Engineering Psychology and Cognitive Ergonomics

Delays between user input and the system’s reaction in control tasks have been shown to have a detrimental effect on performance. This is often accompanied by increases in self-reported workload. In the current work, we sought to identify physiological measures that correlate with pilot workload in a conceptual aerial vehicle that suffered from varying time delays between control input and vehicle response. For this purpose, we measured the skin conductance and heart rate variability of 8 participants during flight maneuvers in a fixed-base simulator. Participants were instructed to land a vehicle while compensating for roll disturbances under different conditions of system delay. We found that control error and the self-reported workload increased with increasing time delay. Skin conductance and input behavior also reflect corresponding changes. Our results show that physiological measures are sufficiently robust for evaluating the adverse influence of system delays in a conceptual vehicle model.

This paper studies the effects of automated support for workload distribution in traffic control teams on human values such as security, autonomy and privacy. The paper describes a workshop in which the support system’s stakeholders, their values, and the effects of the support system on these values were analyzed. The workshop results were used to derive design recommendations that minimize the negative effects on the stakeholders’ values. The main conclusions are that in order to minimize negative impacts on privacy, trust and team spirit, the type and amount of information that is shared to improve workload distribution should be adjustable, depending on the role of the receiving party.

We present an empirical study where the effects of three levels of system transparency of an automated target classification aid on fighter pilots’ performance and initial trust in the system were evaluated. The levels of transparency consisted of (1) only presenting text–based information regarding the specific object (without any automated support), (2) accompanying the text-based information with an automatically generated object class suggestion and (3) adding the incorporated sensor values with associated (uncertain) historic values in graphical form. The results show that the pilots needed more time to make a classification decision when being provided with display condition 2 and 3 than display condition 1. However, the number of correct classifications and the operators’ trust ratings were the highest when using display condition 3. No difference in the pilots’ decision confidence was found, yet slightly higher workload was reported when using display condition 3. The questionnaire results report on the pilots’ general opinion that an automatic classification aid would help them make better and more confident decisions faster, having trained with the system for a longer period.

This paper describes the requirements elicitation process and the subsequent system requirements for an advanced cockpit to reduce crew workload and stress. The paper outlines the need for a step-change in technology and operational practices to ensure the continued safety of a transport system which is predicted to grow. The ACROSS project aims to develop advanced cockpit solutions to reduce workload and stress in an increasingly congested aviation transport system. Six types of requirements were derived including aviate requirements, and navigate, communicate, manage systems, crew monitoring, and crew incapacitation requirements. The research project is currently specifying the human factors requirements for the technologies to achieve improved operational safety.

Workload research in command, information and process-control centers, resulted in a modular and formal Cognitive Load and Emotional State (CLES) model with transparent and easy-to-modify classification and assessment techniques. The model distinguishes three representation and analysis layers with an increasing level of abstraction, focusing on respectively the sensing, modeling, and reasoning. Fuzzy logic and its (membership) rules are generated to map a set of values to a cognitive and emotional state (modeling), and to detect surprises of anomalies (reasoning). The models and algorithms allow humans to remain in the loop of workload assessments and distributions, an important resilience requirement of human-automation teams. By detecting unexpected changes (surprises and anomalies) and the corresponding cognition-emotion-performance dependencies, the CLES monitor is expected to improve team’s responsiveness to new situations.

The laboratory study we are carrying out is aimed at discovering possible correlations between multitasking activity, workload and the attribution of mental states to technological systems. The scores of mental states attribution provided by subjects allotted to three different experimental conditions (one task, two concurrent tasks, three concurrent tasks) have been compared. Preliminary results show an increase in the tendency to attribute mental states as the operational workload increases.

Neuronal workload measurement is a key-technology for optimizing work conditions in human-machine systems. Specific aims are the identification of neurophysiological parameters indicative for workload and their validation by systematic variation of external load conditions.

The battery consists of tasks with diverse complexity and difficulty. The sample consists of 34 people and shows high variability in respect to the cognitive capacity and hence to the experienced mental workload. The electroencephalogram (EEG) as well as further workload relevant bio signal data and the NASA-TLX as a subjective questionnaire method are registered.

Results from the NASA-TLX questionnaire reveal the predominant role of the mental dimension at the implemented task battery. Furthermore, the NASA-TLX indicates the existence of diverse levels of difficulty with several tasks per level. Analysis of EEG spectra demonstrates an increase of frontal theta band power and a decrease of alpha band power with increasing task difficulty level.

As working memory (WM) is compromised with advancing age, older people may have performance deficits in WM tasks. This is probably due to a great number of WM operations which should be performed for extended periods of time. The reduction of a number of these operations was expected to reduce WM load and age-related deficits in WM performance. Fifty younger (29±3 years) and 49 older (55±3 years) healthy employees had to perform a visual 0-back (oddball) task and a 2-back task. Within the 2-back task, the short (3 or 4 items, low WM load) and long (5 or 6 items, high WM load) target-to-target sub-sequences were analysed separately. Older workers performed worse than younger ones at higher WM loads, except for the oddball condition and low WM load condition. The N2 latency of the event-related potentials (ERPs) increased with WM load and was generally longer in older than younger adults. In addition, the N2 latency decreased with WM load in younger adults but did not change in older ones. Older workers also showed a delayed P3a as well as a delayed and reduced P3b. By contrast, age-related enhancements of the occipital N1 and frontal P2 components under WM load were observed. The parietal slow positive wave (SPW) increased under high WM load but did not vary with age. The results indicate that older adults are able to compensate for age-related WM impairments when the amount of WM operations required does not exceed the limits of their WM capacity. The allocation of cognitive resources to stimulus encoding (N1) and memory retrieval (P2) are putative neuronal mechanisms for these WM improvements. However, older adults have maintenance problems at higher WM loads. This is associated with deficits in neuronal processes relating to response selection (N2), detection of changes in WM representations (P3a) and WM updating (P3b). These results provide a basis for the development of work load criteria and training opportunities for older workers who have to do complex work requiring working memory.

Reorganization of a rail control post may affect its ability to cope with unexpected disruptions. The term ‘resilience’, the ability to manage spare adaptive capacity when unexpected events occur, encapsulates this situation. This paper focuses on the workload adaptive capacity through a method for revealing workload weak-resilience-signals (WRS). Three different workload measurements are adapted to identify structural changes in workload. The first, executed cognitive task load, targets system activities. The second, integrated workload scale, is a subjective measure. The last, heart rate variability, identifies physiological arousal because of workload. An experiment is designed to identify the workload change and distribution across group members during disruptions. A newly defined Stretch, the reaction of the system to an external cluster-event, is used to reveal a workload WRS. The method is suitable for real-time usage and provides the means for the rail signaler to influence the system through his subjective workload perception.

Human errors have become the major threat to flight safety. Improper workload imposed on the pilot was supposed to be one of the most critical causes for human error. Among the various workload measure techniques, physiological measures were promising because of its objectivity and capability of continuous measure. However, the mechanisms of the physiological reactions were complex. The sensitivity and diagnosticity of the physiological parameters should be carefully discussed. This paper study several physiological reactions in a simulated flight task including blink, saccade, fixation, pupil diameter, heart rate, respiration etc. Statistical analysis was made to test the sensitivities of the physiological parameters and the diagnosticity of parameter was also discussed in the paper. These results could provide some guidance for the selection of the physiological parameters during the assessment of the pilot’s workload.

The present study attempted to establish an effective discrimination and prediction model that can be applied to evaluate mental workload changes in human-machine interaction processes on aircraft flight deck. By adopting a combined measure based on primary task measurement, subjective measurement and physiological measurement, this study developed both experimental measurement and theoretical modeling of mental workload under flight simulation task conditions. The experimental results showed that, as the mental workload increased, the peak amplitude of Mismatch negativity (MMN) was significantly increased, SDNN (the standard deviation of R-R intervals) was significantly decreased,the number of eye blink was decreased significantly. Finally, a comprehensive mental workload discrimination and prediction model for the aircraft flight deck display interface was constructed by the Bayesian Fisher discrimination and classification method. The model’s accuracy was checked by original validation method. When comparing the prediction and discrimination results of this comprehensive model with that of single indices, the former showed much higher accuracy.

This study examined both the behavioral performance and the brain mechanisms of post-error adjustments under long-term psychological stress by using ERP technique. Forty two participants who had been exposed to long-term exam preparation (versus 21 controls who were not exposed to such exam) performed a Go/NoGo task while electroencephalograms were recorded. We used Cohen’s Perceived Stress Scale to assess their chronic stress level, and results suggested that participants in the exam group had higher levels of perceived stress. Although the behavioral performance of post-error trials had no difference between two groups, the exam group elicited significantly decreased P3 amplitude than the non-exam group in the post-error condition. Furthermore, the P3 amplitude in the post-error condition was negatively correlated with the perceived stress scores, suggesting that long-term psychosocial stress may lead to the decrease of the attention resource after committing an error.

The perception of transparency in architecture poses a major cognitive challenge due to the nearly faultless quality of large-scale light-permeable materials. Previous experiments demonstrated the significance of the virtual image in the perception of light-permeable materials. The separation of this image in photographs of architecture proved crucial in the research, giving the freedom to manipulate independently the components of the perceived scene. In the paper the author presents a new methodology of separating and manipulating real and virtual images, which is based on real-life photographs, not computer-generated scenes. The paper also proposes a step-by-step image processing algorithm which helps to better understand the cognitive processes of the human visual system, and presents sample results of this method.

Mobile computers and smartphones are often used while their users are walking. From an ergonomic viewpoint, this requires a thorough design of the user interface. Although styleguides provide multiple recommendations there is little known about basic human factors’ issues. This study provides recommendations for the visual design by analyzing the influence of walking on visual acuity with a mobile computer. N=22 volunteers participated in the experiment comparing visual acuity during standing, slow walking and fast walking. Additional conditions referred to indoor (treadmill) and outdoor (free walking) situations. The results show that walking speed has a highly significant influence on visual acuity. The results are independent of the indoor or outdoor condition. The decrease of visual acuity is similar to a row on a common eye chart. For compensating this decrease, letters and icons on a mobile device should be enlarged by about 20%.

Initial analysis of a first Remote Control Tower (RTO) field test with an experimental videopanorama system [1] [2] under quasi operational conditions has shown performance deficits quantified by two-alternative aircraft maneuver discrimination tasks [3]. RTO-controller working position (CWP-) performance was compared with that one of the conventional tower-CWP with direct out-of-windows view by means of simultaneous aircraft maneuver observations at both operator positions, and it was quantified using discriminability d’ and Bayes inference. Here we present an extended data analysis using nonparametric discriminability A and we discuss the RTO performance deficit in terms of the information processing (IP) theory of Hendy et al. [4]. As initial working hypothesis this leads to the concept of time pressure (TP) as one major source of the measured response errors. We expect the RTO-performance deficits to decrease with the introduction of certain automation features to reduce time pressure and improve the usability of the videopanorama system. A fit of the experimental data with a modified error vs. TP function provides some evidence in support of the IP/TP-hypothesis, however more specifically designed experiments are required for obtaining sufficient confidence.

A perceptual object is created when an observer perceives a single “thing” even though it is comprised of separately perceptible components. A perceptual object has permanence across changes in position and, within limits, changes in the arrangement or composition of the constituent parts. The present research is an examination of the emergence of perceptual objects solely from the dynamics of data presentation. Ten participants viewed presentations of dot patterns that varied in persistence, color, and opacity. Half the presentations were set to have parameters optimized to promote object perception. The same data were also presented with other (non-optimized) settings. Participants correctly detected about 95% of the targets presented with optimized settings and less than 5% of the same targets with non-optimized settings. There were very few false alarms. Participants perceived a unitary object that was hopping from place to place on display despite changes in speed, direction, or color.

People could extract mean expression of multiple faces pretty precisely. However, the mechanism of how we make such ensemble representation was far from clear. This study aimed to explore how faces in the foveal and extrafoveal vision contribute to the ensemble representation and whether the emotion of faces modulates the contribution. In the experiment, the expressions of foveal and extrafoveal faces were independently manipulated by changing the ratio of happy vs. angry faces. The participants reported whether the overall emotion was positive or negative. The results showed that faces in the foveal vision were given more weight than those in the extrafoveal vision in ensemble emotional representation. In addition, the ensemble perception was more accurate when faces in the extrafoveal vision were positive. These findings have great implications for the emotional design in interactive systems, especially when there are multiple users or multiple avatars presented on the screen.

It has been reported that the increase in driving speed incurs a shortened pupil distance, termed as a visual tunneling phenomenon. However, our experimental investigation shows that the effect of driving speed on driver’s visual attention should be understood in terms of the maximum field of view that can balance against the maximum amount of visual information a driver can take/handle against. More specifically, our experimentation shows the following: For the sake of ensuring safety, drivers tend naturally to take as much visual information as possible, should it be allowed in terms of the maximum amount of visual information they can take/handle. However, the maximum visual information a driver can take/handle is different among individuals according to their level of driving expertise. Since the increase of driving speed increases the amount of visual information to process, a driver may be able to expand their field of view only up to the point where the amount of visual information to process balances the maximum amount he/she can take/handle. Beyond this point, the increased anxiety stress may even further reduce the maximum visual information a driver can take/handle, thus further diminishing the field of view, leading to a tunneling effect.

In this paper we describe the newly created eye tracking annotated database

Eye-Tracking Movie Database ETMD

and give some preliminary experimental results on this dataset using our new visual saliency frontend. We have developed a database with eye-tracking human annotation that comprises video clips from Hollywood movies, which are longer in duration than the existing databases’ videos and include more complex semantics. Our proposed visual saliency frontend is based on both low-level features, such as intensity, color and spatio-temporal energy, and face detection results and provides a single saliency volume map. The described new eye-tracking database can become useful in many applications while our computational frontend shows to be promising as it gave good results on predicting the eye’s fixation according to certain metrics.

The aim of this study is to explore the time course of the selective consolidation process. While maintaining one color, which was selected from an initially array of four colors in three different intervals, participants had to perform a visual search and ignore 1) a singleton distractor that matched the color maintained in WM, 2) a new singleton that did not match the maintained color, or 3) a uniformly colored distractor display (i.e., no singletons). After that, they should match a probe color with the maintained color. WM performance for the color was significantly impaired in the matching and new color distractor conditions relative to the uniformly distractor condition when the interval for selective consolidation was short (35 or 50 ms/item), but was identical across these three conditions when this interval was relative long (65 ms/item). This result indicated an astonishingly fast process of selection before the color consolidation process.

Flight simulators are often assessed in terms of how well they imitate the physical reality that they endeavor to recreate. Given that vehicle simulators are primarily used for training purposes, it is equally important to consider the implications of visualization in terms of its influence on the user’s control performance. In this paper, we report that a complex and realistic visual world environment can result in larger performance errors compared to a simplified, yet equivalent, visualization of the same control task. This is accompanied by an increase in subjective workload. A detailed analysis of control performance indicates that this is because the error perception is more variable in a real world environment.

We examined the function of the central visual field by using the newly developed VR system that was consisted with a wide-view HMD and an eye-tracker for restricting an arbitrary area of human visual field. Subjects were asked to walk through short virtual mazes under different visual condition in which 10 or 20 degree of their central visual field was restricted artificially with the system. Results indicated 1) Times for walking through the entire maze under the visual condition with 10 degrees of the central visual field restricted in synchronization were longer than times under the condition in which 10 degrees of the fixed central area of screen were restricted. 2) For walking through the area with two dead ends, walking times under the condition in which 20 degrees of the central visual field were restricted were longer than under the condition in which 10 degrees of the central visual field were restricted.

Data entry by physicians is a critical aspect in the health care domain, in which errors may lead to severe consequences for patients. This paper describes and discusses these aspects to support human-centered design of appropriate human-computer interaction technology. The following issues will be addressed, including system aim, users’ profiles, interaction devices and environment of use, to cite the most important. Our work is based on a literature survey, questionnaires, and an active participatory design process conducted with healthcare professionals. Since the crucial factor is context of use, we elicited several relevant contextual attributes that enabled us to create and incrementally upgrade a context model. This conceptual model is intended to support a scenario-based design approach of future data entry systems. A few scenarios are provided.

The electrophysiology technique now provides an alternative way to evaluate users’ emotional states in real time, but how to confirm the valence of emotions using these techniques is still a concern to researchers. Frontal alpha asymmetry (FAA) is often used as an index of pleasantness or liking in neuromarketing, but results in related fields are not consistent. In this study, we investigated the emotional states of users interacting with mobile phone applications (APPs) using FAA. Twenty participants participated in this experiment. They were asked to complete several tasks in a scene of everyday life using three APPs of the same type. EEG data and subjective evaluations were recorded during the experiment. The FAA results showed a positive trend when using an APP that provided an excellent user experience. The mechanism of emotional change during interacting with mobile applications and the implications of this research are also discussed in this study.

We propose a new concept, disuse atrophy in cognitive abilities, i.e., cognitive disuse atrophy. Generally, the term “disuse atrophy” has been used to describe physical atrophy, such as muscle wasting. We advance the idea that disuse atrophy appears not only as physical loss but also as a loss of cognitive abilities. To understand the mechanisms underlying cognitive disuse atrophy, we note the duality of cognitive activities such as performance- and learning-oriented activities when engaging in tasks. It is crucial to investigate the balancing of these two types of activities as the assistance dilemma in learning science. We explored principles for controlling this balance based on two theories: cognitive load theory and goal achievement theory. Cognitive load theory distinguishes three types of cognitive loads. This theory proposes to suppress the extraneous load to the minimum, while assigning adequate amounts of the germane load for learning-oriented activities into working memory, and still leave enough resources for the intrinsic load of performance-oriented activities. Goal achievement theory assumes principles from the viewpoint of goal setting. Specifically, orientation to a performance goal activates performance-oriented activities, and orientation to a learning goal causes learners to direct their efforts to learning-oriented activities.

Cognitive engineering aims to provide operators with immediate access to as much relevant information as possible. However, this can encourage display-based strategies that do not involve committing information to memory. To overcome this problem, a somewhat counterintuitive method is discussed, based upon the theory of soft constraints [1], that involves delaying access to some critical information by one or two seconds. This design technique induces a more planful and memory-based strategy that can improve recall, develop more planning behavior, improve problem solving, and protect against the negative effects of interruption. Furthermore, we provide some preliminary results that this more memory-intensive strategy can be trained through past experience with high access cost and then used in situations where access cost is minimal. This was the case when only half of the training trials involved a higher access cost. Further research is needed to ascertain how long training effects last and what are the ideal training regimes for different types of task.

In this paper, the influence of Audio Sampling Rate (ASR) and Frame Loss Rate (FLR) on perceived Quality of Experience (QoE) was studied. The result indicated that users are very sensitive to the damaged auditory quality caused by frame loss at 8 kHz and 12 kHz no matter how much it losses. The perceived damage of auditory quality caused by frame loss at 16 kHz and 24 kHz is also much lower that at 8 kHz and 12 kHz. Users even failed to perceive the negative impact of frame loss on auditory quality at 32 kHz whatever the frame loss rate is. The interaction effect indicates that users are not so sensitive to the negative impact of frame loss when the sampling rates increase to 16 kHz or higher.

The goal of this research is to define the dimensions of User Exprience (UX) with interactive products and systems in order to lay the ground for the construction of a subjective assessment tool for UX. After defining UX, we describe several characteristics of UX and present key elements of some UX models in order to understand the ins and outs and the process of UX. Then we present the results of two empirical studies wherein 77 persons were asked to tell UX stories with products. From their stories we extracted 12 UX dimensions which can be grouped around two poles : Product and User. Thereafter we present the underpinning model and an outline of a new UX subjective assessment tool based on the assessment model of NASA-TLX, a well-known tool for assessing mental workload. As conclusion, we indicate the next steps of the construction and validation of the new tool.

The unreasonable design of interface information has given rise to malfunctions of cognition and decision-making among operators, thus leading users into a complex cognition and finally resulting in serious failures in information recognition and analysis, and even in operation and execution processes, which poses one of the major causes for many accidents. Firstly, there remains an internal relevance between errors and perception, and five error factors i.e., visual confined, visual interference, visual illusion, attention shift and over attention were extracted from the point of visual attention mechanism; Secondly, the cognitive model (theory) and psychological experimental paradigm corresponding to the cognitive level were combed out through explanation of the error level, thus the misperception model was established; Finally, It provides a feasible basis for design improvement of visual interface through behavior and physiological experimental data. This misperception analysis method of visual interface has applied mature psychological experimental paradigm and can favorably analyze the design factors from the aspect of misperception, so as to play a significant role in improving the visual interface design.

It is well documented that affective learning materials can impact learning process, but it is unclear that what the role of affective learning-irrelevant stimuli is. To tackle this issue, this study provided evidence that affective value of irrelevant stimuli can be transferred to learning materials and influences the learning process. Using a variant of minimal affective learning paradigm, the experiment demonstrated that only one occasion of neutral-affective pairing can lead to affective learning, and showed an advantage of positive affective learning on the improvement of face memory. Implications for the design of affective human-computer interactive system are discussed.

Measurements of user experience (UX) in traditional human-computer interaction studies mostly rely on task performance and self-report data. Recent research has showed that physiological measures are good indicators of cognitive involvement and emotional arousal and are suggested being used as a complementary measure of UX. This paper reports a preliminary study to examine the possibility of including physiological measures in the UX evaluation process. In the experiment, participants’ physiological responses, task performance and self-report data were collected and analyzed. It was found that physiological measures varied with task performance, as participants showed greater galvanic skin response (GSR) change in the failed tasks than that in the successful tasks. In addition, correlations were found between GSR and self-report data of user experience. The results demonstrated the potential value of physiological measures as a data source of user experience evaluation. However, further investigations involving variations in tasks and individual difference are required.

The purpose of the current study was to analyze the work of imagery analysts associated with Sagebrush, a Synthetic Aperture Radar (SAR) imaging system, using an adapted version of cognitive work analysis (CWA). This was achieved by conducting a work domain analysis (WDA) for the system under consideration. Another purpose of this study was to describe how we adapted the WDA framework to include a

sequential

component and a means to explicitly represent relationships between components. Lastly, we present a

simplified work domain representation

that we have found effective in communicating the importance of analysts’ adaptive strategies to inform the research strategies of computational science researchers who want to develop useful algorithms, but who have little or no familiarity with sensor data analysis work.

Twenty qualified mission-ready F-16 pilots participated in this research. The ages of participants are between 26 and 46 years old (M=33, SD=6); total flying hours between 400 and 3,250 hours (M=1358, SD=882); F-16 type flying hours between 101 and 2,270 hours (M=934, SD=689). Eye movement data were collected by a head-mounted ASL (Applied Science Laboratory) Mobile Eye which was 76 grams in weight, combined with F-16 flight simulator, a dynamic high fidelity trainer that replicates actual aircraft performance, navigation and weapon systems. The scenario is an air-to-surface task. Participants have to intercept the proper route and turn toward the target at an altitude of 500 feet with speed of 500-KIAS, then performing a steep pop-up manoeuver to increase altitude abruptly for appropriate reconnaissance, following by dive and roll-in toward the target to avoid hostile radar lock-on. When approaching the target, subjects have to roll-out, level the aircraft, aiming at the target, release the weapon, and finally pull-up with a 5 ~ 5.5 G-force to break-away from the range. The results show significant differences in pilots’ number of gaze points among five different AOIs, F (4, 95) = 533.84, p<.001,

η

2

ρ

= .97. Further comparisons using post-hoc Bonferroni adjusted tests showed HUD has a significantly higher numbers of gaze points than ICP, DED, RMFD and LMFD; and ICP has significantly higher gaze points than DED, RMFD and LMFD. Also, there were significant differences in pilots’ number of fixation among five different AOIs, F (4, 95) = 306.98, p<.001,

η

2

ρ

=.94. Further comparisons using post-hoc Bonferroni adjusted tests showed HUD has significantly higher number of fixation than ICP, DED, RMFD and LMFD; and ICP has significantly higher number of fixations than DED, RMFD and LMFD. Pilots have to be able to ‘see and process’ the information to understand the situation, and then, to ‘project’ the situation in the near future. There is a long-standing argument concerning bottom-up or top-down visual processes in the eye movement literature. It is observed in this research that pilots applied both bottom-up and top-down visual processes, depending on the salience of information or previous experience.

To establish the civil aviation student pilot psychological quality evaluation model, investigation was carried out about the psychological quality indicator system. This paper analyzed and summarized the relevant research literature, extracting the psychological quality indicators which effect flight training performance. 20 experts identified 4 categories of 21 psychological quality evaluation indicators, and established the civil aviation student pilot psychological quality evaluation indicators system. Using Delphi’s analysis, the weight of each indicator was determined. The flight psychological quality evaluation model was constructed using fuzzy comprehensive evaluation method and the corresponding internet -based questionnaires were sent out to 100 student pilots. The evaluation results were compared with their flight training results, verifying the correctness and validity of the model indicators. The evaluation model can play an important role in psychological selection and psychological training for student pilots, which could further reduce the grounded rate and avoid unnecessary losses.

The performance of pilots are related to the physiological response. In this study, we investigated the correlation between the pilot performance and physiological meters through recording the multiple physiological parameters in a flight. The results showed that in the flight phase from cutting off auto-pilot and auto-throttle, the performance of the subjects correlated strongly with heart rate and also correlated with fixation time or blink interval while the correlation between the performance and the saccade frequency or blink duration was weaker. When the subjects adjusted the flaps or controlled landing gear, the correlation between the performance and multiple physiological parameters, specially the cardiovascular parameters and fixation time, were more stronger compared with that at the other time.

The EU Vision 2020 sets a goal of reducing the air travel accident rate by 80%. Achieving this vision requires innovation and a different approach. PROSPERO (Proactive Safety Performance for Operations) is an EU FP7 project that will provide an advanced systemic methodology for managing the improvement process to help achieve that goal, as well as delivering a way of measuring progress. The overall objectives of PROSPERO are to; (i) Provide a proactive anticipation of complex system risks; (ii) Ensure more effective management of and enhanced learning from situations where risks cannot be designed out of the operation and (iii) Achieve substantial improvement in the elimination of and recovery from human error. This paper reports on the overall PROSPERO concept and high level system requirements as they emerged from the first research phase that focused on identifying industry needs.

This work describes the participatory design and development of a Virtual Camera (VC) system to improve astronaut and mission operations exploration efficiency and cooperation while exploring in deep space. Advanced interaction media capabilities can improve exploration efficiency and cooperation as the distribution of human space exploration roles change in deep space exploration. This capability was developed in a tablet-based application that was evaluated in the field. The VC can minimize the risk of astronauts exploring unknown reaches of the solar system with limited previous knowledge of the area under exploration. Ground-based expert knowledge can be captured and be easily assessable in the remote deep space environment with the VC. The human-centered method of development and testing is described as well as results.

Fatigue has been a threat to flight safety. Based on former researches about pilots’ fatigue, fatigue risk and prevention measures of fatigue, a flight fatigue dynamic risk index is developed to evaluate the fatigue risk during flight operations. Flight fatigue dynamic risk index is defined as the ratio of required and human alertness during flight operations. The required alertness is the function associated with weather conditions, airport conditions, phase of flight, etc., while pilot alertness is calculated by an alertness prediction model which is based on the concept of alertness energy and constructed by human circadian oscillator, alertness energy consumption and restoration. The flight fatigue dynamic risk index prediction model should improve fatigue risk management system(FRMS).

Recently, the global aviation industry has started to promote Safety Management System (SMS), and thus consider the enhancement of safety culture as an essential. To meet the growing demand of safety culture assessment, we cooperated with China Airlines (CAL) in Taiwan to develop a safety culture questionnaire using both qualitative and quantitative methods to meet the needs of their operation environment and to take the theoretical researches of safety culture into account. During the development process, we continuously integrated the opinions from the industry experts, and effectively established a safety culture assessment tool that not only conforms to the actual operation of the airline but also to the requirement of reliability and validity. Although the statistical results show both the strengths and weaknesses of CAL’s safety culture, the hidden reasons of the shortcomings or unstated ideas of workers are still remained undiscovered in this phase. In the following, we would use the focus group method to help examining the potential causes of those low score facets of safety culture, and to develop follow-up recommendations to improve CAL’s safety performance accordingly.

Hard landing is one kind of typical landing incidents that can cause passenger discomfort, aircraft damage and even loss of life. This paper aimed to find out flight performance and operation features of hard landing incidents by using the methods of variance analysis, regression modeling and flare operation analysis based on flight QAR data. Results showed that pilots need to control the aircraft to an appropriate groundspeed and descent rate before descending to the flare initial point. Then control column and throttle operation in flare maneuver would affect landing performance conjointly. The logistic model showed that the vertical load of touching ground was actually linked with touchdown attitude and configuration closely, including three variables of pitch angle, roll angle and flap degree. These findings were expected to be applied in practice to prevent hard landing incidents and even landing accidents.

Radar situation interface belongs to a sub-interface of a complex system. Because the information in human-computer interaction interface of a complex system is of a large amount and in complicated relationships, it is apt to cause misreading, misjudgment and information omission in the target search. The critical factors causing error problems like information omission and misjudgment in the radar situation interface are analyzed. Based on the behavioral data and the physiological data derived from eye movement tracking, the misperception factors leading to users’ information omission/misjudgment are detected. The experimental results showed that, (1) Both interval size and vision position impose a significant influence on the visual cognition of target search. The interval should not be too large for target search in the situation interface, otherwise it may result in long reaction time and omission and misjudgment. (2) During the target search in the upper vision, lower vision and peripheral vision, the reaction time and the error rate present significant changes, and the reaction time of peripheral vision achieves the longest. The vision position also exerts a remarkable influence on the first saccade latency. The fixation duration and fixation point number display obvious changes, and the mean fixation duration of the lower vision is the longest while its fixation point number is the smallest, which is apt to cause misjudgment and omission of information. (3) Eye movement plots can effectively reflect the process of information search, and the gaze plot and the heat point map can present the relevant factors of information omission. And the conclusion reached can be used as reference for the information design and layout of the situation interface of future complex system, so as to effectively improve the misperception problems like omission and misjudgment in the target search process.

To probe pilot’s attention allocation, workload and cognition by eye movement indexes analysis. Six subjects participated the experiment. They were asked to fly three simulation scenarios: landing, climbing and cruise flight. Five eye movement indexes which they were the percentage of fixation point, the percentage of dwell time, average fixation duration, average pupil size and average saccade amplitude were recorded and analyzed. The result indicated that eye movement data was obviously different in out view and instrument panel; also it was different among three tasks. Conclusions can be made from the result: subjects spent most time on outside view while leaving small time to make quick crosschecking to the instrument; Subjects show different pattern of attention allocation through three flight tasks; the recorded eye movement indexes are the good indicators to pilots’ attention allocation, workload and cognition.

Human factors have been the main reason of flight accidents, in which misoperation of pilots plays an important role. According to Statistics, in the field of accidents caused by human error accounted for about 80%. The layout design of aircraft cockpit and different controller locations result in different situations of muscle fatigue. Since the situation of muscle fatigue has an effect on the response accuracy, the probability of misoperations increases. The situation of muscle fatigue can be reflected by variables of the sEMG signal. This paper aims to verify the relationship between muscle fatigue and response accuracy based on analysis of sEMG signal. In the experiment, we investigated changes of response accuracy using joystick controller to trace the static object and dynamic object, along with the change of the situations of muscle fatigue. The terminal experiment result can provide information and design method of flight deck to make flight deck safer and more comfortable.

In aviation, the Standard Operating procedures (SOPs) provides typically a list of action items that allowing the pilots to complete tasks in flight environment. Therefore, the complexity of SOPs should be appropriate to guarantee the flight safety. In this paper, step complexity (SC) from nuclear power plant is introduced to evaluate complexity in flight in nine tasks selected from SOPs. The verification measurement of SC is difference of heart rate (HR-D) of pilots. From experiment result, SC is correlative to HR-D. However, the correlation is not significant enough. Thus, to evaluate complexity in flight efficiently, the SC measure should be modified.

Introducing new train traffic management technologies can activate undesirable changes to operational safety. Therefore, it can be useful to understand how to expedite technology adoption in control rooms to strengthen the collaborative efforts of the human-automation team in support of resilient processes. This exploratory study presents factors that impact technology adoption. Results revealed that end-user buy-in was considered critical. Participants revealed that buy-in can be undermined when end-user expertise is not or under utilised and when horizontal communication channels are restricted. Technology issues arise when end-user work needs are not supported and when insufficient time, training or support slow adoption processes. Finally, organizational factors included: weak commitment and leadership to resource and drive project processes and dishonesty and lack of open accountability. Finally, stakeholders recognized that new projects are frequently managed from the top-down and that contributions from the bottom-up can add significant advantages toward expediting system changeovers.

This paper addresses the challenges of high system complexity within rail traffic control. Based on resilience engineering principles, the different types of traffic control technology are analysed in order to identify either their contributions or hindering factors towards system resilience. Throughout four main generations of technology in traffic control, whilst there is a clear path towards increased automation, evidence from recent research in this domain suggests that the introduction of automation does not necessarily contribute to enhanced resilience. Despite its contributions to efficiency by placing larger areas under the supervision of each control post, it has introduced many new complexities in traffic control decision making. In many cases, automation has created a gap between rail operations and those in charge of their control. Beyond basing their decisions on operational needs and priorities, Traffic Controllers must take into account the possible responses that automated systems might initiate. So far, traffic control technologies are unable to deal with disruptions and much of the variability inherent to complex operations such as the railway but future generations of rail signalling systems may be able to better support resilience if appropriately designed.

Six-dimensional tracking and control task within an Integrated Cognitive Architecture, as a makeup for automated Six-dimensional tracking and control task default. is a common yet highly complex space operation, challenging the human workload. For space exploration system safety, workload is a critical factor in task design and implementation. This research integrates two cognitive architectures: Queuing Network (QN) & Adaptive Control of Thought-Rational (ACT-R) to develop a rigorous computational model for Six-dimensional tracking and control task cognition process. ACT-R represents the human mind as a production rule system. Experiments are set up to build Six-dimensional tracking and control task cognition model and afterwards to validate feasibility of the proposed integrated cognition architecture. Ten subjects of similar training level are chosen to finish manual Six-dimensional tracking and control task with three task difficulty level: one only with displacement margin, one only with posture margin and one with displacement and posture margin. Cognition task analysis is firstly conducted on task performance of subjects. Cognition model of manual Six-dimensional tracking and control task is then built up based on the proposed integration architecture. The proposed integration model developed in the ACTR-QN describes component processes of tracking, decision making and controlling in a 3D environment by ACT-R production rules within QN network. Workload index for each cognition module is calculated based on sector utility throughout the whole task. Human results are compared with the modeled results in the dimension of task time and displacement/posture control trajectory deviation. Workload index is calculated based on the percentage of each module in the time dimension.

This paper presents a methodology for measuring Crew Resource Management (CRM) parameters as applied to a pilot decision-making task. Six teams of pilots took part in a desk-top decision-making exercise. Flight crew performance was observed by human factors researchers and was measured on a number of parameters pertaining to communication, situational awareness, decision-making, mission analysis, leadership, adaptability and assertiveness. This methodology facilitated the mapping of decisions in the context of the overall process. The communication analysis can be considered more objective than standard CRM expert rating. This methodology could be used to examine CRM for training, recruitment, incident and accident analysis, identifying degraded performance on the flight-deck and has further implications for multi-team co-ordination. It could also be used to provide a sound contribution to the design of automatic means of detection for CRM metrics on the flight deck.

The purpose of this study is to suggest a framework to assess diagnosis error of operators in nuclear power plants. In nuclear power plants, human error caused by inappropriate performance due to inadequate diagnosis of situation by operators have been considered to be critical since it may lead serious problems. In order to identify and estimate the human errors, various human error analysis methods were developed so far. Most human error analysis methods estimate diagnosis error through time reliability curve or expert judgments. In this study, a new framework to assess diagnosis error was suggested. It is assumed that diagnosis error is caused by inadequate quality of data and diagnosis error can be observed by using information processing model of human operators. Based on this assumption, we derived the assessment items for the quality of data and diagnosis error taxonomy here.

Foraging tasks, such as search and rescue or reconnaissance, in which UVs are either relatively sparse and unlikely to interfere with one another or employ automated path planning, form a broad class of applications in which multiple robots can be controlled sequentially in a round-robin fashion. Such human-robot systems can be described as a queuing system in which the human acts as a server while robots presenting requests for service are the jobs. The possibility of improving system performance through well-known scheduling techniques is an immediate consequence. Unfortunately, real human-multirobot systems are more complex often requiring operator monitoring and other ancillary tasks. Improving performance through scheduling (jobs) under these conditions requires minimizing the effort expended monitoring and directing the operator’s attention to the robot offering the most gain. Two experiments investigating scheduling interventions are described. The first compared a system in which all anomalous robots were alarmed (Open-queue), one in which alarms were presented singly in the order in which they arrived (FIFO) and a Control condition without alarms. The second experiment employed failures of varying difficulty supporting an optimal shortest job first (SJF) policy. SJF, FIFO, and Open-queue conditions were compared. In both experiments performance in directed attention conditions was poorer than predicted. A possible explanation based on effects of volition in task switching is proposed.

Railway traffic control faces the challenge of ensuring a high infrastructure capacity to maintain a constant train traffic flow. The current study assesses the situation awareness (SA), as a predictor of decision-making, of train traffic controllers to gain novel insights in their cognition. This study puts emphasis on levels of implicit and explicit situation awareness in a monitoring mode, through measures of SAGAT, MARS and performance. A human-in-the-loop simulator, called the PRL game is used to simulate the workspace of train traffic controllers. Initial findings indicate rather low levels of explicit SA, on the contrary to higher subjective SA scores through MARS and observer ratings, and a high performance on the punctuality and unplanned stops of trains.

It has been demonstrated that cell-phone conversations impair driving performance. However, it is unclear whether the difficulty of naturalistic phone conversations can modulate driving performance. The present study used a simulator to investigate whether the cognitive load of phone conversations (non-conversation, multiple choice and open question conversations) influence highway driving performance. The results showed cell phone conversations with open questions led to most aggressive driving with highest speeds and shallowest braking. Furthermore, open question conversations led to the smallest route deviations. These results suggested that a drivers’ capability for monitoring speed and making manoeuvre decreases as the difficulty of a phone conversation increases. The implications of this study for driving safety are discussed.

Auditory warning signals are common features in industrial control rooms. Finding sound signals that convey higher degrees of urgency while keeping the potential for annoyance low is challenging. In the present study, evaluations were performed on four different types of auditory displays. The displays were all designed to convey three levels of urgency. The examination focused on the following questions: (1) “How reliably can the operators identify the three levels of urgency?” and (2) “How annoying do the operators find the sound signals?”. Fourteen operators participated in the study. For every signal within each auditory display, the participants were asked to rate the level of urgency and annoyance. The results show that one can design auditory displays that employ appropriate urgency mapping while the perceived annoyance is kept at a low level. The work also suggests that involving the end users in the design process could be advantageous.

Imagery analysts are given the difficult task of determining, post-hoc, if particular events of importance had occurred, employing Synthetic Aperture Radar (SAR) images, written reports and PowerPoint presentations to make their decision. We were asked to evaluate the current system analysis process and make recommendations for a future temporal geospatial analysis prototype that is envisioned to allow analysts to quickly search for temporal and spatial relationships between image-derived features. As such, we conducted a Hierarchical task analysis (HTA; [3], [6]) to understand the analysts’ tasks and subtasks. We also implemented a timeline analysis and workload assessment [4] to better understand which tasks were the most time-consuming and perceived as the most effortful. Our results gave the team clear recommendations and requirements for a prototype.