This paper provides a review of the analysis and modeling of human spatial planning, perception, and learning based on the dynamics of agent-environment interactions. The approach stems from an analysis and modeling framework that was previously conceived using interaction patterns emerging from system-wide interactions as the basic unit of analysis. The paper first discusses the rationals for using patterns in agent-environment interactions as units of organization of behavior, and as functional units of the modeling framework. These concepts are then illustrated through two applications using experimental data from a first-person flight simulator that implements agile obstacle navigation tasks. The first application focuses on the analysis of the formation and evolution of interaction patterns over successive trials, and the use of these patterns as basic elements of the task environment representation, enabling the evaluation of the learning process and assessment of the operator performance. The second application focuses on the analysis of interaction patterns as functional units supporting the modeling of the underlying perceptual guidance and control mechanisms. These examples demonstrate the relevance of dynamics in agent-environment interactions for studying a wide range of functions across the human control hierarchy.
Bibliographical noteFunding Information:
This research work was made possible thanks to the financial support from the National Science Foundation ( CMMI-1002298 and Career Grant CMMI-1254906 ) and the Office of Naval Research (Grant 11361538 ).
© 2017 Elsevier Ltd
- Human motion planning
- Human motor skills
- Human performance modeling
- Perceptual guidance