Lauri Oksama¹ (), Jaakko Kulomäki¹, Jukka Hyönä², Esa Rantanen³; ¹Finnish Defence Research Agency, Finland, ²University of Turku, Finland, ³Rochester Institute of Technology

Multitasking is an essential skill in many job environments where multiple subtasks must be performed simultaneously. However, when several subtasks occupy the same sensory channel simultaneously, such as displays placed at different points in the visual field, operators of the systems must temporarily choose which subtask to monitor and pay closer attention to. In real-life multitasking, operators must prioritize subtasks, allocating their attention to the most important tasks at the right time, even at the expense of less important tasks. In demanding safety-critical environments, such as aviation, it is necessary to be able to measure the ability of those aspiring to the field to cope with challenging multitasking environments. Recently, we devised a demanding multitask environment that requires subtask prioritization, intelligent and flexible attention control, and time-sharing for successful performance (see Kulomäki et al., 2022). The task consists of four subtasks (like dynamic meters) requiring continuous monitoring and precise motor responses at exactly the right time. In our multitasking environment, subtasks had distinct event rates defining their priorities; the subtask with the highest event rate requires more attention and responses than the subtask with a lower event rate. To motivate participants to perform at their maximum level, a game-like composite scoring system was used, based on penalty and reward scores derived from the accuracy of responses to the subtasks. Here, we present evidence of the content validity of the composite score of the new multitask by examining its correlation with other tests measuring cognitive constructs, such as fluid intelligence, and with other multitasking environments like SYNWORK (Elsmore, 1994). We report correlational results from Finnish Air Force applicants (two data sets, N=456 and N=196). Our findings provide evidence that our new multitask environment shows better discriminant validity (lower correlations) with fluid intelligence compared to previous multitasks like SYNWORK (Elsmore, 1994).