With the increasing development and implementation of Automated Driving Systems (ADS), understanding driver reactions during the transition from automated to manual driving has become a critical research area. The transition from automated to manual driving is a challenging task that requires the driver to take over control of the ADS-equipped vehicle in a safe and timely manner. Previous studies have identified various variables associated with driver reactions, but a comprehensive classification is lacking. This paper addresses this gap by systematically identifying and classifying the independent and dependent variables that affect drivers’ reactions in ADS-equipped vehicles during the transition from automated to manual driving. This is achieved through an extensive review of major findings in the designs of driving simulator experiments in this field. Additionally, an analysis of five hundred on-road collision reports from 2014 to 2023 involving ADS-equipped test vehicles is conducted. Finally, a comprehensive overview of the identified dependent and independent variables from both studies, along with their synergies and shortcomings, is presented. The variables are categorised and mapped, highlighting key research gaps. The main research gaps were identified by comparing the variables extracted from reviewed papers with the statistical analysis in the DMV reports. Some gaps include the need for incorporating real-world scenarios into experiments, driver-initiated take-over requests and applying physiological measures to assess driver-centric factors. This detailed identification and classification of variables assists in designing a range of future experimental scenarios to assess drivers’ reactions to the transition of control in ADS-equipped vehicles.

Artificial General intelligence (AGI) offers enormous benefits for humanity, yet it also poses great risk. The aim of this systematic review was to summarise the peer reviewed literature on the risks associated with AGI. The review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Sixteen articles were deemed eligible for inclusion. Article types included in the review were classified as philosophical discussions, applications of modelling techniques, and assessment of current frameworks and processes in relation to AGI. The review identified a range of risks associated with AGI, including AGI removing itself from the control of human owners/managers, being given or developing unsafe goals, development of unsafe AGI, AGIs with poor ethics, morals and values; inadequate management of AGI, and existential risks. Several limitations of the AGI literature base were also identified, including a limited number of peer reviewed articles and modelling techniques focused on AGI risk, a lack of specific risk research in which domains that AGI may be implemented, a lack of specific definitions of the AGI functionality, and a lack of standardised AGI terminology. Recommendations to address the identified issues with AGI risk research are required to guide AGI design, implementation, and management.

Engine failure events during critical phases of flight are a rare yet very real possibility on commercial aircraft. The response of the pilots to these events is vital to minimizing possible adversity. New technologies are in development that provide enhanced information on the status of the aircraft engine after sustaining engine damage in order to guide pilot’s decision making. It is important that this enhanced information is aligned with the requirements of the user. This paper describes how user input can effectively be incorporated early on in the design process of new technologies to understand how current systems may be improved. The Perceptual Cycle Model (PCM) is applied to in-depth interview reports from commercial airline pilots on their responses to different engine failure events caused by bird-strikes to the engine. Application of the PCM demonstrates where current processes may be better supported to enhance pilot decision making from such events. From this, areas where improvements could be made using future avionic systems are presented, with user-led design recommendations.

As flight deck touchscreen implementation continues to rise, more research is required to understand how task performance is impacted by turbulence. Previous work found that for basic dragging gestures on a touchscreen, performance and usability declined, and workload increased, with rising levels of turbulence. The current study extended on this work by examining common flight deck tasks using various gestures and input methods. Twenty-six participants completed two tasks (panning and number entry), in different formats, at four levels of turbulent vibration (control, light chop, light turbulence, and moderate turbulence). Performance was assessed objectively (time to completion and additional interactions with the screen), and subjectively (workload, usability, and comfort). Across the tasks, increased turbulence negatively impacted performance, overall workload, and usability. Design recommendations are made for how to best implement pan and number entry tasks on a touchscreen in the flight deck.

Eco-driving assistance systems that encourage drivers to engage in fuel-saving behavior are effective at improving energy-efficiency, with recent research directed towards incorporating predictive models of energy losses in these systems to optimize recommendations. In this article, we evaluate a predictive eco-driving assistance system on three powertrains: a combustion engine-driven vehicle, a parallel hybrid electric vehicle, and a battery electric vehicle. In each case, energy consumption is found by applying a quasi-static model to driving simulator data for a simulated route including urban, rural, and highway sections. We find that both assisted and unassisted eco-driving has a beneficial effect in all cases, with the assistance system leading to reductions in energy usage of 6.1%, 15.9%, and 16.6% for the combustion engine, hybrid electric, and battery electric powertrains, respectively.

This paper assessed four types of human-machine interfaces (HMIs), classified according to the stages of automation proposed by Parasuraman et al. [“A model for types and levels of human interaction with automation,” IEEE Trans. Syst. Man, Cybern. A, Syst. Humans, vol. 30, no. 3, pp. 286-297, May 2000]. We hypothesized that drivers would implement decisions (lane changing or braking) faster and more correctly when receiving support at a higher automation stage during transitions from conditionally automated driving to manual driving. In total, 25 participants with a mean age of 25.7 years (range 19-36 years) drove four trials in a driving simulator, experiencing four HMIs having the following different stages of automation: baseline (information acquisition-low), sphere (information acquisition-high), carpet (information analysis), and arrow (decision selection), presented as visual overlays on the surroundings. The HMIs provided information during two scenarios, namely a lane change and a braking scenario. Results showed that the HMIs did not significantly affect the drivers' initial reaction to the take-over request. Improvements were found, however, in the decision-making process: When drivers experienced the carpet or arrow interface, an improvement in correct decisions (i.e., to brake or change lane) occurred. It is concluded that visual HMIs can assist drivers in making a correct braking or lane change maneuver in a take-over scenario. Future research could be directed toward misuse, disuse, errors of omission, and errors of commission.

This paper presents formative research to inform the design of intelligent automotive user interfaces. It describes an evaluation of dialogue-based interfaces, mediating the driver to take back control from the autonomous mode of a car. Four concepts designed to increase driver Situation Awareness were evaluated in a driving simulator. They used dialogue-based interaction, where driving-related information was either asked from or repeated by the driver, with the alternative of a countdown-based interface with no additional information. An inclusive set of participants, with a wide age spectrum, tested the interfaces. The shorter and simpler interaction of the countdown timer was most accepted. The interface seeking answers to driving-related questions came next, and the interface requiring repetition of driving-related information, even when augmented by visual and tactile cues, was least accepted. Design guidelines on utilizing dialogue as a means of keeping the driver in the loop during a takeover were thus derived.

Controlling vehicle velocity, by coaching the driver to eco-drive with an advanced driver assistance system (ADAS), is a promising method to decrease fuel consumption and greenhouse gas emissions for combustion engine-driven road vehicles. By using optimal control techniques, such a system may find velocity profiles in real-time that minimize fuel consumption. This is particularly useful to recommend the optimal time to initiate coasting, which is otherwise difficult to estimate by a driver. However, this ADAS should not choose velocities and accelerations that the driver will dislike, such as those that leave too much or too little space to the preceding vehicle, or those that take corners at high speed. To remedy this, we introduce an optimal control model of acceleration that mimics drivers' behavior and combine this with a model of fuel consumption to trade-off driver preferences and fuel savings. We give examples of the velocity profiles recommended in a typical driving scenario to demonstrate the potential fuel savings. Finally, we give details of a prototype system, which has recently been implemented in the driving simulator at the University of Southampton.

Submarine control room operations have not changed much over the past 50 years, despite introduction of new technologies. This study sought to catalog current operations as a baseline for comparing new ways of working. Three scenarios were selected to be examined in both high and low demand: returning to periscope depth, dived tracking, and inshore operations. The scenarios were run in a submarine simulator with currently serving submariners from the Royal Navy. The flow of throughout the submarine command team was examined using event analysis for systemic teamwork (EAST). EAST models collaborative teamwork via three networks: task, social, and information. Results show that the social interactions, information transition, and focus of tasks are different depending on the particular operation being completed and the work demand placed on the command team. There are particular information elements that are fundamental across all scenario types. Task and communication load is not evenly distributed across the team, with potential bottlenecks identified between the sonar controller and operations officer roles. Implications of the results are discussed alongside recommendations for future research.

This paper presents an experimental evaluation of an in-vehicle eco-driving support system that provided auditory, visual, and vibrotactile stimuli for the discouragement of harsh accelerations, and to encourage maximization of the duration of the coasting phase of the vehicle. Behavior when driving normally was compared to that exhibited when participants were asked to drive economically, and to that exhibited when provided with feedback in the three sensory modes, individually and in all combinations thereof. Results suggest that participants were already aware that harsh accelerations are to be avoided when eco-driving; however, additional eco-driving support, particularly that which involved the auditory and haptic modalities, further discouraged these behaviors. The eco-driving information also supported significantly greater coasting distances (when approaching slowing events), a behavior not spontaneously produced by participants when asked to drive economically. Few differences were seen between the effects of the different modes and combinations; however, results taken together suggest that visual-only information may be less effective at encouraging compliance across all participants. The auditory stimulus, although it encouraged compliance, was not well received by participants.

Since first receiving attention in the literature almost 25 years ago, ecological interface design has been applied to a wide variety of man-machine systems across a range of domains. The design framework has its theoretical basis in Gibsonian ecological psychology, and its founding principles draw heavily on Jens Rasmussen's skills, rules, and knowledge (SRK) taxonomy. This paper presents a comprehensive review of the framework's applications since Vicente and Rasmussen's 1992 seminal article detailing the theoretical foundations of the method. There is variation in terms of both the use of the two fundamental components of the method as it was first described, and how it has been supplemented with other phases of the cognitive work analysis; this review highlights these variations with regard to how the design framework has been applied and how these applications have been reported in the literature. The importance of the SRK taxonomy to the framework is specifically discussed following the finding that 40% of reviewed applications do not cite this component despite its centrality to ecological interface design. Attention is drawn to the method's flexibility and adaptability, to its contribution to the content and form of an interface, and a point is made about the importance of being clear and consistent when reporting how the method has been applied and, where appropriate, adapted.

The purpose of this study was to investigate whether the NATO Problem-Approach Space could be mapped to optimal performance of different team structures. The results show that the five team communication structures (chain, Y, circle, wheel, and all-connected) did not generally perform as predicted based on findings in the literature. The team structures all performed most optimally in the same Problem Space conditions: static rate of change, strong information position, and familiarity with the task. Moreover, contrary to predictions, the all-connected team structure did not perform particularly well at all. Instead, the Y team structure produced the highest levels of performance and was, therefore, judged to be the most successful team structure overall. The Y team structure can be seen as a simplified form of typical military command structures including staff officers and subordinate units. Therefore, the findings of the study serve as a reinforcement of the effectiveness of the classic command and control structure.

Private vehicle use contributes a disproportionately large amount to the degradation of the environment we inhabit. Technological advancement is of course critical to the mitigation of climate change, however alone it will not suffice; we must also see behavioural change. This paper will argue for the application of Ergonomics to the design of private vehicles, particularly low-carbon vehicles (e.g. hybrid and electric), to encourage this behavioural change. A brief review of literature is offered concerning the effect of the design of a technological object on behaviour, the inter-related nature of goals and feedback in guiding performance, the effect on fuel economy of different driving styles, and the various challenges brought by hybrid and electric vehicles, including range anxiety, workload and distraction, complexity, and novelty. This is followed by a discussion on the potential applicability of a particular design framework, namely Ecological Interface Design, to the design of in-vehicle interfaces that encourage energy-conserving driving behaviours whilst minimising distraction and workload, thus ensuring safety.

The purpose of this study was to investigate novices learning to use a mission planning system. Novice participants received one training session, followed by three test sessions. This was compared to expert performance. During the test sessions, all participants were required to ‘think aloud’, based on Ericsson and Simon's (Ericsson, K.A. and Simon, H.A., 1993. Protocol analysis: verbal reports as data. Revised ed. Cambridge, MA: The MIT Press) description of the technique. Objective measures of time to complete the task, number of errors made, mouse travel and number of mouse clicks were also recorded. It was hypothesised that error, time and mouse data would reflect the acquisition of skill and improved problem-solving abilities. The verbal data were analysed in an exploratory fashion to see what changes individuals experienced. The results of the study showed that participants acquired a level of skill across sessions, and that the cognitive processes reflected in the verbal protocols changed as a function of this skill acquisition. These changes in cognitive processing mirrored a number of the differences between novices and experts.

In-Vehicle Information Systems (IVISs) integrate most of the secondary functions available within vehicles. These secondary functions are aimed at enhancing the driving experience. To successfully design and evaluate the performance of these systems, a thorough understanding of the task, user, and system is required. This article presents a review of these three variables in the context of IVISs, which aims to enhance understanding of this specific task–user–system interaction. A framework for modeling system performance for the task–user–system interaction is also proposed. This will allow designers and evaluators of IVISs to make predictions about system performance and to design systems that meet a set of criteria for usable IVISs.

This article presents a prototypical model of how decision making can proceed within an environment, independent of situation or actor. Based upon Rasmussen's decision-ladder, an approach for capturing formative descriptions of existing decision-making processes is presented. The example of land-based combat identification is used to demonstrate this approach. Two new representations are introduced to aid domain understanding and support the design of decision support tools. The first maps the links between information elements, system states, and options. The second clusters elements in the decision-making process in terms of their location in the world. The approach presented is not intended to replace existing decision-making analysis techniques, rather, based on similar data collection procedures, its aim is to compliment them with a more formative integrant. By considering decision making independently from actor and context, a flexible approach is presented that is applicable complex sociotechnical systems.

Abstract Digitized mission support systems are currently being introduced in the military arena. The projected benefits include quicker, better informed, more efficient decision making by the teams using them; however, these claims are often made without appropriate scientific testing. A live operational field trial of a land warfare digital mission support system was observed, and an integrated framework of Human Factors methods was used to evaluate performance with the mission support system. The findings indicated that, as a corollary of various flaws associated with the digital system, decision making was found to be more difficult, more drawn out, and more susceptible to error. Rather than augment the decision-making process, in some cases these flaws were seen to create further decision-making requirements for users. In closing, the implications of this study and the wider naturalistic decision making literature for the design of digitized mission support systems are discussed. This research from the Human Factors Integration Defence Technology Centre was part-funded by the Human Sciences Domain of the UK Ministry of Defence Scientific Research Program. The work described was undertaken while all authors worked in the Ergonomics Research Group at Brunel University. We thank two anonymous reviewers for their helpful comments on an earlier, shorter version of this article for the NDM 09 conference. Notes

Naturalistic Decision Making (NDM) has a special place in the hearts of researchers because of its exploratory nature, and the emphasis on investigating how experts make decisions in “real” situati...

Interface problems have been cited as critical factor in the suboptimal performance of a large-scale digital command and control system. A live field study involving fully functioning Brigade and Battlegroup headquarters was observed. More than 3,000 communications events were extracted and analyzed in terms of their quantity, direction, and content. The effect of the interface problem was pronounced. Voice mediated communications (conducted by radio and avoiding the interface entirely) were superior at converting “data” into “information.” In cases where the interface was relied upon, users seized on a highly simplistic comms. facility and put it to use in ways that were not anticipated. The findings added significant value in terms of the phased, real-world delivery of this system. Very little human factors analysis had been performed previously. The current analysis, therefore, was able to provide a considerable amount of insight and justification for further design iterations. The take-away message: Neither networked technology nor the large quantities of raw data carried by it are sufficient to guarantee successful naturalistic decision making.

Persuasive Technology has the potential to influence user behavior for social benefit, e.g. to reduce environmental impact, but designers are lacking guidance choosing among design techniques for influencing interaction.

Computers are ubiquitous, in terms that they are everywhere, but does this mean the same as ubiquitous computing? Views are divided. The convergent device (one-does-all) view posits the computer as a tool through which anything, and indeed everything, can be done (Licklider & Taylor, 1968). The divergent device (many-do-all) view, by contrast, offers a world where microprocessors are embedded in everything and communicating with one another (Weiser, 1991). This debate is implicitly present in this issue, with examples of the convergent device in Crook & Barrowcliff's paper and in Gay et al's paper, and examples of the divergent devices in Thomas & Gellersen's paper and Baber's paper. I suspect both streams of technology are likely to co-exist.

Cars offer an excellent example of ubiquitous computing, and a technological revolution is currently underway that will eventually see in-vehicle computers empowered with increasingly complex sections of the driving task. In this article, we critically review the effect of ubiquitous computing in cars with reference to the psychology of the driver and present a survey of automotive researchers drawn from five major carmakers. The results illustrate the role of the computer in vehicles over the short, medium, and long term. Systems that are likely to be fitted into vehicles in the next 5 years include sophisticated electronic architectures and greater penetration of navigation and telematics systems. In the next 5 to 15 years drive by wire and collision sensing are anticipated. In the long term, 15 years and beyond, advanced driver-assistance systems will increasingly automate the driving task, and in-car personal computers and Internet will be commonplace. We conclude that the increased complexity and prominence of computing in cars requires further investigation of the needs, abilities, and limitations of the driver if the aims of safety, efficiency, and enjoyment, as well as greater ubiquity, are to be realized.