Click for next page ( 15

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement

Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 14
14 CHAPTER 3 Static Screening Experiment The results of the literature review and survey were used Format, Intensity, Color, and Flash Pattern to develop and conduct a static screening experiment in LED and Halogen Panel Lights. The LED and halogen which 41 lighting configurations were tested in daytime panel-mounted lights were only capable of illuminating at one and nighttime with regard to attention-getting, discomfort intensity setting. The high-intensity condition was achieved glare, meaning, and ability to detect using peripheral vision. by activating two adjacent lights at once, which then flashed The lighting conditions of color, intensity, flash pattern, simultaneously (for the flashing conditions). Using a light and lamp type were all investigated in this experiment using controller, four different flash patterns were tested: (1) steady, human participants in a static (stationary vehicle) test sce- (2) synchronous at 1 Hz (the light or lights on each side flash- nario. The four lighting configurations were then carried ing at the same time at 1 time per second), (3) asynchronous forward into a dynamic experiment. A complete description at 1 Hz (the lights or lights on each side flashing in an alter- of the experimental methods and the results is included as nating pattern), and (4) asynchronous at 4 Hz. The two inten- Appendix D. sities and four light patterns made up eight conditions. Four colors were also tested (red, blue, amber, and white), adding Experimental Methods an additional six conditions. The lighting panel layout did not allow both of these types of lights to be mounted at once; Experimental Design therefore, half of the participants received the LED-light con- The choice of independent variables (factors that were ditions and half received the halogen-light conditions. Each manipulated during the experiment) was driven by factor participant received the same condition (LED or Halogen) at rankings suggested by the knowledgeable practitioners de- both nighttime and daytime. scribed in "Identification of Relevant Factors" in Chapter 1 and shown in Appendix B. Variables considered included Strobe Panel Lights. The strobe lights mounted on the between-subjects variables (gender and age, each with two panel were capable of being set at two different intensities levels) and within-subject variables (intensity with two levels, (low and high). The high-intensity condition was achieved format with five levels, contrast with two levels, flash pattern by activating two adjacent lights at once, which then flashed with four levels, color with four levels, and position with two simultaneously. Two flash patterns were selected to test these three different intensities. The first pattern was a double flash, levels). Because of the large number of identified variables, a while the second pattern was a quad flash. The same four colors mixed-factor partial factorial design was used to allow explo- were tested (red, amber, blue, and white). ration of the most relevant main effects and interactions. In other words, because it was not feasible to present every par- Rotating Beacons. The rotating beacons selected for this ticipant with every possible combination of the independent study were capable of rotating at two speeds (slow and fast, variables in the time allotted, only the most relevant com- equating to two flash patterns) and had two intensity settings binations were used. The final experiment design resulted (low and high). The rotating beacons were tested with four in a total of 82 conditions for each participant: 41 during colors (red, amber, blue, and white). the day, and 41 during the night. The daytime and night- time conditions were essentially the same except for time of Passive Format. The passive condition refers to the use presentation. of retroreflective tape on the black test panel. Two strips of

OCR for page 14
15 retroreflective tape were positioned on either side of the test conducted in the daytime, when contrast of the lights was at its panel at 90 angles. They were covered when they were not lowest (this represents the worst situation for detecting lights being tested. using peripheral vision). Recognition. The recognition dependent variable inves- Location tigated how subjects classify the warning-light scheme relative "Daylighting" refers to placing the lights on top of the black to a list of typical vehicle functions. This questionnaire was panel in order to test how well they can be seen against the sky. only administered at night, when the reaction was expected In comparison, placing the lights on a shelf in front of the black to be more due to the lights (number, type, arrangement, and panel might increase their conspicuity because of increased color) than to external cues such as the vehicle type holding contrast. One strobe beacon, one halogen beacon, and one LED the lighting panel. beacon were tested both on top of the panel and on a shelf in front of the panel. All beacons either rotated or gave the Glare. The glare dependent variable measured the discom- impression of rotation by sequential activation of lights. fort subjects experienced when presented with the warning- light configurations. A nine-point rating scale ranging from "not noticeable" to "unbearable" was used to capture the rat- Randomized Assignment of Treatment Orders ings. This scale was administered only at nighttime, when dis- Treatment orders were randomly assigned to participants comfort glare is at its worst because of the high degree of to reduce variation in responses that may have arisen from contrast between the lights and their background. factors not considered in the experiment, such as fatigue. Six random orders were used for the 24 participants. Subjects were Participants tested during the daytime before being tested at nighttime. Twenty-four subjects, 12 males and 12 females, were selected to participate in this study. The subjects were evenly represented Dependent Variables in two age groups of 25 to 35 years old (younger) and 65+ years Five dependent variables (those factors measured during old (older). Institutional Review Board (IRB) approval for the the experiment) were used in this experiment. The daytime use of human participants was obtained prior to recruiting and nighttime attention-getting ratings measured how well subjects. Once subjects arrived for the first session, they read the lighting conditions caught the participants' attention. The and signed an informed consent form before beginning any horizontal maximum peripheral detection angle was measured experimental activities. Subjects were paid $20/h, and they only during daytime conditions, while the recognition and glare were allowed to withdraw at any point in time, with payment rating dependent variables were measured only at nighttime. adjusted accordingly. Attention-Getting. To establish the effectiveness of the Apparatus lighting and marking configuration, a metric for attention- getting was developed. This metric used a seven-point rating Test Road scale ranging from "not at all attention getting" to "extremely attention getting"; it was administered in both the daytime The experiment took place on a bridge to an unfinished and nighttime tests. highway adjacent to the Virginia Tech Transportation Institute (VTTI) and the Virginia Smart Road. A degree of control was Horizontal Maximum Peripheral Detection Angle. The attained by not allowing public vehicles and pedestrians to enter horizontal maximum peripheral detection angle measured the the bridge. However, participants could see passing traffic on maximum horizontal angle at which the warning light could a lower-level highway during nighttime conditions. be detected (in other words, how well the lights could be seen using peripheral vision). Subjects were asked to look at pre- Test Vehicles determined positions located 15 apart (from 0 to 90 away from the forward view) and state whether they could detect the A Virginia Department of Transportation (VDOT) no- warning light. The progression of detection angles was con- longer-in-service dump truck was obtained for this experi- ducted both going up (further away from the forward view) ment. A large panel was used to mount the different lighting and going down (closer to the forward view), and three re- configurations; the panel was attached to the rear of the dump sponses were then averaged. Discrepancies of more than 30 truck (Figure 2). Small shelves were used to place the beacons among the three responses were retested. This test was only on the lower shelf position during the testing.

OCR for page 14
16 Rotating beacons placed on the top of the panel. Rotating beacons placed on shelves. Figure 2. VDOT dump truck outfitted with warning lights. For each grid of lights, the two left-most columns housed was Avery Dennison DOT Type C2 material. The retroreflec- LEDs (or halogens for one-half of the participants), while the tive tape was placed underneath the LED and strobe lights, and two right-most columns housed strobe lights. The LED and was covered with a magnetically fastened black strip of rubber strobe lights were thus evenly spaced between the left and right when it was not being tested. side, and to the subject viewing them from 400 ft, their location in space appeared to be the same. All lights were manually Methods controlled by an operator who sat behind the panel in the bed of the dump truck. There was radio communication between Subjects were greeted and asked to show their driver's li- the experimental vehicle and the test truck. Participants were censes. The purpose of the study was explained and they then stationed inside a 2002 Cadillac Escalade, along with an in- read and signed the informed consent form. Three vision tests vehicle experimenter who provided directions and recorded were administered: the Snellen vision test, contrast sensitivity responses on a laptop computer. test, and color blindness test. Subjects had to have a minimum of 20/40 vision using the Snellen acuity test in order to par- ticipate. Prior to the test trials, time was taken to orient sub- Lights jects to the study, the vehicle, and the procedures. Subjects The lights used for this experiment were commercially avail- were shown how to adjust the position of the seat in order to able light sources acquired from manufacturers. The light be comfortable. sources were selected for the ease of use, photometric charac- Instructions were provided once subjects were seated in the teristics, and suitability for the experiment (available in the re- vehicle. During the daytime session, subjects were instructed quired colors, with various intensities and flash patterns also to rate the conspicuity of the warning lights. Subjects then available). The retroreflective tape used in this experiment performed a maximum peripheral detection task to assess