Role of multiple resources in predicting time-sharing efficiency: evaluation of three workload models in a...

Area of Science:

• Human-Computer Interaction
• Cognitive Psychology
• Human Factors Engineering

Background:

• Workload models are crucial for predicting task performance and system design.
• Existing models like Time-Line Analysis and Prediction, VACP, and W/INDEX have distinct assumptions.
• Validating these models is essential for their practical application in complex environments.

Purpose of the Study:

• To assess the validity of three prominent workload models: Time-Line Analysis and Prediction, VACP, and W/INDEX.
• To evaluate the predictive accuracy of these models against empirical data from a flight simulation.
• To identify key features contributing to model fit and compare them to a basic time-line model.

Main Methods:

• A low-fidelity flight simulation was used with 16 subjects performing tracking, monitoring, and decision tasks.
• The decision task featured 16 variations across input modality, processing code, difficulty, and response modality.
• Tracking performance was analyzed using ANOVA and correlational analysis against workload model predictions.

Main Results:

• Dual-task costs were observed solely in the tracking task.
• Tracking performance improved with easier concurrent decision tasks and vocal responses.
• All three workload models effectively predicted between 56% and 84% of the variance in the experimental data.

Conclusions:

• Multiple-resource coding is a critical feature for accurate workload modeling.
• The specific coding features within multiple-resource models require further investigation.
• Task demand level coding appears less critical for predictive accuracy compared to resource coding.