The primary objective of this research is to evaluate the effectiveness of dynamic information and warning systems. Examples of such systems include Dynamic Message Signs (DMS) providing real time traveler information. A full-scale, fixed-base, fully interactive dynamic driving simulator is used to study the effect of the content and format of a message on drivers, identify the content and format that maximize comprehension for older adults, and determine the content and format that produce the ideal distribution of route choices. Central to this research will be an evaluation of the effects of such dynamic systems on elderly drivers. For comparison purposes the research will also examine the effects of these systems on other drivers.

Experimental Design. The factors that are varied in the experiment are the familiarity of the drivers with the alternative route and the information presented on the DMS. In another condition, the drivers were provided with travel time information before they enter the highway. We assume that all drivers are familiar with the main route and will tell them beforehand that it takes one hour to travel.

An underlying aim of the research is to determine the extent to which these systems tend to contribute to improvements in safety and mobility for all transportation user groups and in particular older drivers.

Objective: Driving simulation has been offered as a method of evaluating the functional design, ergonomics, cognitive demands and safety of in-vehicle information systems (IVIS), which are rapidly moving from novelties to a common feature of both consumer and fleet vehicles. However, relatively little research has addressed whether simulators provide accurate models of behavior in the real world. This study assessed the validity of driving simulation for evaluating different IVIS interface designs.

Methods: A surrogate destination entry task was developed which required drivers to enter the first two letters of each address component (state, city, street). Three interfaces were evaluated: two paralleling methodologies used by various automobile manufacturers (touch screen and distributed rotational controller) and a third based upon the cellular phone keypad. Results from a simulator and an on-road study were compared to assess the validity of measures including visual attention, driving performance, and task performance. Two independent sets of participants between the ages of 22 and 28 were tested, 30 in the simulator and 28 in the field.

Results: Most visual attention measures and destination entry task performance indices, such as glance frequency, total glance duration, and mean response time, mapped almost identically from simulation to field. Driving performance, including standard deviation of velocity and lateral offset, were less consistent and less sensitive at discriminating between devices.

Conclusions: Medium fidelity, fixed based driving simulation appears to be a safe and effective method of evaluating task performance and modeling visual interaction with in-vehicle interfaces that occur in the field.

With the rapid increase of devices used by drivers and associated concerns about safety, it is of utmost importance to properly identify their effects on road safety. A growing body of research suggests that distraction at the wheel does not only increase the chance of being involved in a crashes but could also affect traffic flow. Moreover, with the aging of the population, there is a need to understand more broadly the impact of age on cognitive demand or distraction regarding driving behavior.

This study evaluated the effect of age and the addition of a secondary cognitive task while driving on the number of lane changes executed, lane choice, and the time spent in the leftmost lane. Three age groups (20’s, 40’s and 60’s) were monitored in an instrumented car (Aware Car) while performing a cognitive task. Video records of each subject were reviewed in order to identify the onset and the number of lane changes and the time spent in the left lane.

During baseline driving, both group showed similar number of lane changes. During the task, older drivers showed a reduction in lane changes compared to the two other groups. They also spent less time in the left lane during baseline and while involved in a cognitive task.

More studies need to be conducted in order to evaluate if these reduction in lane change showed a conservative approach to driving under dual task or a saturation of the driver cognitive capacities. Important knowledge concerning traffic flow and safety could emerge from these data.

Operator interaction with complex systems requires perceiving information from the system, decision making based on that information, and the selection and execution of appropriate action. In many applications it is important to make an accurate and timely decision. We aim to develop and implement a mathematical model of human perception of nonstandard system behavior, as well as human decision making about whether that behavior constitutes a system failure due to bias/noise increase, omission of information, or sensor degradation. This model will be integrated with an existing system design tool to simulate total system performance, incorporating a variety of failure modes for each component. The integration of human performance models (HPMs) is the beginning of a series of work to create an early-stage design tool for understanding human interaction with complex systems. This approach will allow designers to analyze the anticipated closed-loop effects of human decisions and actions on total system performance, and make informed design changes or recommendations to proposed systems. The selected case study is an integrated clinical environment manager, where individuals are monitoring patient health and status. However, this modeling approach will be robust and extensible to a variety of possible domains. Extensive simulation and sensitivity analysis will be conducted to characterize and identify future human subject experimentation for model parameter verification and validation. Future HPMs to be included will continue building on the capability to model human decision making, operator response selection and execution, and incorporate an estimation of operator context.

Cellular phones capable of transferring data have become a common platform to perform daily tasks for a wide variety of today’s work force. While designing effective user-interfaces for pedestrian use has been widely explored, a different challenge is posed when dealing with mounted police patrols. This study will focus on determining what makes a user-interface suitable for mounted police patrols using personal data assistants (PDAs) capable of performing remote database queries via cellular data transmissions.

Our approach is to deploy PDAs to a New Hampshire local police department. The police officers will use the Project54 software to query driver license and motor vehicle registration databases. User performance will be analyzed using metrics such as time to complete record checks, number of interactions with the device, and time spent looking at the device. In conjunction with subjective user opinions, we hope to use this data to determine the different parameters that make a mobile application’s user-interface suitable for mounted police patrols.

Statistics show that over 10,300 elderly deaths resulted from falls in the year 2000 which cost around $179 million in incidence and medical costs (Stevens et al, 2006). One of the quality improvement efforts used to reduce falls is filing an incident report to identify the causes and take preventive measures against future falls. The process of filing an incident report is often by paper-based form. Paper-based forms require a great deal of manual data entry and as the form exchanges many hands throughout the organization, it can potentially be lost. As health information technology (HIT) progresses, many hospitals are beginning to replace paper-based forms with electronic systems. However, for small scale health institutions such as nursing homes, implementing HIT is expensive and may not fit the nurses and their workflow.

In my research, I designed an easy-to-use electronic falls reporting system using low-cost applications. This research aims to evaluate the system through four data collection approaches: return on investment, cognitive task analysis, technology adoption, and a holistic human factors evaluation. The evaluation is based on how well each data collection method evaluates the system in terms of the acceptability of the system for each of the stakeholders at Jewish Geriatric Services. Through the data collection methods, factors that contribute to user acceptance and physical, cognitive, and macroergonomic challenges that the user encountered would be identified. Then, redesign of the system is possible to reduce these challenges and enhance the fit of the technology with the user and user’s workflow.

In the United States, 78% of the pedestrian cashes occur at non-intersection crossings, making mid-block crosswalks a prime target for remediation. Almost two-thirds of pedestrian fatalities occur in urban areas and the percentage is higher for crashes in rural areas. Pedestrian collisions at unmarked crosswalks are more likely than at unmarked crosswalks when the locations are uncontrolled. This project will study the effects of alternative advance yield markings and signs on both drivers’ and pedestrian crosswalks, especially in those multi-threat situations where the driver’s or pedestrian’s view of critical information is obstructed by one or more vehicles. This critical situation is often what creates the most danger for pedestrian in the crosswalk.

The objectives of this research will be achieved through four specific research experiments: Using the driving simulator, we will measure potential vehicle-pedestrian conflicts as well as pedestrian-vehicle conflicts. These measures will be obtained by monitoring the driver’s point of gaze to determine if the drivers approaching an unsignalized mid-block crosswalk look for pedestrian more frequently in high risk scenarios when advance yield markings and associated signs are present than they do when just the crosswalk is marked. Field studies outside and inside the vehicle will also be conducted to determine whether drivers on the road approaching signalized intersection yield more often; also using head mounted eye tracker to monitor drivers’ head and eye movement we will determine whether the driver gaze at potential pedestrian threats sooner, on average, when the advance yield markings are present.

Given the level of convenience that personal navigation devices offer, it is obvious why they became so proliferated in vehicles nowadays. However, the problem is that most of these devices depend on an in-car screen which displays maps and navigation instructions and therefore requires drivers to take their eyes off the road. This reduces visual attention and may lead to accidents.

One way of solving this problem would be to use a head-up display (HUD) augmented reality (AR) navigation aid, which seamlessly integrates navigation instructions on the windshield. Since this solution does not require drivers to take their eyes off the forward roadway, we hypothesize that HUD AR may offer better driving performance and visual attention in comparison to standard navigation aids. One presumably simpler way of tackling the same problem would be to use a head-down display (HDD) AR navigation aid, which would display a live video feed of the road ahead with overlayed navigation instructions. Since this solution also requires looking at the in-car screen, we hypothesize that it should not offer any advantages over the standard navigation aid.

As a first step in this research, our goal was to compare driving performance and visual attention in a simulator-based pilot study between HUD AR and standard navigation aids. Our preliminary results suggest that there is no difference in driving performance, however there is a significant difference in visual attention, favoring HUD AR. Further study will explore HDD AR and possible drawbacks of the HUD AR navigation aid.

In 2008, the World Health Organization (WHO) published a surgical checklist as part of their ‘Safe Surgery Saves Lives’ patient safety campaign to improve surgical care. The WHO surgical safety checklist has been shown to reduce surgical complications and increase adherence to standards of care at eight pilot sites distributed across the globe. Since its publication, hospitals worldwide have worked to integrate the surgical checklist into existing workflow. The research presented here attempts to uncover the barriers to implementing the surgical checklist in an academic healthcare institution as this integration process unfolds. This discovery can be approached by considering the operating room as a joint cognitive system in which the checklist process and the practitioners involved are directly affected by the larger organizational policies and procedures as well as resources and constraints. In approaching the research from this framework we can identify solutions to overcome some of these barriers to checklist implementation and also reveal the demands of practice in the operating room which are inherent to the field and telling of how the system functions.

A future concept of operations for controlling unmanned vehicles has a single, forward-deployed soldier supervising multiple, heterogeneous unmanned vehicles. This operator would collaborate with an automated planner that presents the operator with a choice of potential plans for how the unmanned vehicles could efficiently search the dynamic environment for new targets, track identified targets, and destroy hostile targets. A key step towards this future concept is to determine how often the automated planner should provide the operator with potential plans without causing a detrimentally high workload for the operator, which could have negative performance consequences. This research describes the Onboard Planning System for Unmanned Vehicles Supporting Expeditionary Reconnaissance and Surveillance (OPS-USERS) testbed and an experiment to measure operator workload and performance in a set of scenarios with different replanning intervals. The results show that the replan interval has a significant effect on workload and performance with significantly lower performance at the lowest replan interval. Replanning too frequently in a human-on-the-loop collaboration can cause higher levels of workload and lower performance. Further research will determine the impact of long duration, low workload missions with a highly autonomous system on operator performance. Also, collaboration through operator manipulation of the objective function of an automated planner deserves further research. Finally, a simulation of human-in-the-loop control will validate the merit of the human supervisory control paradigm in OPS-USERS.

While Consumer Health Informatics (CHI) Interventions aim to facilitate communication between patients and care providers, these systems are often of a poor design, increase the chance of user error and amount of work, and end in frustration by patients and care providers. Our research team’s goal is aimed towards pointing out and resolving these areas of concern in the devices and applications.

We developed scenarios involving patient and care provider usage of existing CHI devices and applications – specifically diabetics with high blood pressure and post-bariatric surgery patients. We then conducted in-depth 10-day trials based on the developed scenarios, using multiple health care applications and devices, and taking note of every challenge encountered under three main categories: macroergonomic, physical, and cognitive. Our group is now identifying mediation strategies for challenges found when using the CHI interventions. The scenario-based testing results will be compiled to inform health care technology specialists of the weaknesses and strengths of the CHI devices and systems. These assessments can be presented to decision makers, pointing out the areas they should invest the most time in improving the intervention. We will also begin to mediate these challenges by creating supplementary handbooks to the intervention devices and applications.

The concept of optic flow, as a person or vehicle moves through an environment, was first introduced by Gibson (1979). Optic flow can be measured as a function of both driving speed and the driver’s direction when one is driving. Lee (1980) found that optic flow is useful to drivers when driving on curves and avoiding collisions with other objects. The amount of optic flow is found to be greatest directly in front and to the sides of the vehicle, and decreases to zero at the point of expansion.

The optic flow presented in a driving simulator display does not correspond with optic flow found in real-world driving. Many studies have been conducted to monitor realistic optic flow in a driving simulator and few of them can do an adequate job of providing a 180 degree or larger field-of-view (FOV). Asano and Uchida (2005) have demonstrated a solution to simulating real optic flow for curves by using a vehicle turning driving simulator (VTDS).

The application of optic flow in driving simulator includes the problem of inferring not only the motion of the driver and objects in the scene, but also the structure of objects and the three-dimensional environment. It has been concluded that optic flow can be utilized in a number of driving tasks: (1) to obtain heading information when driving on straight and curved roads; (2) to steer around parked vehicles, roadside trees, etc; and (3) in braking and/or steering to avoid collisions when approaching other moving vehicles. 

References:
Gibson, J.J. 1979. The Ecological Approach to Visual Perception. Houghton Mifflin: Boston, MA.
Lee, D.N., 1980. The optic flow field: the foundation of vision. Phil. Trans. Royal Society London B 290, 169-179.
Asano, Y. & Uchida, N. 2005. Improvement of Driver’s Feeling by Turning Cabin Driving Simulator. Proceeding of the Simulator Conference – North America, Orlando, Florida, 230-239. Driving Simulator

Younger drivers are at a greater risk of getting into fatal crashes than more experienced ones. One reason for this, according to research, is that inexperienced drivers often fail to identify hazardous situations on the road, making them more susceptible to a crash. The Human Performance Laboratory at the University of Massachusetts has developed a PC platform Risk Awareness and Perception Training program (RAPT) to train novice drivers to scan roadways for potential hazards. With this study, using a driving simulator, we want to evaluate the effect of the RAPT program on scene scanning patterns of younger drivers.

Two groups of participants will be used in this study. One group will be trained using a simulator-based version of RAPT and the other, the “placebo” group, will read a driving manual instead of going through the training/receiving the training. Following training, participants will be evaluated on a second driving simulator. During evaluation, participants will be tested on a variety of near and far training transfer scenarios, some of which will closely resemble the training scenarios, while others will be unfamiliar. During the experimental drives participant’s eye movements will be recorded.

We expect to see a significant difference in the number of Participant gazes in the potential risky areas of the road between the two groups.

In storytelling, a picture is worth a thousand words. The way in which digital pictures are used to tell a story greatly impacts their effectiveness. We hypothesize that storytelling with digital photos can be enhanced by combining three commercially available technologies: virtual globes, geotags, and multitouch displays. Virtual globes add visual geographical information to the story, making it more immersive. Geotags are combined with compass orientation to further increase the immersiveness of the story, by placing the photos at the location they were taken, in the direction the storyteller took the picture. Multitouch displays allow for the natural manipulation of the photos and globe.

Our prototype system uses photos that are captured on a GPS enabled mobile phone. In order to tell the story, the photos are displayed using the Google Earth application, which communicates with the multitouch table. In pilot studies, we compared our application to the traditional slideshow storytelling method. After analyzing the data logged from the manipulation of the table by the storyteller as well as the audience questionnaires, early results show a preference for the Google Earth application over the slideshow method. We plan to explore the advantages and limitations of the proposed system using online videos and questionnaires as well as in lab human subject experiments.

Objective: The study was conducted to assess the physical/ergonomic exposures that may lead to musculoskeletal injuries of piledrivers.

Methods: A hierarchical taxonomy for piledriving work was developed with tasks and activities defined within each of eight main pile driving operations. Exposures were characterized for the piledriving work with the PATH (Posture, Activity, Tools, and Handling) work-sampling observation method. Data on working posture were collected for three main body parts: legs, arm and trunk. Data on activity performed, tools used and loads handled were also collected for each observation.

Results: A total of 8301 observations were made on 29 piledrivers, on a total number of 6 work sites. The lagging operation had the highest percentage of observations with non-neutral trunk (46.8%), and leg (41.0%) postures, as well as one of the lowest percentages for working on stable ground (9.0%) as observed during the lagging operation. The bracing operation had the lowest percentage for working on stable ground (0.3%). The slurry wall operation also had a low percentage of work on stable ground (6.0%). The arm postures were less frequently non-neutral. Manual material handling (MMH) activities (lifting, lowering, carrying, moving and pushing/pulling) were observed less frequently in any of the operations.

Conclusion: The results indicate that significant exposures exist that could lead to musculoskeletal injuries of the back and legs for the piledrivers. The results from the study could help target specific hazardous tasks and design ergonomic interventions for those tasks.

There are three aspects of driving skill: knowledge, practical skill, and hazard perception. Knowledge can be learned from books; practical skill can be improved through driving lessons and driving experience. Compared with the other two aspects, hazard perception is difficult to attain in normal driving lessons. We believe that hazard perception is very important for safe driving. The United Kingdom has had a hazard perception test as part of its driving licensing procedure since 2002[1].

Hazard perception [2] is the ability to detect and react to potential hazards. There are some training programs [3] that use video clips to improve a user’s hazard perception skills. Although it is realistic, video clips lack control ability and flexibility. We are now working on a hazard perception training system, which is focused on a variety of complex road conditions and potential hazards. The system includes realistic road environments, different weather conditions, intelligence autonomous vehicles, and a hazards database. The proposed system would have a number of levels: from simple hazard detection in low complexity driving environments, to hazard detection in very fast paced driving environments. We are in the process of defining the specific driving hazards to use in our hazard training perception system. Once a preliminary system has been defined, it will be tested and refined using novice drivers.

Reference
[1] http://en.wikipedia.org/wiki/United_Kingdom_driving_test
[2] Chan, Alan H S, Ng, Annie W Y, Perceptions of implied hazard for visual and auditory alerting signals, Safety Science, Vol. 47 No. 3, pp. 346-352
[3] Vlakveld, W P, Testing and training of hazard perception: study of the testability and trainability of hazard perception in young novice drivers in 2007. Stichting Wetenschappelijk Onderzoek Verkeersveiligheid, D-2008-2