Delegation to automation: performance and implications in non-optimal situations

  • Authors:
  • Christopher A. Miller;Tyler H. Shaw;Joshua D. Hamell;Adam Emfield;David J. Musliner;Ewart De Visser;Raja Parasurman

  • Affiliations:
  • Smart Information Flow Technologies, Minneapolis, MN;Human Factors and Applied Cognition Program, George Mason University, Fairfax, VA;Smart Information Flow Technologies, Minneapolis, MN;Human Factors and Applied Cognition Program, George Mason University, Fairfax, VA;Smart Information Flow Technologies, Minneapolis, MN;Human Factors and Applied Cognition Program, George Mason University, Fairfax, VA;Human Factors and Applied Cognition Program, George Mason University, Fairfax, VA

  • Venue:
  • EPCE'11 Proceedings of the 9th international conference on Engineering psychology and cognitive ergonomics
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

We have previously advocated adaptable interaction with automation through approaches derived from human-human delegation and using the metaphor of a sports team's "playbook". In work sponsored by the U.S. Army's Aeroflightdynamics Directorate (AFDD), we have been studying the effects of play-based delegation on human-machine system performance. Of particular interest is performance with plays in "non-optimal play environments" (NOPE) where no, or only poorly fitting, plays exist to achieve needed behaviors. Plays have been shown to offer benefits in situations for which they are customized, but more interesting is whether complacency, expectation, loss of training, and automation bias might affect performance when plays do not perfectly fit. We provide a taxonomy of NOPE conditions and report on the exploration of some of these conditions in a series of three experiments performed to date.