| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
|
|
|
|||||||
|
|||||||||
a Faculty of Physical Education and Recreation Studies, University of Manitoba, Winnipeg, Canada
Michelle M. Porter, Faculty of Physical Education and Recreation Studies, 207 Max Bell Centre, University of Manitoba, Winnipeg, MB, Canada R3T 2N2 E-mail: portermm{at}ms.umanitoba.ca.
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Methods. Twenty-four subjects who were young (20 to 29; n = 6), middle-aged (30 to 64; n = 8), and older (65 years of age and older; n = 10) drove their own vehicles on a 30-km route of various types of roads, with a GPS receiver and video camera recording.
Results. The combination of GPS and video data allowed for the determination of many age-related driving differences. The young subjects drove faster, had a shorter deceleration distance and time, as well as a shorter acceleration time. Young subjects also had a substantially higher number of infraction demerit points primarily due to speeding, not stopping fully at stop signs, and following too closely. Although the older subjects had a smaller number of demerit points assessed, they tended to make different types of errors than the young subjects, including not stopping at all at a stop sign and turning errors.
Conclusions. GPS and video technology offer new opportunities for the assessment of age-related driving performance.
DRIVING is a complex task that requires many attributes for its successful execution. The skills involved include vision, attention, memory, and perceptual motor skills. Many of these attributes decline with age and, therefore, may affect driving ability. For example, reductions in visual processing as well as eye disease increase the crash risk of older drivers (1). Also, older adults are known to initiate and execute movements more slowly as well as imprecisely, which may impair driving skill (2). In addition, medical conditions and/or medications may predispose older adults to increased crash risk (3), and cognitive impairment is also a causative factor in motor vehicle crashes of older drivers (4).
To date, most assessments of driving and aging have been epidemiological or laboratory-based. Few studies examine actual in-vehicle performance, and those that do usually rely upon an observer inside the vehicle (5) or outside making subjective evaluations (6). More tools are required in order to assess driving in an objective or blinded fashion. The global positioning system (GPS) combined with video technology could provide just such a tool.
GPS is used to measure position through its array of navigational satellites (7). Timing signals and precise clock data are transmitted by each satellite in the GPS receiver's field of view. These data are used to perform a triangulation calculation to determine position anywhere on earth, under any weather conditions. A minimum of four satellites' signals are required to obtain a three-dimensional position. In addition, some GPS receivers measure velocity through Doppler methods. From velocity and time data, acceleration and deceleration can be calculated.
The advantages of using GPS for assessing driving include that objective data can be acquired; no observer or operator is needed during driving; the setup is completed with ease in 10 minutes or less; no calibration of equipment is required; an instrumented vehicle is not required; no data acquisition system is needed other than the GPS receiver; power supply is easily handled by batteries or the lighter of the car; the GPS setup will not in any way physically distract the driver; and the bulk of the infrastructure is provided and maintained without cost to the researcher. Commercially available, relatively low-cost GPS receivers offer new possibilities for the assessment of driving. Because an extremely expensive instrumented vehicle is not required, large groups of individuals could feasibly be examined while driving their own vehicles, under different road conditions or constraints. This information used alone or in combination with other data collected could be advantageous in detecting driving behaviors or performance of different drivers. It also could be used to examine the effect of various age-related physiological or medical conditions, retraining or rehabilitation interventions, and road design or sign placement on driving behaviors. All of this can be done without an observer in the vehicle, which means that the driving can be done in a more naturalistic way and the analysis can be done in a blinded fashion. The other advantage in providing a naturalistic assessment tool is that any subject's vehicle can be easily equipped with the equipment.
GPS on its own, though, does not provide a context to driving situations. Information can be attained on how hard someone braked at a stop sign, but it will not be clear what the reason was for this behavior. For example, was the road clear, or was there a sudden movement of another vehicle that caused the individual to brake hard? It is proposed here that a video camera in the car filming the situations that the driver faces provides valuable information along with the objective GPS data. In-vehicle video taping has been used in past studies to investigate driving (8), typically using an instrumented vehicle. However, we are unaware of any studies that have investigated driving using a combination of GPS and video data. The purpose of this article is to establish the use of GPS along with video technology for age-related driving performance evaluation. In this study, vehicle velocity, acceleration, and deceleration, as well as driving infractions, were examined in young, middle-aged, and older drivers to determine if age-related differences could be detected with this technology. It was hypothesized that this technique would be able to detect age-related differences.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Driving Course
The 26-km course included the following types of roads in Winnipeg, Manitoba, Canada: residential two-lane, urban four-lane, urban six-lane, and rural four-lane (the route as depicted by GPS position data is shown in Fig. 1). Posted speed limits were 50, 60, 70, 80, and 100 km/h. Subjects were instructed to drive as they would normally drive.
|
In addition to the GPS receiver, a Canon Optura digital video camera (Canon Inc., Tokyo, Japan), was mounted on a specialized vehicle tripod (Gruppo Manfrotto, Bassano del Grappa, Italy) attached to the front passenger window, about where a passenger's head would be. The camera recorded the whole driving scene in the widest angle possible, such that vehicles and stop signs in front of the vehicle could be seen. In this study, the camera did not monitor what the subject was physically doing.
Data Analysis
The GPS file was differentially corrected offline after testing using a base station file downloaded from the Clay County, Minnesota, Reference Station (http://www.gis.co.clay.mn.us/trshome.htm) and Pathfinder Office v2.51 software (Trimble). Differential correction improves accuracy of the GPS data from about 20 m to 1 m. For this research, the system was set to Universal Transverse Mercator, zone 14 North, and datum WGS 1984. This expresses the data in meters in the east-west (easting) and north-south (northing) direction.
The data files were converted into ASCII format using the DOS program SSFTOASC (Trimble). The ASCII file of date, time, position, and velocity was then imported into SigmaPlot 2000 (SPSS Inc., Chicago, IL) for further analysis. Fig. 1 shows the position data for the whole course as driven by one driver. Fig. 1 shows the velocity profile for the whole course. From the velocity and time data, acceleration/deceleration variables (average, instantaneous, distance, and time) could be calculated for stop signs. Mean values were calculated for each variable based on the data from all stop signs. All stop sign positions were mapped using GPS on a separate occasion by the investigators.
Both the GPS data as well as the video were used for examining driving infractions during the route. The Manitoba Highways and Transportation Driver and Vehicle Licensing Road Test form for Traffic Class 5 was utilized. Demerits were given for specific infractions, according to the guidelines, whereby minor infractions score five demerits and more major infractions are assessed 10 demerits. Categories for errors were starting, stopping, signal violations, vehicles moving on roadway, passing, speed, turning, and inattention. Subjects were rated on total errors and specific errors, by watching the videotape and also by using the GPS data for speed-related infractions and positional information (e.g., velocity as it related to specific roads and also where stop signs were located). Up to 30 demerits were allowed to pass the test, according to Manitoba guidelines. For the purposes of this study, a running tally was given for the whole 30-km course.
Analysis of variance was performed to determine differences between the age groups for velocity, acceleration, and deceleration variables, as well as traffic infractions using SigmaStat (SPSS Inc.). When post hoc tests were indicated by a significant group effect (p < .05), Tukey tests for multiple pairwise comparisons were performed.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
GPS Data
Table 1 shows the means and standard deviations for the GPS data for velocity and stop signs for the three groups. Overall velocity on the whole course was significantly higher in young than in middle-aged or older subjects, but there was no significant difference between the middle aged and older subjects. Stop sign deceleration time also was significantly different in the young compared with the middle-aged and older groups, but no difference was seen between the middle-aged and older subjects. For deceleration distance and acceleration time, the younger subjects had lower values than the older subjects, but no other group-related differences were significant. No other significant differences were seen for the other stop sign variables (average acceleration/deceleration, instantaneous acceleration/deceleration, acceleration distance).
|
|
|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
GPS provides a feasible way to measure position, velocity, and acceleration of a vehicle under typical driving conditions without an observer present. GPS is advantageous in that it can be used in any vehicle with minimal equipment, 10 minutes or less are required for setup, and several minutes of data can be collected without requiring additional power sources or data acquisition systems. From a research design perspective, the method of combining GPS data with video allows for blinding observers to the age, gender, or other characteristics of the subject when examining driving-related behaviors including infractions. This is not only useful for making group comparisons, but it would also be very attractive in assessing interventions in a blinded fashion. Also, because the analysis is not done with an observer in the vehicle, we feel that the driver is more likely to drive as he or she would normally drive, and not as he or she would drive when being observed directly. This remains to be confirmed experimentally. Some subjects exceeded the speed limit by more than 30 km/h (e.g., 118 km/h in an 80 km/h zone), which would have been very unlikely to occur with a driver examiner or any observer in the vehicle, but was likely representative of how this individual would drive daily. Interestingly, there was a much higher proportion of drivers in this study who failed the "test" in comparison to a study where an evaluator was present in the vehicle and was clearly rating the driver (5). Also, in the study by Dobbs and colleagues (5), there were no differences between the age groups in speed-related errors. This demonstrates that the GPS and video method of the present study is potentially better able to capture day-to-day driving behaviors than direct examiner evaluations.
Our finding of the young subjects driving faster than others has been found in the literature through self-report (10). Whereas older subjects are at an increased risk of fatalities and crashes, young drivers are known to be at an increased risk of crashes and fatalities, but for different reasons, including risk-taking behavior and inexperience (9). Our results seem to support the contention that younger drivers like to take risks (i.e., drive fast) and that older drivers are more likely to avert risk but have perceptual/judgment problems (11). Although our findings on age-related differences in driving are preliminary at this time because of the small number of subjects and the nonrepresentative sample, this study indicates the future promise for using this methodology in a larger, more representative group of drivers, across the lifespan. In the future, gender differences should be examined, because differences in psychomotor variables, such as reaction time, and psychological variables, like risk-taking, may lead to the discrepancies in crash and fatality involvement rates of men and women across the lifespan (9).
In conclusion, this study demonstrated the unique use of GPS, in combination with video recording, for assessing driving behaviors. This preliminary study found that younger subjects drove faster than the two older age groups and made more total traffic infractions, while older subjects made errors of omission like unintentionally failing to stop at a stop sign. This method is feasible, is relatively low-cost compared to an instrumented vehicle, and provides many advantages from a research design perspective. The advantages include the ability to provide objective data, as well as the ability to blind observers to more objective data like traffic infractions. Future research utilizing this method may include studies examining age-related or disease-related differences in driving behaviors as well as determining the effectiveness of rehabilitation or retraining programs on driving performance.
|
Acknowledgments |
|---|
Received December 20, 2001
Accepted April 9, 2002
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. T. Grossberg and A. K. Desai Management of Alzheimer's Disease J. Gerontol. A Biol. Sci. Med. Sci., April 1, 2003; 58(4): M331 - 353. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| All GSA journals | The Gerontologist |
| Journals of Gerontology Series B: Psychological Sciences and Social Sciences | |
