An overview of eye-movement analyses measures for validating a cognitive pilot model

Abstract

The aim of the paper is an introduction to eye-data analysis in a high-fidelity cockpit simulator for the validation of the cognitive modeling approach CASCaS (Cognitive Architecture for Safety Critical Task Simulation). Within the European Project HUMAN, founded by the EC (2008-2011) a cognitive human–machine model of a pilot crew is developed. Such interactive models that account for the cognitive aspects of human information processing, claim supporting the development and design of new systems by error detections in an early design phase. This paper discusses the validation aspects of these complex human-machine models, focusing on eye movement analyses. Therefore a high-fidelity human-in-the-loop simulation was accomplished, integrating an Advanced-HMI/FMS with the capability for 4D-trajectory gate-to-gate negotiations via datalink. Requirements, difficulties and methods for the analysis of eye movement data are discussed in respect to validate a cognitive pilot (crew) model. A literature overview of eye-data analyses in the cockpit is given. Based on a case study the eye-data set of one pilot is analyzed, showing the analyses methods to validate aspects of information acquisition of formal cognitive human-machine models. The paper makes clear that for validating of a pilot (crew) model, a comprehensive analysis of simulation events, operators’ tasks and eye data becomes more important. The application of eye-data for validation aspects of cognitive models in complex 3d-environments is given.