Model Based Development of a Mental Workload Sensitivity Index for Subject Clustering

Abstract

Considering individual differences in personality, personal characteristics, or abilities is common praxis in several areas of research. The clustering of subjects related to individual variables is particularly important in order to value identified content, recognize relationships, analyse and derive complex concepts as well as improve our understanding of potential issues. In this context, the concept of mental workload comprises a number of individual characteristics but is rarely used for subject clustering. In our article, we introduce an approach for the calculation of a workload-sensitivity index that can be used for such purpose. We present a two-parametric logistic model that predicts workload (WL) sensitivity parameters for sample means across participants. Together with a linear approximation it provides estimates for individuals. Thereby it takes into account cognitive resource limitation and the specific scale limits of WL-metrics as well as domain expert knowledge as prior information. Experimental evidence is provided by means of a human-in-the-loop simulation experiment with 21 air traffic controllers through measuring of the WL-effects under eight different task load levels (scenarios) realized by traffic flow n (aircraft/hour) and a non-nominal event using subjective Instanteneous Self Assessment (ISA) metrics. We analyse the ISA measures and show that the theoretically predicted ISA vs. n characteristic exhibits surprisingly good agreement with the experimental parameter estimates when based on the ISA scenario averages despite large inter-individual variance. To sum up, our model based subject clustering allows for defining subgroups of different WL-sensitivity based on a single dimensionless sensitivity index.