Modeling Human Performance in Complex Systems

Workload.

“…The model's demand levels were established through benchmarks, as revealed by consensus judgments of experts in the workload field. For example, detection has a value of 1 within the visual channel, whereas reading a symbol has a value of 5.1 (Laughery, Plott, Matessa, Archer, & Lebiere, 2012). Thus, in the VACP model, a red line value is established within each channel, and whenever the combined demand of tasks within a channel exceeds that value, the red line is crossed, and performance is predicted to decline.…”

All models are wrong, some are useful, but are they reproducible? Commentary on Lee et al. (2019)

“…The best example COMMENTARY ON LEE ET AL. (2019) 6 of such a framework may be ACT-R (Anderson & Lebiere, 1998), which has proven enormously useful and influential in the human factors literature (e.g., Laughery, Plott, Matessa, Archer, & Lebiere, 2012). We appreciate that in some cases, cognitive models cannot be easily generalized beyond the task in which they were developed.…”

Mental workload prediction based on attentional resource allocation and information processing

“…Cognitive structure analysis utilizes foundational cognitive theory to establish a frame structure representing a potential object, principle and the operator's performing mechanism; then the human behavior in the complex system can be expressed as relevant factors in the structure [3]. For instance, Xie and Salvendy presented and verified a MW conceptual frame under single or multiple task environments, including instantaneous MW, average MW, accumulated MW and total MW, as well as a prediction model [8].…”

“…Task network analysis does not generally need special mathematical modeling technology, but instead requires detailed task sequence and demanded time information [3]. For example, the Timeline Analysis and Prediction (TLAP) method worked out the timeline through elaborate task analysis and calculated MW in a certain time unit, which was the TR/TA (Time Required/Time Available) in the time unit [4].…”

“…Prior analysis of the operator's MW in the humanmachine system can be adopted in both MW prediction and evaluation. The widely used MW prediction modeling methods include task network analysis and cognitive structure analysis [3].…”