Whether or not it is possible to predict military performance using laboratory measures constitutes an important question. There are indications that humans possess a common multitasking ability enabling them to perform complex behaviors irrespective of task requirements. Working memory processing abilities likely illustrate cognitive substrates thereof. Thus, it should be possible to predict military performance by means of laboratory multitasking via working memory processing abilities. To investigate this, we recruited 40 officer cadets and assessed their laboratory multitasking proficiency using the Multi-Attribute Task Battery and their performance in a simulated military operation. We then tested if the laboratory measure predicted their military performance and if this relationship was mediated by working memory processing abilities using Bayesian procedures. We also controlled if demographics, military characteristics, media preferences, or social/personality traits affected any of these measures. In contrast to our expectations, the associations between laboratory and military multitasking and working memory were weak. Furthermore, the participants did not display multitasking decrements but improvements as a function of time on task in the military setting. Moreover, we found a positive association between the time officer cadets had already served in the military and military performance. We discuss the role of learned task representations in this regard and conclude that it might be more reasonable to investigate cognitive functions as co-variates of associations between military characteristics (e.g., military service duration) and military performance in future research than to focus on laboratory measures as predictors of military performance.
Keywords: Bayesian statistics; Military performance; Working memory.
© 2025. The Author(s).