Human-automation Interaction Research Articles

The Effect of Correlated Observations on Human-Automation Interaction in a Signal Detection Task

Automated signal detection aids assist human operators in various activities (e.g., medical diagnosis). However, operators tend to interact with these aids in suboptimal ways. An under-considered factor that might influence operators’ aid use is the correlation between the aid’s and the operator’s observations. Although prior research has generally assumed that human operators and automated aids rely on independent information sources, correlated observations may be common in naturalistic automation-aided detection tasks. The present study explored the effects of correlated observations on automation-use by analyzing performance in a numerical signal detection task using the Contingent Cutoff (CC) model, a statistical cognitive model of automation-aided decision making. Participants completed the task with and without assistance of an aid, and in one of two conditions: correlated observations vs. uncorrelated observations. The CC model accurately described observed performance for both conditions. Overall aid use efficiency was unaffected by correlated observations.

Performance and Relative Risk Dynamics during Driving Simulation Tasks under Distinct Automation Conditions

Risk has been a key factor influencing trust in Human-Automation interactions, though there is no unified tool to study its dynamics. We provide a framework for defining and assessing relative risk of automation usage through performance dynamics and apply this framework to a dataset from a previous study. Our approach allows us to explore how operators’ ability and different automation conditions impact the performance and relative risk dynamics. Our results on performance dynamics show that, on average, operators perform better (1) using automation that is more reliable and (2) using partial automation (more workload) than full automation (less workload). Our analysis of relative risk dynamics indicates that automation with higher reliability has higher relative risk dynamics. This suggests that operators are willing to take more risk for automation with higher reliability. Additionally, when the reliability of automation is lower, operators adapt their behavior to result in lower risk.

How Do Advanced Driver Assistance Systems Fit into Level of Automation Frameworks?

Advanced driver assistance systems (ADAS) have been increasingly incorporated in cars for nearly four decades and have changed the relationship of the driver to the driving task substantially. Over this period, original equipment manufacturers (OEMs) have developed similar ADAS functions (e.g., adaptive cruise control, lane keeping assist, forward collision warning), but these functions lack uniformity in their implementation such that there is the possibility of negative transfer of learning across different implementations of the same ADAS function. This brief theoretical paper aims to highlight issues around using some existing human-automation interaction (HAI) frameworks that have been used to classify vehicle automation and discuss considerations for better ADAS classification to inform their design and support safe and satisfactory use.

Trustworthy UAV Relationships: Applying the Schema Action World Taxonomy to UAVs and UAV Swarm Operations

Human Factors play a significant role in the development and integration of avionic systems to ensure that they are trusted and can be used effectively. As Unoccupied Aerial Vehicle (UAV) technology becomes increasingly important to the aviation domain this holds true. This study aims to gain an understanding of UAV operators’ trust requirements when piloting UAVs by utilising a popular aviation interview methodology (Schema World Action Research Method), in combination with key questions on trust identified from the literature. Interviews were conducted with six UAV operators, with a range of experience. This identified the importance of past experience to trust and the expectations that operators hold. Recommendations are made that target training to inform experience, in addition to the equipment, procedures and organisational standards that can aid in developing trustworthy systems. The methodology that was developed shows promise for capturing trust within human-automation interactions.

Detecting Human Trust Calibration in Automation: A Convolutional Neural Network Approach

There is a general lack of studies that are aimed at monitoring and detecting an operator's trust calibration, even though detecting someone's adjusted trust towards automation is essential to prevent misuse and disuse of automation. The goal of this article is to propose a convolutional neural network (CNN) based framework to estimate operators’ trust levels and detect their trust calibration in automation using image features of electroencephalogram (EEG) signals preserving temporal, spectral, and spatial information. Thirteen participants performed a set of automated Air Force multiattribute task battery tasks that differed in reliability (High/Low) and credibility (High/Low) levels. The proposed framework was compared with three machine learning methods—naïve bayes, support vector machine, multilayer perceptron—in terms of accuracy, sensitivity, and specificity of trust estimation and detection of trust calibration. Results of this article showed that the proposed framework had the highest performance of both trust estimation and detection of trust calibration in automation compared to the other comparison methods. This indicates that the proposed framework using the CNN classifier with the image-based EEG features could be an applicable model for estimating multilevel trust and detecting trust calibration during human-automation interaction. Also, it can help to prevent disuse and misuse of automation by estimating operators’ trust levels and monitoring their trust calibration in automation.

Human-automation interaction for helicopter flight: Comparing two decision-support systems for navigation tasks

This paper investigates the effects of different automation design philosophies for a helicopter navigation task. A baseline navigation display is compared with two more advanced systems: an advisory display, which provides a discrete trajectory suggestion; and a constraint-based display, which provides information about the set of possible trajectory solutions. The results of a human-in-the-loop experiment with eight pilot participants show a significant negative impact of the advisory display on pilot trajectory decision-making: out of the 16 encountered off-nominal situations across the experiment, only 6 were solved optimally. The baseline and constraint-based display both lead to better decisions, with 14 out of 16 being optimal. However, pilots still preferred the advisory display, in particular in off-nominal situations. These results highlight that even when a support system is preferred by pilots, it can have strong inadvertent negative effects on their decision-making.

Insight into User Acceptance and Adoption of Autonomous Systems in Mission Critical Environments

With Industry 4.0 the immense progression of Artificial Intelligence (AI) technology has introduced new challenges for engineers to effectively design human-automation interaction in autonomous systems that are mission critical. Although various autonomous systems are currently being utilized in mission critical environments, there is limited literature and research on which factors affect the acceptance and adoption of said systems. Understanding which factors are most critical for the human-automation interaction could lead to seamless acceptance and adoption and more effective and less expensive missions. Findings of 47 semi-structured interviews revealed ease of use and system reliability to be significant factors for the acceptance and adoption of autonomous systems independent of the level of automation. Through our findings we expand on the current technology acceptance models by including mission critical factors. Emphasis is given to the discussion and consideration of the human factors and engineering approaches associated with the design of autonomous systems for mission critical environments that are needed to empower tomorrow’s users with effective AI systems technology.

Human-centred risk assessment for a land-based control interface for an autonomous vessel

Autonomous ferries are providing new opportunities for urban transport mobility. With this change comes a new risk picture, which is characterised to a large extent by the safe transition from autonomous mode to manual model in critical situations. The paper presents a case study of applying an adapted risk assessment method based on the Scenario Analysis in the Crisis Intervention and Operability study (CRIOP) framework. The paper focuses on the applicability of the Scenario Analysis to address the human-automation interaction. This is done by presenting a case study applying the method on a prototype of a Human–Machine Interface (HMI) in the land-based control centre for an autonomous ferry. Hence, the paper presents findings on two levels: a method study and a case study. A concept of operation (CONOPS) and a preliminary hazard analysis lay the foundation for the scenario development, the analysis, and the discussion in a case study workshop. The case study involved a Scenario Analysis of a handover situation where the autonomous system asked for assistance from the operator in a land-based control centre. The results include a list of identified safety issues such as missing procedures, an alarm philosophy and an emergency preparedness plan, and a need for explainable AI. Findings from the study show that the Scenario Analysis method can be a valuable tool to address the human element in risk assessment by focusing on the operators’ ability to handle critical situations.

Psychometric properties of the Chinese version of the trust between People and Automation Scale (TPAS) in Chinese adults

Trust in automation plays a leading role in human-automation interaction. As there lack of scales measuring trust in automation in China, the purpose of this study was to adapt the trust between People and Automation Scale (TPAS) into Chinese and to demonstrate its psychometric properties among Chinese adults. A total of 310 Chinese adults were randomly selected as sample 1, and 508 Chinese adults as sample 2. Results of the item analysis revealed that each item had a good quality, and the exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) suggested that the two-factor model with 12 items was the best fitting model. In addition, the TPAS was positively correlated with Interpersonal Trust Scale (ITS), proving good evidence based on relations to other variables to support the TPAS. In sum, the study suggested that the Chinese version of the TPAS could be used as an effective tool to assess trust in automation in the Chinese context.

Providing different levels of accuracy about the reliability of automation to a human operator: impact on human performance

Previous research has suggested that supervising automation can lead to a decrease in human performance, especially when automation is not totally reliable. Providing context-related information about reliability can help operators to better adjust their behaviour in a human–automation interaction context. However, previous studies have not specified the level of accuracy that this information should provide to the human operator. The objective of this study was to investigate the effects of different levels of information accuracy about an automation’s reliability on human performance. Results showed that accuracy of information about reliability improves performance when specific percentages of reliability were given to the participants. Participants had a better performance in the condition of high accuracy of information. A link between perceived reliability and trust was found: the more the trust in automation increased, the more the perceived reliability increased. Practitioner summary: The experiment dealt with how accurate information about automation’s reliability influences people’s performance when supervising an automated task. Overall, this research suggests that designing systems that provide accurate, useful information can reduce the frequency of automation bias. Trust and perceived reliability of automation are related. Abbreviations: MATB: multi attribute task battery

How the initial level of trust in automated driving impacts drivers’ behaviour and early trust construction

Trust in Automation is known to influence human-automation interaction and user behaviour. In the Automated Driving (AD) context, studies showed the impact of drivers’ Trust in Automated Driving (TiAD), and linked it with, e.g., difference in environment monitoring or driver’s behaviour. This study investigated the influence of driver’s initial level of TiAD on driver’s behaviour and early trust construction during Highly Automated Driving (HAD). Forty drivers participated in a driving simulator study. Based on a trust questionnaire, participants were divided in two groups according to their initial level of TiAD: high (Trustful) vs. low (Distrustful). Declared level of trust, gaze behaviour and Non-Driving-Related Activities (NDRA) engagement were compared between the two groups over time. Results showed that Trustful drivers engaged more in NDRA and spent less time monitoring the road compared to Distrustful drivers. However, an increase in trust was observed in both groups. These results suggest that initial level of TiAD impact drivers’ behaviour and further trust evolution.

Aptitude Requirements for Human Operators in Human–Automation Interaction

Abstract. The purpose of this work was to identify individual differences that affect aptitude requirements for jobs involving autonomous systems and human–automation interaction (HAI). This was addressed in two stages. First, we conducted a literature review of task demands and operator states relevant to HAI. On the basis of this review, we formed a model for understanding performance as a composite of operator states, operator behaviors, and distal outcomes. Second, we conducted a meta-analysis of correlations between individual differences and criteria reflecting job demands of an HAI environment. Results suggest cognitive skills such as working memory are important to performance in an HAI context. Inconsistent findings for personality across studies underscore the need for more research. Measurement challenges and research gaps are identified.

Human Factors Considerations and Metrics in Shared Space Human-Robot Collaboration: A Systematic Review.

The degree of successful human-robot collaboration is dependent on the joint consideration of robot factors (RF) and human factors (HF). Depending on the state of the operator, a change in a robot factor, such as the behavior or level of autonomy, can be perceived differently and affect how the operator chooses to interact with and utilize the robot. This interaction can affect system performance and safety in dynamic ways. The theory of human factors in human-automation interaction has long been studied; however, the formal investigation of these HFs in shared space human-robot collaboration (HRC) and the potential interactive effects between covariate HFs (HF-HF) and HF-RF in shared space collaborative robotics requires additional investigation. Furthermore, methodological applications to measure or manipulate these factors can provide insights into contextual effects and potential for improved measurement techniques. As such, a systematic literature review was performed to evaluate the most frequently addressed operator HF states in shared space HRC, the methods used to quantify these states, and the implications of the states on HRC. The three most frequently measured states are: trust, cognitive workload, and anxiety, with subjective questionnaires universally the most common method to quantify operator states, excluding fatigue where electromyography is more common. Furthermore, the majority of included studies evaluate the effect of manipulating RFs on HFs, but few explain the effect of the HFs on system attributes or performance. For those that provided this information, HFs have been shown to impact system efficiency and response time, collaborative performance and quality of work, and operator utilization strategy.

Does automation trust evolve from a leap of faith? An analysis using a reprogrammed pasteurizer simulation task

Trust is a critical factor that drives successful human-automation interaction in a myriad of modern professional environments. One seminal work on human-automation trust is Muir and Moray (1996) showing that human-machine trust evolves from faith, then dependability, and finally predictability in a simulated supervisory control task. However, our recent work failed to replicate the finding of the original study, calling for further replication efforts. Experiment 1 aimed to fully replicate Muir and Moray (1996) where participants performed a simulated pasteurizer task. Experiment 2 attempted to replicate Experiment 1 using participants who major in Engineering as used in the original study. Both experiments showed that dependability was the best initial predictor of trust, building later to predictability and faith. Two experiments consistently failed to support both the hypothesis proposed by Muir and Moray (1996), that trust develops from predictability to dependability to faith, and their original findings that trust develops initially from faith. The results of the current experiments challenge this widely cited view of how human-machine trust develops. Modern automation designers should be aware that dependability might control initial trust development for general users and incorporate dependability information into their designs.

Human-Automation Interactions in Automated Logistics

Human-Automation Interactions in Automated Logistics

Convention-based Intent Integration in a Haptic Shared Control Paradigm

This paper establishes a platform to study the principles of convention formation in a haptic shared control framework wherein both humans and automation collaboratively control the steering of a semi-automated ground vehicle. We then apply these principles to determine optimal strategies for exchanging the control authority between human drivers and an automation system. In the first step, we introduce a modular structure to separate partner-specific conventions from task-dependent representations. Next, we develop a map that connects different forms of conventions to the outputs of human-automation interaction. Finally, using the convention map, we developed a reinforcement-learning-based model predictive controller to enable the automation system to learn complex policies and adapt its behavior accordingly. We applied the proposed platform to the problem of intent negotiation for resolving a conflict. Specifically, we considered a scenario where both humans and automation detect an obstacle but choose different paths to maneuver around the obstacle. The simulation results demonstrate that the convention-based handover strategies can successfully resolve a conflict and improve the performance of the human-automation teaming.

Requirements for Computational Approaches to Analyzing Resilience in Human-Machine Teams

Human-machine teams (HMTs) in complex work domains need to be able to adapt to variable and uncertain work demands. Computational modeling and simulation can provide novel approaches to the evaluation of HMTs performing complex joint activities, affording large-scale, quantitative analysis of team characteristics (such as system architecture and governance protocols) and their effects on resilience. Drawing from literature in resilience engineering, human-automation interaction, and cognitive systems engineering, this paper provides a theoretical exploration of the use of computational modeling and simulation to analyze resilience in HMTs. Findings from literature are summarized in a set of requirements that highlight key aspects of resilience in HMTs that need to be accounted for in future modeling and evaluation efforts. These requirements include a need to model HMTs as joint cognitive systems, the need to account for the interdependent nature of activity, the temporal dynamics of work, and the need to support formative exploration and inquiry. We provide a brief overview of existing modeling and simulation approaches to evaluating HMTs and discuss further steps for operationalizing the identified requirements.

Studying Robot Vacuums Using Online Retailer Reviews to Understand Human-Automation Interaction

Robot vacuums are a type of every-day automation and may serve as a possible domain for studying human-automation interactions. Online reviews of four models of autonomous robot vacuums, representing different levels of price and automation, were collected from multiple retailers to understand user interactions and issues encountered. Using a thematic analysis, reviews were categorized into three themes: prior experience with robot vacuums, navigation, and mapping, and troubleshooting issues. The results suggested that new users were more satisfied with their robots than experienced users. Furthermore, price and brand name may influence users’ mental models of their vacuums. Owners of more expensive vacuums appeared less willing to change their behavior to accommodate the vacuum’s limitations. These findings suggest that robot vacuums may be a useful gateway domain for understanding everyday-users’ interactions with automation, and that online reviews are a potential source of information on user experiences.

Dynamics of Trust in Automation and Interactive Decision Making during Driving Simulation Tasks

As technological advancements and lowered costs make self-driving cars available to more people, it becomes important to understand the dynamics of human-automation interactions for safety and efficacy. We used a dynamical approach to examine data from a previous study on simulated driving with an automated driving assistant. To maximize effect size in this preliminary study, we focused the current analysis on the two lowest and two highest-performing participants. Our visual comparisons were the utilization of the automated system and the impact of perturbations. Low-performing participants toggled and maintained reliance either on automation or themselves for longer periods of time. Decision making of high-performing participants was using the automation briefly and consistently throughout the driving task. Participants who displayed an early understanding of automation capabilities opted for tactical use. Further exploration of individual differences and automation usage styles will help to understand the optimal human-automation-team dynamic and increase safety and efficacy.

Automation as an equal team player for humans? – A view into the field and implications for research and practice

The practical reality and feasibility of Human-Autonomy Teaming (HAT) are analyzed from an experts' point of view, considering current possibilities of various fields. We aim to find out whether the topics discussed scientifically are also practically relevant, to identify requirements for successful HAT, and to derive further research needs. Intensive guideline-based interviews with 28 experts from different industries are conducted and compared to the results of our literature review. The topics discussed scientifically are also practically relevant. Today's technology is far from being able to meet the practical requirements for successful HAT, as postulated in the literature. Contrary to the Human-Automation Interaction, the concept of HAT is hardly applied in the field. Identified key aspects for successful HAT are converted into a model. Future research needs with practical impact exist especially in the area of heterarchy, system knowledge, anticipation of mental states, and consideration of human needs and emotions.