Article #10
Newsletter list 


Previous Article 

Next Article 


Situation Awareness

Alex Kirlik
Atlanta, GA

My student, Richard Strauss, completed a dissertation this year viewing the human factors problem of "situation awareness" (SA) from a Brunswikian perspective. Rich developed a framework in which SA is conceived as a set of relations between a performer's state of knowledge and an environmental situation. He conducted three experiments in the context of a naval submarine detection task in order to evaluate his approach. Stewart and Lusk's expanded lens model (ELM) was used for statistical analysis and modeling.

The first experiment evaluated whether perceptually augmented displays could be used to enhance SA and whether ELM parameters would be sensitive to this manipulation. The results of this experiment were mixed in terms of SA enhancement, but unequivocal when evaluated in terms of the ELM. Display augmentation improved participants' abilities to perceptually measure cue values but caused participants to display a significantly greater regression bias than participants using a baseline (unaugmented) display. The bias of the display augmentation group was toward overweighting situation specific cues at the expense of base rate information. This finding is in accord with the suspicion of some human factors researchers that the use of increasingly rich displays for decision support may cause performers to insufficiently attend to information from other sources.

In the second experiment, a fitted ELM model for each experimental participant was used to bias the presentation of display information in a participant-specific manner. This manipulation significantly increased environmental predictability for 13 out of 16 participants and increased the achievement of 8 of 16, leaving the other 8 unchanged. The biasing adjusted cue values to levels where the ELM model predicted the participant would render a perfect judgment, and had the effect of acting as a filter on environmental noise.

The final experiment focused on individual differences, in particular, the ELM's ability to diagnose the underlying differences between the highest and lowest performing experimental participants. In this task, high and low performers did not differ in terms of regression or base rate biases, or in terms of task knowledge. However, high and low performers differed significantly in terms of both consistency of cue acquisition and consistency of information processing.

Strauss (2000). A methodology for measuring the judgmental components of situation awareness. Unpublished doctoral dissertation. School of Industrial & Systems Engineering, Georgia Institute of Technology.

Contact Alex Kirlik

Previous Article                Next Article

Newsletter list