Proceedings of the Human Factors and Ergonomics Society Annual Meeting

The Head Mounted Display (HMD) is discussed as a subset of Mixed Reality (MR) displays. A definition of MR is given, in terms of image mixtures along a Reality-Virtuality (RV) Continuum, including the subclasses of augmented reality (AR) and augmented virtuality (AV). In relation to actual task execution, the relative need for local guidance information versus more global planning and navigational information is discussed. A taxonomic framework for classifying MR systems is presented, in terms of not only the RV continuum, but also the degree of centricity of the observer relative to a nominal viewpoint and the extent of control-display congruence. Several practical examples of MR systems are presented, all from the domain of surgery, and for each a volume within the MR design space is proposed.

Eight subjects' abilities to detect changes in system latency during voluntary lateral hand movement of virtual objects were studied in an immersing virtual environment. A two-alternative forced choice procedure was used in which discrimination of latency was studied with respect to three reference latencies: 27, 94, and 194 msec. Results show that subjects are able to reliably detect changes definitely less than 33 msec and probably less than 16.7 msec. Strikingly, for the short latencies we examined, subjects' ability to detect latency changes does not depend upon the base latency we used as a reference. Thus, the discrimination we studied does not appear to follow Weber's law and may provide evidence for quick adaptation to the reference latencies used.

Helmet-mounted displays (HMDs) may become a primary source of head-up information in future tactical aircraft. In evaluations, HMDs have produced improved situation awareness and mission performance. To realize this improved performance, HMD equipped pilots spend significantly more time looking off-axis, away from traditional displays. Accordingly, it is important that HMD information be designed so targeting and ownship status symbology is effective and safe. The U. S. Air Force Research Laboratory is working to develop an optimized HMD symbology set. Toward this end, symbol designs must accommodate the conditions of intended use, technology limitations, and integration with other displays. This paper discusses efforts to design target tracking/location and ownship status symbologies for the day/night all weather fixed-wing tactical aircraft application. Design principles derived from both laboratory and flight test are presented. These principles relate to symbology frame of reference, orientation, compression, and line of sight mechanization. Evaluation methodologies are also discussed.

Over-reliance occurs when the human responsible for completing a task allows a task aid to have too much authority. Over-reliance, long recognized as a problem with automation and decision aiding systems, now appears to also be an issue for task guidance systems. Task guidance systems are computer-based systems that guide a user through a task. A recent investigation of how the capabilities of a wearable computer may be used to provide task guidance in mobile environments demonstrated the possibility for over-reliance in these systems. Specifically, this investigation examined a wearable computer inspection aid for preflight inspection of general aviation aircraft. Pilots using the wearable computer inspection system were less thorough in their inspection than the pilots completing the inspection from memory. Deficiencies in the procedures used on the task guidance systems made it clear that the pilots with the wearable computer inspection aid were over-relying on its information. Details of this investigation and goals for future investigations into an interface design process to mitigate over-reliance in task guidance systems are discussed in this paper.

Does the addition of audio enhance visual perception and performance within a virtual environment? To address this issue we used both a questionnaire and an experimental test of the effect of audio on recall and recognition of visual objects within different rooms of a virtual environment. We tested 60 college-aged students who had normal visual acuity, color vision, and hearing. The between-participants factor was audio condition (none, low fidelity, and high fidelity). The questionnaire results showed that ambient sounds enhanced the sense of presence (or “being there”) and the subjective 3D quality of the visual display, but not the subjective dynamic interaction with the display. We also showed that audio can enhance recall and recognition of visual objects and their spatial locations within the virtual environment. These results have implications for the design and use of virtual environments, where audio sometimes can be used to compensate for the quality of the visual display.

The ability of listeners to detect, identify, and monitor multiple simultaneous speech signals was measured in free field and virtual acoustic environments. Factorial combinations of four variables, including audio condition, spatial condition, the number of speech signals, and the sex of the talker were employed using a within-subjects design. Participants were required to detect the presentation of a critical speech signal among a background of non-signal speech events. Results indicated that spatial separation increased the percentage of correctly identified critical speech signals as the number of competing messages increased. These outcomes are discussed in the context of designing binaural speech displays to enhance speech communication in aviation environments.

The potential of virtual audio display technology to provide operators with veridical spatial cues may be substantially constrained by factors that are common in many operational settings – i.e., high noise level, limitations in the bandwidth of the audio source and/or display. The purpose of this study was to examine the effects of varying the bandwidth of a virtual sound source in the presence of broadband noise in a reverberant environment. Specifically, the signal-to-noise ratio (SNR) was varied from +50 dB to −10 dB, and the signal was low pass filtered at 1.6 kHz, 4 kHz, 8 kHz, and 15 kHz. Correlational analyses comparing actual and perceived sound source location revealed that both signal bandwidth and signal-to-noise ratio influenced auditory localization acuity, and that even under optimal bandwidth and noise conditions (15 kHz and +50 dB) localization in elevation was extremely poor. These findings have numerous implications for the design of spatial audio displays, especially those meant to be used in noisy environments.

Most Virtual Reality applications involve manual manipulation and suffer from lags. Effects of lag in a virtual environment have received much attention. However, effects of target distance and width had not been analyzed in separation. A study has been conducted to investigate the effects of, and interactions among, hand movement related lag, target distance, and target width on manual task performance in a virtual environment. With a constant hand-related lag, hand Movement Time obeys Fitts' law (R²>0.85). Lags of 110 ms or above can significantly increase hand Movement Times and their effects have significant interactions with the effects of target width but not with target distance. This finding indicates that, in the presence of lag, the effects of target distance and width should be analyzed in separation.

An instrumented prop was used to rotate a virtual object in an immersive virtual environment. For each experimental trial, two virtual objects were displayed adjacent to each other. With an emphasis placed on speed and accuracy, participants rotated one object until its orientation matched the orientation of the other object. Participants used an efficient path of rotation more often when the axes of the object, the participant, and the shortest path of rotation were aligned. In relative terms, these paths were more efficient when there was a large angular difference between the objects' orientations. The rate at which participants rotated the virtual object increased substantially with the magnitude of the initial angular difference between the two orientations. By contrast, in most mental rotation studies the rate of rotation remained constant for variations in angular disparity.

Viewing wide field-of-view scene movement with a Virtual Reality (VR) display can cause symptoms of cybersickness (e.g., nausea and headache). It has been known that cybersickness is a type of vection-induced motion sickness. Since vection is associated with the perception of moving scene, two studies have been conducted to investigate the effects of different scene movements on levels of cybersickness. The velocity and complexity of the scene movements were quantified by a previously reported metric called ‘spatial velocity (SV)’ while the levels of cybersickness were measured in terms of nausea ratings and Simulator Sickness Questionnaire (SSQ) scores. Results showed that in both experiments, both the nausea ratings and the SSQ total sickness scores increased significantly with increasing SV in the dominant axes of scene movement. Potential uses of the SV metric for formulating a CyberSickness Dose Value (CSDV) are discussed.

The use of a virtual environments (VE) for training purposes has the potential to create motion sickness-like symptoms in participants; in some, these symptoms are so severe that the user drops out of the VE training. The purpose of this study is to characterize the frequency of VE dropouts as well as the thresholds and associated symptoms experienced by those who cannot complete their VE exposure. When 245 students interacted with a VE for varying amounts of time performing a variety of tasks, 19% consistently could not complete their assigned VE exposure, even though all participants were exposed to the same virtual world and completed the same tasks. Dropouts experienced symptoms such as nausea, oculomotor disturbances, and disorientation. Insights into the potential differences between dropouts and finishers emerge via an examination of their interaction with the virtual world and suggest that controlling, streamlining, and limiting VE interaction may reduce the number of dropouts.

We conducted a study to examine the effects of target cueing and conformality with a hand-held or head-mounted display to determine their effects on visual search tasks requiring focused and divided attention. Eleven military subjects were asked to detect, identify, and give azimuth information for targets hidden in terrain presented in a simulated far domain environment while performing a monitoring task in the near domain using either a helmet-mounted display or hand-held display. The results showed that the presence of cueing aided the target detection task for expected targets but drew attention away from the presence of unexpected targets in the environment. This effect was mediated by the display used, such that attentional tunneling was reduced when subjects were using the hand-held display.

Haptic interfaces are becoming more widely used in virtual reality simulations because of their enhancement to performance in training and simulation. Unintended vibration degrades the effectiveness of a haptic device and can reduce the user's ability to detect small details in the surface of simulated hard objects. Two important design factors for controlling haptic device vibrations are servo update rate and resistive force magnitude. This work employs a signal detection and receiver operator characteristic methodology to evaluate the interaction of these two factors on a user's ability to perceive small details in a hard surface. In a two-factor, full factorial experiment, six male research participants rated their confidence in detecting a 5mm upward step along a simulated hard surface. The independent variables were resistive force magnitude (1.63N or 4.88N) and update rate (468Hz or 1630Hz). The results indicate a significant interaction of the two variables. Participants demonstrated a 69% success rate with the strong force, slow update rate compared with a 86% success rate in the other 3 conditions. The participants may have employed a strategy of detecting the lack of oscillations when entering the surface and then noted their kinesthetic motion while moving back to the oscillating state experience when sliding along the surface of the simulated edge. Either increasing the update rate or reducing the resistive force magnitude can mitigate the effects of vibration when using a haptic device.

Haptic Displays present information to the sensory modality of touch. Tele-robotic tasks that are usually conducted in the absence of information that we gather by the sense of touch will benefit greatly by the inclusion of such data. The limited size of haptic displays and the serial nature of the information processing make it difficult to present many kinds of haptic information simultaneously. Display designers will benefit from knowing which types of haptic information (information usually gathered by touch) can be presented using cross modal displays (displays that use modalities other than touch) without causing reduced task performance. This study investigates the efficacy of cross modal displays of texture information. Three displays were designed in each of the following categories: Visual - Pictorial, Visual - Graphical and Auditory. Three control displays - Touch Only, Sight Only and Touch and Sight were also used. Data from displays with matched sensitivity values were used to analyze the effect of modality on three representative tasks: search, averaging and optimization. Results indicate that low cost, low bandwidth cross modal displays are viable alternatives for tasks with an embedded texture perception component.

Retention of spatial knowledge was compared after acquisition using virtual reality (VR), a map, and a real building. Participants were trained to navigate a route and to note the locations of landmarks. Two weeks later they were tested on their ability to navigate the route and to point to the landmarks they had learned. They were tested in either an aligned or contra-aligned orientation. The orientation specificity for VR observed immediately after training in our previous experiment (Clawson, Miller, Knott, & Sebrechts, 1998) was magnified after retention. On the route test, the VR group performed better than the map group, but only when tested in the aligned orientation. On the survey test, the VR group was less accurate than the map or real building groups, especially in the contra-aligned condition. Potential cognitive and perceptual causes of this orientation specificity are discussed.

This study examines potential mappings between the subjective experience of speed when mediated through visual display devices and the physical velocity of a vehicle, especially when the apparatus for experiencing speed and the actual vehicle are not in the same scale. Such mappings are important for creating psychological reality in telepresence environments, such as a driving simulator. Subjects made magnitude estimations of perceived speed by watching video clips of forward motion taken from vehicles of two different heights. A 2.625:1 ratio in “virtual” eyeheights was approximately maintained in the magnitude estimates, indicating that speed is experienced as a scaled quantity relative to virtual eyeheight. When magnitude estimates for the lower virtual eyeheight were scaled, perceived speed at that height could be predicted for any actual velocity in this experiment. Building on this knowledge, telepresence designers must account for the scale factor in predicting other behaviors such as distance for braking and turning the vehicle.

Virtual reality (VR) applications that enhance user presence have been hypothesized to be effective training tools for promoting virtual task performance and transfer of skill to reality in operation of complex systems. Research has focused on use of VR for training navigation skills, but has not described how systems should be configured to increase presence and provide for optimum learning. We present a study comparing the effect of display types, as well as navigational aids, on performance, presence, and workload in training navigation of a telerobotic vehicle through a virtual maze. The relationships among performance, presence, workload, and human factors, such as spatial ability and immersive tendencies, were also examined. The findings of this research may be used to formulate VR application design guidelines for enhancing user presence in virtual environments and, possibly, promoting training of real skills.

Objective: To determine whether egocentric distance judgments are accurate in a virtual environment using a perceptual matching task. Methods: Observers were immersed within a virtual environment consisting of an L-shaped room with a test object (a column) located down one corridor and a comparison object (a flagpole) located down the other. Perceived distances were measured by having the observers move the comparison object to match the distance to the test object. Textures in each corridor were independently varied, as was the size of the test object. Results: Significant effects were found for the texture pattern under the test object and for distance, but not for the texture under the comparison object or column size. Performance was best with a regular (brick) pattern. Overall, observers' judgments overestimated distance to the test object by approximately 7%. Conclusions: Egocentric distance judgments were relatively accurate compared to past research. Texture affected performance, with a regular pattern producing highest accuracy.