Understanding multitasking through parallelized strategy exploration and individualized cognitive modeling

Human multitasking often involves complex task interactions and subtle tradeoffs which might be best understood through detailed computational cognitive modeling, yet traditional cognitive modeling approaches may not explore a sufficient range of task strategies to reveal the true complexity of multitasking behavior. This study proposes a systematic approach for exploring a large number of strategies using a computer-cluster-based parallelized modeling system. The paper demonstrates the efficacy of the approach for investigating and revealing the effects of different microstrategies on human performance, both within and across individuals, for a time-pressured multimodal dual task. The modeling results suggest that multitasking performance is not simply a matter of interleaving cognitive and sensorimotor processing but is instead heavily influenced by the selection of subtask microstrategies.