They found that as time-on-task increased, operators’ gaze to both sides of the excavator (looking at the side mirrors) decreased while gaze was shifted more towards the front of sight. Thus, as this type of fatigue increased under this simulation, people spent more time looking forward rather than around them.
This effect makes sense, and the authors refer to evidence around how mental fatigue leads to “an aversion to a further investing effort in maintaining task performance process, the increase in operators’ mental fatigue makes them reluctant to pay more attention to the edges far from the visual center” (p11). This results in an impaired distribution of visual attention to the surroundings.
Also, It was really cool seeing the visual heat map data that came from the eye tracker and how gaze changed over the task.
Finally, the usual limitations apply – small sample size, simulation setting (which is known to influence results compared to real world settings) and other stuff.
Authors: Jue Li, Heng Li, Hongwei Wang, Waleed Umer, Hong Fu, Xuejiao Xing, 2019, Automation in Construction
Study link: https://doi.org/10.1016/j.autcon.2019.102835
Link to the LinkedIn article: https://www.linkedin.com/pulse/evaluating-impact-mental-fatigue-construction-ability-ben-hutchinson
SafetyInsights.org 