影响驾驶安全的驾驶员注意模式研究述评

驾驶员的视觉注意模式对于汽车安全驾驶具有非常重要的意义。本文主要从驾驶员的路面视觉扫描模式和整体视觉场景的注意模式两个方面对国内外有关研究进行了概览。主要包括: (1)视觉扫描模式对驾驶安全的影响研究; (2)驾驶员视觉注意分配模型研究; (3)以及针对驾驶员视觉扫描模式的培训研究。并提出进一步研究有待于深入探讨影响驾驶员视觉注意模式的影响因素以及开展适合我国汽车驾驶特点的本土化研究。     

音乐对驾驶警觉的影响及其理论模型

驾车时听音乐非常普遍。唤醒理论认为, 音乐可以使驾驶员处于最佳唤醒水平, 有利于驾驶警觉的维持。但是, 根据中枢容量有限理论, 音乐也会使驾驶员分心, 危害道路安全。动态模型在已有研究的基础上提出, 音乐对驾驶警觉起到唤醒抑或分心影响, 取决于音乐复杂性、驾驶情境要求及驾驶员个体差异等诸多因素的动态调节。最新的补偿控制理论则强调, 驾驶员自我调控是恢复动态平衡的关键。而未来的研究应关注驾驶员心理负荷的测量、智能车载音乐系统的开发以及驾驶员培训等领域。     

自我评估的能力与驾驶经验对危险知觉的影响

结合反应时和眼动研究方法,采用驾驶经验2（新手,有经验）×自我评估的能力2（高,低）×危险类型2（明显,隐藏）的混合实验设计,探索驾驶员驾驶能力的自我评估和驾驶经验对危险知觉及视觉注意的影响。采用驾驶能力量表和基于动态交通视频的危险知觉任务对86名驾驶员测试,并使用Tobbi T120记录眼动数据。结果发现,新手和有经验驾驶员自我评估的能力与危险知觉反应时间之间相关不显著,这说明两组驾驶员对自己驾驶能力的认识和评估不准确。有经验驾驶员对两类危险的反应比新手快。新手自我评估的能力过高,他们对两类危险的反应比同龄驾驶员和有经验驾驶员慢。然而,有经验驾驶员自我评估的能力与危险知觉反应时间之间不存在显著差异。此外,与新手相比,有经验驾驶员对两类危险的首次注视较快,总注视时间更长。与同龄驾驶员相比,新手自我评估的能力过高,他们对危险的首次注视较慢,对危险的总注视时间更少。这些研究结果表明,新手自我评估的能力过高,可能会提高他们的风险接受阈限,由此降低了他们对危险的反应速度。未来驾驶训练应当采取措施减少新手对驾驶能力的自我评估偏见,改善他们的视觉搜索模式以降低事故风险。     

不同驾驶条件对于驾驶员情景意识的影响

本文详述了两种驾驶条件对于驾驶员情景意识的影响。手动驾驶是目前最为常用的驾驶方式,驾驶员需要手脚配合来完成驾驶任务,并且要时刻关注于道路情况。高度自动化驾驶系统(HAD)是基可以同时控制机动车的纵向运动和横向运动。根据之前的研究,手动驾驶带有HAD辅助系统的驾驶的情况下,可以将更多的注意资源分配在与驾驶无关的其他任务上。驾驶员驾驶带有HAD辅助系统的车辆与手动驾驶相比,驾驶员识别环境中客体的能力有所提升。但是在驾驶员驾驶过程当中同时执行与驾驶无关的任务,驾驶带有HAD系统与手动驾驶相比,情景意识较差。     

影响驾驶安全的驾驶员危险知觉研究述评

危险知觉是指驾驶员对交通情境中的潜在危险进行识别并做出反应的一种能力。目前,驾驶员危险知觉测量主要使用反应时、眼动或主观评定三种测量方法。在测量中,这三种方法各有优缺点,并出现相互结合的研究趋势。为了提高我国交通安全性,结合国外在提高驾驶员危险预期能力和改善视觉搜索模式两个方面的应用研究,未来国内研究可以在探索危险知觉的认知加工特点、影响因素及加快测量技术的商品化应用等方面加以深入。     

基于认知机制的有条件自动驾驶接管中的人因问题

自动驾驶能在很大程度上缓解现代交通问题并提升驾驶舒适度。有条件自动驾驶下, 驾驶员可执行非驾驶相关任务但需要在系统无法处理的状况下接管车辆。在这一关键过程中, 驾驶员需要进行注意转换并获得情境意识以成功接管。已有研究表明, 接管请求、非驾驶相关任务、驾驶情景及驾驶员因素是影响接管过程重要因素。未来可从认知机制角度研究各因素对接管过程产生的影响, 以及探究接管过程中各因素之间可能存在的交互作用。     

驾驶经验和危险类型对年轻驾驶员危险知觉的影响

结合反应时和眼动测量方法,采用驾驶经验分组2（年轻新手驾驶员,年轻有经验驾驶员）× 危险类型2（明显危险,隐藏危险）的混合实验设计,探索年轻驾驶员危险知觉的经验优势是否随着危险类型的不同而变化。使用一个基于真实交通情境视频的驾驶员危险知觉任务,对27名年轻新手驾驶员和22名年轻有经验驾驶员进行测试,并使用Tobbi T120记录眼动数据。结果发现,总体上,年轻新手驾驶员对潜在道路危险的反应次数比年轻有经验驾驶员更少,反应时间也更慢。两组驾驶员反应时间的差异,随着危险类型的不同而来源于不同的加工阶段。具体来说：一方面,年轻新手驾驶员对明显危险的反应时间慢,是因为他们对该类危险的评估时间长。另一方面,年轻新手驾驶员对隐藏危险的反应时间慢,是因为他们对该类危险的识别时间长。两组驾驶员对危险的平均注视时间受测试中危险类型的影响但不受驾驶经验的影响。这些研究结果表明,在对年轻新手驾驶员进行危险知觉训练或测试时,应重视危险类型对他们危险知觉的影响。     

驾驶经验和危险类型对年轻驾驶员危险知觉的影响（英文）

结合反应时和眼动测量方法,采用驾驶经验分组2(年轻新手驾驶员,年轻有经验驾驶员)×危险类型2(明显危险,隐藏危险)的混合实验设计,探索年轻驾驶员危险知觉的经验优势是否随着危险类型的不同而变化。使用一个基于真实交通情境视频的驾驶员危险知觉任务,对27名年轻新手驾驶员和22名年轻有经验驾驶员进行测试,并使用Tobbi T120记录眼动数据。结果发现,总体上,年轻新手驾驶员对潜在道路危险的反应次数比年轻有经验驾驶员更少,反应时间也更慢。两组驾驶员反应时间的差异,随着危险类型的不同而来源于不同的加工阶段。具体来说:一方面,年轻新手驾驶员对明显危险的反应时间慢,是因为他们对该类危险的评估时间长。另一方面,年轻新手驾驶员对隐藏危险的反应时间慢,是因为他们对该类危险的识别时间长。两组驾驶员对危险的平均注视时间受测试中危险类型的影响但不受驾驶经验的影响。这些研究结果表明,在对年轻新手驾驶员进行危险知觉训练或测试时,应重视危险类型对他们危险知觉的影响。.     

伪忽视行为表现与神经基础及交通驾驶应用

伪忽视是健康个体存在轻微偏左的不对称空间注意。影响这个现象的因素主要包括刺激材料、年龄、认知负荷和知觉负荷等。其中, 眼动搜索习惯曾经被认为是伪忽视现象的主要因素, 甚至提出伪忽视是眼动偏好现象。但是, 最近文献研究普遍认为, 伪忽视不是眼动行为, 而是注意特质。近10年的大脑神经层面的证据发现, 右脑腹侧注意网络对左侧空间注意偏好有较大的贡献, 而且在神经传导束的证据中也发现右脑大纵束二型对左侧空间注意的激活程度高于其他两种类型的传导束。另外, 注意偏好的时间分辨率研究发现, 注意的左侧空间偏向发生在注意发生后的100~200 ms之内, 眼动搜索时间则是在1500 ms之内出现最大左偏。该现象的应用研究主要体现在交通驾驶员的注意能力和驾驶视觉搜索模式的关系研究。研究意义在于为安全交通提供两条原则, 首先, 道路交通驾驶环境设置要满足驾驶员的注意特质的需求; 其次, 通过合理干预, 可以抑制视觉搜索习惯, 培养安全的驾驶搜索模式。     

次任务测定技术在心理负荷评定中的作用

次任务测定技术是心理负荷评定的重要方法之一,它是以“有限无差异加工资源”假设为基础提出的,这一技术存在“侵入性”、“难度不敏感性”及“结构特异性”等问题,在实际情境中进行测评有时有一定的危险,但是,它具有其它测定技术不具备的“诊断性”特点,作者认为,只要次任务选择适当,次任务测定技术在评定心理负荷中的效用仍是十分明显的.     

Situation Awareness and Workload in Aviation

A pilot faces special challenges imposed by the need to control a multivariate lagged system in a heterogeneous multitask environment. The time lags between critical variables require prediction in an uncertain world. The interrelated concepts of situation awareness and workload are central to aviation psychology. Three components of situation awareness are spatial awareness, system awareness, and task awareness. Each of these components has real–world implications, spatial awareness for instrument displays, system awareness for keeping the operator informed about actions that have been taken by automated systems, and task awareness for attention and task management. Task management is directly related to mental workload, as the competing demands of tasks for attention exceed the operator's limited resources.     

Using agent transparency to support situation awareness of the Autonomous Squad Member

Agent transparency has been proposed as a solution to the problem of facilitating operators’ situation awareness in human-robot teams. Sixty participants performed a dual monitoring task, monitoring both an intelligent, autonomous robot teammate and performing threat detection in a virtual environment. The robot displayed four different interfaces, corresponding to information from the Situation awareness-based Agent Transparency (SAT) model. Participants’ situation awareness of the robot, confidence in their situation awareness, trust in the robot, workload, cognitive processing, and perceived usability of the robot displays were assessed. Results indicate that participants using interfaces corresponding to higher SAT level had greater situation awareness, cognitive processing, and trust in the robot than when they viewed lower level SAT interfaces. No differences in workload or perceived usability of the display were detected. Based on these findings, we observed that transparency has a significant effect on situation awareness, trust, and cognitive processing.     

Urgent and non-urgent takeovers during conditional automated driving on public roads: The impact of different training programmes

Supplying training to drivers that teaches them about automated driving and requests to intervene may help them to build and maintain a mental representation of how automation works and thereby improve takeover performance. We aimed to investigate the effect of different types of training programmes about the functioning of automated driving on drivers’ takeover performance during real driving. Fifty-two participants were split into three groups for training sessions: paper (short notice), video (3-minute tutorial) and practice (short drive). After the training, participants experienced automated driving and both urgent and non-urgent requests to intervene in a Wizard-of-Oz vehicle on public roads. Participants’ takeover time, visual behaviour, mental workload, and flow levels during the requests to intervene were assessed. Our results indicated that in urgent circumstances, participants’ takeover response times were faster in the practice training condition compared to the other training conditions. Nevertheless, the practice training session did not present any other positive effect on drivers’ visual behaviour. This could indicate that prior training, particularly when reinforcing drivers' motor skills, improved their takeover response time at the latest motor stages rather than in the early sensory states. In addition, the analysis of in-vehicle videos revealed that participants’ attention was captured in the first place by the in-vehicle human-machine interface during the urgent request to intervene. This highlights the importance for designers to display information on the HMI in an appropriate way to optimise drivers’ situation awareness in critical situations.     

工作记忆在监控作业情境意识保持中的作用

本研究通过模拟复杂人机系统的监控作业,考察了长时工作记忆与短时工作记忆在情境意识保持中的作用。研究采用模拟的飞行相撞判断任务,通过考察中断任务对情境意识的影响,分析情境意识在记忆中的存储地点,以及保持时间是否会突破1分钟的限制。本研究选择了心理旋转和算式判断两种中断任务。实验结果表明,在问题回答之前的中断阶段,无论施加心理旋转任务,还是施加算式判断任务,无论中断任务的加工负荷与记忆负荷有多高,熟练被试的情境意识始终保持于较高的水平,并没有因受到中断任务的干扰而出现下降。新手被试的情境意识受到心理旋转任务和算式判断任务的显著影响,而且两种任务的执行速度越快、记忆负荷越大,情绪意识水平就越低。本研究结果说明熟练被试可以利用长时工作记忆存储情境意识,新手被试主要利用短时工作记忆存储情境意识。     

性别和情境危险程度对行人危险知觉的影响：来自ERP的证据

采用事件相关电位技术,探讨性别和情境危险程度对行人危险知觉的影响。实验采用三类真实交通情境图片作为材料,包括无危险情境图片、低度危险情境图片和高度危险情境图片。结果发现：低度危险情境条件下,女性的N1潜伏期更短;在N1波幅上,高度危险情境大于无危险情境或低度危险情境;在P3波幅上,男性大于女性,低度危险情境大于无危险情境或高度危险情境。说明早期视觉加工阶段,女性对低度危险情境更敏感;男性比女性给予交通情境更多的注意;情境危险程度能够调节危险知觉过程中的注意分配。     

Attentional processes in time perception: effects of mental workload and event structure

The role of attention in timing was evaluated in 2 experiments. In Experiment 1, participants reproduced the durations of melodies with either a coherent or an incoherent structure. Participants were tested under control (timing only) and detection (timing plus target detection) workload conditions. Reproductions were shorter and more inaccurate under detection conditions, and incoherent event structure extended the effect to a wider range of durations. In Experiment 2, participants reproduced the durations of auditory prose passages that represented 3 levels of mental workload and 3 levels of event structure. Both increases in workload and the degradation of structure led to inaccurate reproductions. The results point to the central role of attention in temporal experience.     

Risk behaviour and mental workload: Multimodal assessment techniques applied to motorbike riding simulation

We present data from an ongoing research project on the cognitive, emotional and neuropsychological basis of risk behaviour. The main aim of the project is to build a model of risk behaviour so that if we know certain cognitive, behavioural and emotional variables, we will be able to predict decisions made in the face of uncertainty and risk, with the final goal of designing programs for evaluating, preventing and controlling risk behaviour. The objective of the present study was to look for individual differences in hazard perception during a static riding simulation and their relationship with mental workload. We used a multidimensional methodology, including behavioural, subjective and physiological data. The behavioural measures were obtained in a static riding simulation during eight hazard situations. We evaluated whether eye activity measures correlated with cognitive workload and different types of risky behaviours. Eye movement parameters were measured using a video-based eye tracking system. We found that risk-prone individuals showed specific patterns of risky behaviours and that peak of saccadic velocity and subjective mental workload indexes were both reliable indicators of risk proneness. Mental workload was higher for participants showing attitudes to risk behaviours probably because of a lack of conscious awareness of specific cues indicating dangerous scenarios.     

Mental workload and visual impairment: differences between pupil, blink, and subjective rating

This research has two aims: (a) To study the concurrent validity of three measures of mental workload, NASA TLX rating scale, pupil dilation and blink rate, testing the hypothesis that they will provide convergent results using a single-task, and dissociative results for dual-task; and (b) To analyse their capability to predict visual search impairment. These three measures were analyzed in the same cognitive tasks in single-task and dual-task (cognitive task and visual search) conditions in a within-subjects experiment with twenty-nine participants. Mental workload measures showed concurrent validity under single-task condition, but a complex pattern of results arose in the dual-task condition: it is suggested that NASA TLX would be a subjective addition of the rating of each task; pupil dilation would measure the average arousal underlying the cognitive tasks; and the blink rate would produce opposite effects: whereas mental workload of cognitive tasks would increase blink rate, visual demand would inhibit it. All three measures were good predictors of visual impairment. The soundness of these measures is discussed with regard to the applied field of driving and other activities.