SANLab-CM: a tool for incorporating stochastic operations into activity network modeling

Area of Science:

• Cognitive Science
• Human-Computer Interaction
• Psychology

Background:

• Activity network modeling is crucial for understanding task performance.
• Previous models often used static paths, potentially oversimplifying complex human-computer interactions.
• Variability in cognitive, perceptual, and motor processes significantly impacts task execution.

Purpose of the Study:

• Introduce the Stochastic Activity Network Laboratory for Cognitive Modeling (SANLab-CM).
• Demonstrate SANLab-CM's capability to model complex task dynamics.
• Distinguish between emergent variability and distinct task strategies.

Main Methods:

• Developed SANLab-CM, integrating stochastic operations into activity network modeling.
• Expanded a static model of telephone operator-customer interaction into a stochastic model.
• Generated 55 unique paths with varying frequencies and qualitative properties.

Main Results:

• The stochastic model generated diverse task paths, highlighting variability.
• SANLab-CM differentiated between frequent critical paths and underlying process variability.
• Identified critical paths as emergent properties of basic process variations.

Conclusions:

• SANLab-CM provides a more nuanced understanding of task performance by incorporating stochasticity.
• The tool helps avoid misinterpreting emergent path variations as deliberate strategic choices.
• This approach enhances cognitive modeling by accounting for inherent process variability.