Computational GOMSL Modeling towards Understanding Cognitive Strategy in Dual-Task Performance with Automation

Abstract

The objective of this study was to assess the use of a computational cognitive model for describing human performance with an adaptively automated system, with and without advance cueing of control mode transitions. A dual-task piloting simulation was developed to collect human performance data under auditory cueing or no cueing of automated or manual control. GOMSL models for simulating user behavior were constructed based on a theory of increased memory transactions at mode transitions. The models were applied to the same task simulation and scenarios performed by the humans. Comparison of results on human and model output demonstrated the model to be generally descriptive of performance; however, it was not accurate in predicting timing of memory use in preparing for manual control. Interestingly, the human data didn't reveal differences between cued and no cue trials. A refined GOMSL model was developed by modifying assumptions on the timing and manner of memory use, and considering human parallel processing in dual-task performance. Results revealed the refined model to be more plausible for representing behavior. Computational cognitive modeling appears to be a viable approach to represent operator performance in adaptive systems.