Before the era of fully automated vehicles, human is consistently an indispensable part of the driving system. Various studies have investigated drivers' cooperation with the vehicle under different conditions. In this article, we analyzed how non-driving-related tasks (NDRT) influence takeover time (TOT) by conducting a meta-analysis on 37 related papers. NDRTs were transcoded into combinations of five basic dimensions to unify and demonstrate their effects on drivers. In order to interpret experimental data comprehensively, we implemented three methods. A synthetical analysis was conducted to compare the effect size between each study and subgroup. Studies with eligible control groups have been examined by the two-group analysis, followed by moderator analysis on seven variables. The results from the two-group analysis showed that both visual-mental-motoric and visual-mental tasks have significant negative effects on the takeover time and the previous type had a larger effect than the latter one. Moreover, the subgroup comparison and meta-regression in the meta-analysis part revealed the correlation between moderators and the effect size, in which the Driving Experience and the Automation Level affected the relation between NDRT and TOT. The findings of this paper can contribute to the improvement and new directions for further scientific research and engineering design.

翻译：在完全自动驾驶时代到来之前，人类始终是驾驶系统中不可或缺的一部分。已有诸多研究探讨了不同条件下驾驶员与车辆的协同配合。本文通过对37篇相关文献进行元分析，研究了非驾驶相关任务（NDRT）如何影响接管时间（TOT）。我们将NDRT转码为五个基本维度的组合，以统一并揭示其对驾驶员的影响。为全面解读实验数据，我们采用了三种分析方法：通过综合分析比较各研究及亚组间的效应量；对具备合格对照组的实验进行双组分析，并对七个变量进行调节变量分析；最后通过元分析中的亚组比较与元回归探究调节变量与效应量的关联。双组分析结果表明，视觉-心智-运动型任务与视觉-心智型任务均对接管时间产生显著负面影响，且前者的影响程度大于后者。此外，元分析中的亚组比较与元回归揭示了调节变量与效应量之间的相关性，其中驾驶经验与自动化水平显著影响NDRT与TOT之间的关系。本文的研究发现可为后续科学研究与工程设计的改进及新方向提供参考。