The Effect of Information Level on Human-Agent Interaction for Route Planning

Report No. ARL-TR-7563
Authors: Julia L Wrigh; Michael W Boyce; Jessie YC Chen; Michael J Barnes
Date/Pages: December 2015; 50 pages
Abstract: This study investigated how differing levels of information affected decisions human operators made in a route-selection task. Experiment 1 examined how information about resource usage/requirements affected route-selection decisions for a remotely based supervisor guiding a dismounted Soldier unit through an urban environment. Experiment 2 included all the information from Experiment 1 but added a robotic asset and its resource usage/requirements information. Results suggest that a low level of information corresponds to a significantly lower decision time (DT) than its counterparts, and DT at all information levels increased with the addition of the robotic asset. Participants exhibited sensemaking by reducing DTs as the experiment progressed. In addition, as the amount of information increased, preference for specific information sources began to vary. In the condition with the greatest amount of information available, participants displayed no clear consensus as to preferred information source, with many indicating they preferred sources that were unsuitable for successful mission completion. Future research could investigate further into the complexity of appropriate display for user interfaces involving robotic assets.
Distribution: Approved for public release
  Download Report ( 1.023 MBytes )
If you are visually impaired or need a physical copy of this report, please visit and contact DTIC.
 

Last Update / Reviewed: December 1, 2015