Abstract
This study examined how neural engagement differs between training and application under varied cognitive workload in a virtual-reality assembly task. Twenty-three participants completed four within-subject conditions (low vs. high intrinsic load and low vs. high extraneous load). We recorded three electroencephalography-derived engagement indicators: Cognitive Effort (parietal α/frontal θ), Sustained Attention (β/(α + θ)), and Integration & Execution (γ power). Changes in these indicators during learning were expected to reveal changes in engagement linked to unique cognitive resources recruited by workloads and phases of learning tasks. Analyses revealed a robust task phase effect: application activation exceeded training for Sustained Attention and Integration & Execution. Significant task phase workload condition interactions emerged for these indicators as well. Findings indicated that unguided application amplifies working-memory and integrative engagement and that this amplification depends on workload type. Our results inform the design of adaptive training systems monitoring phase-dependent neural engagement to optimize learning transfer.
Get full access to this article
View all access options for this article.