Leveraging the Second Strategic Highway Research Program Naturalistic Driving Study: Examining Driver Behavior When Entering Rural High-Speed Intersections

508 CAPTIONS

Figure 1. Map. Crossings through intersection 1. This map provides an overhead view of a four–way intersection. The road running from left to right across the photo is highlighted.

Figure 2. Map. Crossings through intersection 2. This map provides an overhead view of a four–way intersection. The road running from left to right across the photo is highlighted, and the highlighting partially extends from the intersection down the road extending toward the bottom of the photo.

Figure 3. Map. Crossings through intersection 3. This map provides an overhead view of a four–way intersection. The road running from the top of the photo to the bottom is highlighted, and the highlighting extends from the intersection to the road extending toward the upper right–hand of the photo.

Figure 4. Map. Crossings through intersection 4. This map provides an overhead view of a four–way intersection. The road running from left to right across the photo is highlighted, and there are two cars parked near the bottom right–hand of the intersection.

Figure 5. Photo. Example of glance to far left. This photo shows a close–up of a driver in a car who is looking off to her left, almost over her left shoulder.

Figure 6. Photo. Example of glance to near left. This photo shows a close–up of a driver in a car who is looking slightly off to her left.

Figure 7. Photo. Example of glance to road ahead. This photo shows a close–up of a driver in a car who is looking straight ahead through the windshield of the car.

Figure 8. Photo. Example of glance to rearview. This photo shows a close–up of a driver in a car who is looking into the rearview mirror in order to see behind her.

Figure 9. Photo. Example of glance to near right. This photo shows a close–up of a driver in a car who is looking slightly off to her right.

Figure 10. Photo. Example of glance to far right. This photo shows a close–up of a driver in a car who is looking off to her right, almost over her right shoulder.

Figure 11. Photo. Example of glance to cell phone. This photo shows a close–up of a driver in a car who is looking down at a cell phone.

Figure 12. Illustration. Traffic vehicle path approaching from left. This illustration shows a traffic vehicle path approaching from the left. It features a rectangle in the middle. There is a letter outside each side of the rectangle: A (top), L (left), R (right), and D (bottom). Inside the rectangle, there are three arrows extending from L. The top arrow curves upward and points at A, the middle arrows extends straight across the rectangle and points at R, and the bottom arrow curves downward and points at D. Outside of the rectangle in the corners of the figure, there are blue quarter circles. There is a stop sign inside of the upper left and lower right quarter circles.

Figure 13. Illustration. Traffic vehicle path approaching from the driver. This illustration shows a traffic vehicle path approaching from the driver. It features a rectangle in the middle. There is a letter outside each side of the rectangle: A (top), L (left), R (right), and D (bottom). Inside the rectangle, there are three arrows extending from D. The left arrow curves to the leftward and points at L, the middle arrows extends straight across the rectangle and points at A, and the right arrow curves rightward and points at R. Outside of the rectangle in the corners of the figure, there are blue quarter circles. There is a stop sign inside of the upper left and lower right quarter circles.

Figure 14. Illustration. Traffic vehicle path approaching from right. This illustration shows a traffic vehicle path approaching from the right. It features a rectangle in the middle. There is a letter outside each side of the rectangle: A (top), L (left), R (right), and D (bottom). Inside the rectangle, there are three arrows extending from R. The top arrow curves upward and points at A, the middle arrows extends straight across the rectangle and points at L, and the bottom arrow curves downward and points at D. Outside of the rectangle in the corners of the figure, there are blue quarter circles. There is a stop sign inside of the upper left and lower right quarter circles.

Figure 15. Illustration. Traffic vehicle path approaching from ahead. This illustration shows a traffic vehicle path approaching from ahead. It features a rectangle in the middle. There is a letter outside each side of the rectangle: A (top), L (left), R (right), and D (bottom). Inside the rectangle, there are three arrows extending from A. The left arrow curves leftward and points at L, the middle arrows extends straight across the rectangle and points at D, and the right arrow curves rightward and points at R. Outside of the rectangle in the corners of the figure, there are blue quarter circles. There is a stop sign inside of the upper left and lower right quarter circles.

Figure 16. Illustration. Traffic vehicle path approaching from unknown direction. This illustration shows a traffic vehicle path approaching from an unknown direction. It features a rectangle in the middle. There is a letter outside each side of the rectangle: A (top), L (left), R (right), and D (bottom). In the middle of the rectangle is a U with a question mark next to it to represent unknown. There are four arrows extending from the U? (one in each direction–L, A, R, and D). Outside of the rectangle in the corners of the figure, there are blue quarter circles. There is a stop sign inside of the upper left and lower right quarter circles.

Figure 17. Graph. Brake distance (non–queued crossings only). This histogram shows brake distance for non–queued crossings only. The y–axis shows count and ranges from 0 to 50. The x–axis shows brake distance relative to intersection and ranges from 0 to 500 ft. The first bar denotes 1 observation in the 25–50 ft range; the rest of the bars denote the following: 3 observations in the 50–75 ft range, 2 observations in the 75–100 ft range, 3 observations in the 100–125 ft range, 7 observations in the 125–150 ft range, 16 observations in the 150–175 ft range, 15 observations in the 175–200 ft range, 21 observations in the 200–225 ft range, 48 observations in the 225–250 ft range, 37 observations in the 250–275 ft range, 28 observations in the 275–300 ft range, 22 observations in the 300–325 ft range, 16 observations in the 325–350 ft range, 23 observations in the 350–375 ft range, 15 observations in the 375–400 ft range, 16 observations in the 400–425 ft range, 12 observations in the 425–450 ft range, 11 observations in the 450–475 ft range, 35 observations in the 475–500 ft range, 41 observations in the 500–525 ft range, 6 observations in the 525–550 ft range, 1 observation in the 550–575 ft range, 0 observations in the 575–600 ft range,

Figure 18. Equation. Estimated brake distance (ft) as a function of brake speed (mi/h). Brake distance equals 7.6 times brake speed minus 142.3.

Figure 19. Graph. Mean brake distance (ft) by speed at brake point (mi/h). This graph shows the mean brake distance by speed at brake point. The y–axis shows brake distance and ranges from 0.0 to 500 ft. The x–axis shows speed at brake application and ranges from 40 to 80 mi/h. The bars increase in value from left to right, and each bar has an error bar that extends approximately equally below and above the top of each bar.

Figure 20. Graph. Mean brake distance (ft) by speed at brake point (mi/h) and age. This graph shows mean brake distance by speed at brake point and age. The y–axis shows brake distance and ranges from 0 to 500 ft. The x–axis shows speed at brake application and ranges from 40 to 80 mi/h in increments of 10 mi/h. For each of the five speeds shown on the x–axis, there are two bars: one for the older age group (45–84 years old) and one for the younger age group (16–44 years old). Both sets of bars increase in value in speed at brake application, but the older age group bars increase at a greater rate than the younger age group bars. Each bar has an error bar that extends approximately equally below and above the top of each bar.

Figure 21. Graph. Minimum speed (all crossings). This histogram shows the minimum speed for all crossings. The y–axis shows count and ranges from 0 to 200. The x–axis shows minimum speed and ranges from 0 to 15 mi/h. The first bar appears at 0 mi/h with a count of 200, and the rest of the bars appear as follows: 1 mi/h (count of 5), 2 mi/h (count of 35), 3 mi/h (count of 25), 4 mi/h (count of 30), 5 mi/h (count of 25), 6 mi/h (count of 12), 7 mi/h (count of 25), 8 mi/h (count of 12), 9 mi/h (count of 10), 10 mi/h (count of 8), 11 mi/h (count of 8), 12 mi/h (count of 8), 13 mi/h (count of 10), 14 mi/h (count of 4), and 15 mi/h (count of 7). There is also a bar past the 15 mi/h increment with a count of 1.

Figure 22. Graph. Minimum speed by traffic condition. This boxplot shows the minimum speed by traffic condition. It has four increment labels on the y–axis (from top to bottom): no cross traffic (CT) or queue, queue without CT, CT without queue, and CT and queue. The x–axis shows minimum speed and ranges from 0 to 15 mi/h. Vertical lines appear in the middle of the boxes for no CT or queue and queue without CT, denoting the median values. Each box has a line extending from the right side for various distances; the no CT or queue box also has a short line extending from the left size of the box. The right–hand line for the no CT or queue box extends almost all the way to the right side of the figure, but the lines for the other three boxes are followed by lines of dots.

Figure 23. Graph. Probability of making a complete stop by AAM for each definition of “complete.” This graph shows the probability of making a complete stop by average annual mileage (AAM) for each definition of a complete stop. The y–axis shows Pr(complete stop) and ranges from 0.00 to 1.00. (“Pr” was defined as “probability” in the text.) The x–axis shows complete stop definition threshold and has two levels: 0 and 3 mi/h. There are two bars shown: 10,000 to 20,000 mi and 20,000 to 30,000 mi. Both sets of bars increase in value from left to right, although the bars representing the higher AAM have significantly higher values than the lower AAM bars. Each bar has an error bar. Those for the lower AAM bars extend further above the top of the bars than they do below. The error bars for the higher AAM bars extend further below the top of the bars than they do above.

Figure 24. Graph. Probability of making a rolling stop by expressed risk associated with performing a rolling stop for each definition of “complete.” This graph has two sets of bars. The y–axis is labeled as “Pr(rolling stop)” and ranges from 0.00 to 1.00. (“Pr” was defined as “probability” in the text.) The x–axis is labeled as “complete stop definition threshold (mi/h)” and has three levels: 0, 3 and 6. The top of the graph is labeled “rolling stop risk” and includes the key: the pink bar with diagonal white lines represents medium risk, and the blue bar represents high risk. Both sets of bars decrease in value from left to right, although the pink bars ultimately decrease in greater magnitude than the blue bars. Each bar has an error bar; the error bars for the pink bars extend farther above the top of the bars than they do below the top of the bars, and the error bars for the blue bars extend farther below the top of the bars than they do above the top of the bars.

Figure 25. Graph. Probability of making a rolling stop by expressed tendency to perform a rolling stop for each definition of “complete.” This graph has two sets of bars. The y–axis is labeled as “Pr(rolling stop)” and ranges from 0.00 to 1.00. (“Pr” was defined as “probability” in the text.) The x–axis is labeled as “complete stop definition threshold (mi/h) and has three levels: 0, 3, and 6. The top of the graph is labeled “rolling stop tendency” and includes the key: the pink bar with diagonal white lines represents never/rarely, and the blue bar represents often/sometimes. Both sets of bars decrease in value from left to right, although the right–most pink and blue bars are almost equal in value. Each bar has an error bar. The error bars for the first two pink bars extend farther below the top of the bars for than they do above the top of the bars, and the error bar for the third pink bar extends slightly farther above the top of the pink bar than it does below the top of the pink bar. The error bar for the first blue bar extends farther below the top of the bar than it does above the top of the bar, the error bar for the second blue bar extends approximately equally above and below the top of the blue bar, and the error bar for the third blue bar extends farther above the top of the blue bar than it does below the top of the blue bar.

Figure 26. Graph. Discrete eyeglances along intersection approach. This scatterplot shows discrete eyeglances along the intersection approach. It includes translucent markers that form eight lines that are represented on the x–axis and show glance targets, which includes, from left to right, respectively: far left, near left, forward, rearview, near right, far right, cell, and other. The y–axis shows distance and ranges from 0 to 492.1 ft. Each line’s intensity of color varies. The line for far left is solid from 0 to 8.4 ft and light in intensity from 98.4 to 196.9 ft., and it varies from light to very light in intensity from 196.9 to 492.1 ft.  The line for near left is light in intensity around 0 ft, solid from approximately 5 to 98.4 ft., and it varies from light to very light in intensity from 98.4 to 492.1 ft. The line for forward is solid with no variation from 0 to 492.1 ft. The line for rearview is only solid near the 20 to 50 ft mark and very light in intensity everywhere else. The line for near right is solid from 0 to 98.4 ft and is light in intensity from 98.4 to 492.1 ft. The line for far right is similar to that of near left. The line for cell is light in intensity from 0 to 98.4 ft, and very light in intensity everywhere else. The line for other is light in intensity and varies only slightly from 0 to 492.1 ft.

Figure 27. Heat map. Mean total glance duration among all crossings. This heat map shows the mean total glance duration among all crossings. It shows a grid with five rows and eight columns. At the bottom of each column is a glance target, which includes, from left to right, respectively: far left, near left, forward, rearview, near right, far right, cell, and other. The left side of the grid is labeled “Segment,” and each row is numbered from 1 to 5 from top to bottom, respectively. The top of the grid shows a legend labeled “Mean Total Glance Duration (s)” with a key indicating that the darker the color, the larger the number of seconds. The key goes from 1 to 4 s. Each cell in the heat map includes a number, and the higher the number, the darker the fill color is. The far left column numbers are (from top to bottom) 2.40, 0.12, 0.06, 0.03, and 0.03. The near left column numbers are (from top to bottom) 0.39, 0.10, 0.04, 0.04, and 0.01. The forward column numbers are (from top to bottom) 4.15, 2.14, 1.74, 1.42, and 1.34. The rearview column numbers are (from top to bottom) 0.08, 0.01, 0.01, 0.00, and 0.01. The near right column numbers are (from top to bottom) 0.48, 0.08, 0.07, 0.08, and 0.07. The far right column numbers are (from top to bottom) 1.83, 0.03, 0.02, 0.03, and 0.00. The cell column numbers are (from top to bottom) 0.03, 0.04, 0.01, 0.01, and 0.00. The other column numbers are (from top to bottom) 0.12, 0.07, 0.06, 0.04, and 0.05.

Figure 28. Heat map. Mean total glance durations among crossings with no CT or vehicle queues. This heat map shows the mean total glance duration among all crossings. It shows a grid with five rows and eight columns. At the bottom of each column is a glance target, which includes, from left to right, respectively: far left, near left, forward, rearview, near right, far right, cell, and other. The columns for rearview and cell are blank (no gridlines or numbers). The left side of the grid is labeled “Segment,” and each row is numbered from 1 to 5 from top to bottom, respectively. The top of the grid shows a legend labeled “Mean Total Glance Duration (s)” with a key indicating that the darker the color, the larger the number of seconds. The key goes from 0 to 2 s. Each cell in the heat map includes a number, and the higher the number, the darker the fill color is. The far left column numbers are (from top to bottom) 1.30, 0.13, 0.06, 0.05, and 0.05. The near left column numbers are (from top to bottom) 0.21, 0.14, 0.03, 0.07, and 0.00. The forward column numbers are (from top to bottom) 2.40, 1.79, 1.55, 1.30, and 1.28. The near right column numbers are (from top to bottom) 0.22, 0.06, 0.09, 0.12, and 0.13. The far right column numbers are (from top to bottom) 0.95, 0.06, 0.04, 0.03, and 0.00. The other column numbers are (from top to bottom) 0.03, 0.04, 0.07, 0.04, and 0.03.

Figure 29. Graph. Estimated mean total glance duration for collapsed ROIs across segments. This graph shows the mean total glance duration for collapsed regions of interest (ROIs) across segments. It is split into three sections: forward (left), scanning (middle), and other (right). The three sections share a y–axis label (total glance durations that ranges from 0.00 to 5.1 s and an x–axis that shows segment, which is numbered incrementally from 5 (left) to 1 (right). In each section, the five bars increase in value from left to right, although in the forward and scanning sections, the increase is dramatic for segment 1 bars, while there is relatively little increase over the course of the bars in the other segment. Each bar has an error bar that extends approximately equally below and above the top of each bar.

Figure 30. Graph. Estimated mean total glance duration for collapsed ROIs across segments among crossings with no CT or vehicle queues. This graph shows the mean total glance duration for collapsed regions of interest (ROIs) across segments with no CT or vehicle queues. It is split into three sections: forward (left), scanning (middle), and other (right). The three sections share a y–axis label (total glance durations (s) that ranges from 0.00 to 2.7 s and an x–axis that shows segment, which is numbered incrementally from 5 (left) to 1 (right). In the forward scanning section, the five bars increase in value from left to right, and each bar has an error bar that extends approximately equally below and above the top of each bar. In the scanning section, there is a slight increase for the 4 bar, a slight decrease for the 3 bar, another slight increase for the 2 bar, and then a dramatic increase for the 1 bar. Each bar has an error bar; the error bar extends farther above the top of the bar than it does below the top of the bar for bars 5 through 2, and the error bar for bar 1 extends approximately equally below and above the top of each bar. In the other section, bars 5 and 4 are approximately equal in value, there is a slight increase for bar 3, there is a very slight decrease for bar 2, and then there is another decrease for bar 1.

Figure 31. Graph. Pre–stop scanning percentage by stop type for each definition of “complete”. This graph shows the pre–stop scanning percentage by stop type for each definition of a complete stop. The y–axis shows pre–stop scanning percentage and ranges from 0 to 80 percent. The x–axis shows complete stop definition threshold and has three levels: 0, 3, and 6 mi/h. For each complete stop definition threshold, there are two bars: complete stops and rolling stops. Both stop types increase in value from left to right, and each bar has a very narrow error bar.

Figure 32. Graph. Pre–stop scanning percentage by stop type and traffic condition for each definition of “complete”. This graph shows the pre–stop scanning percentage by stop type and traffic condition for each definition of a complete stop. It is split into three horizontal sections by complete stop definition threshold (from top to bottom): less than or equal to 0 mi/h, less than or equal to 3 mi/h, and less than or equal to 6 mi/h. For each section, the y–axis shows pre–stop scanning percentage and ranges from 0 to 75 percent, and the x–axis shows traffic conditions, which includes no cross traffic (CT) on queue, queue without CT, CT without queue, and CT and queue. For each traffic condition, there are two bars: complete stops and rolling stops. In the less than or equal to 0 mi/h section, the complete stops bar increases from the first, and then the third and fourth complete stops bars both decrease. The first and second rolling stops bars are roughly equivalent, and then there are slight decreases for the third and fourth rolling stops bars. In the less than or equal to 3 mi/h section, the complete stops bar increases slightly from the first, and then the third and fourth complete stops bars both decrease. The first and second rolling stops bars are roughly equivalent, and then there are slight decreases for the third and fourth rolling stops bars. In the less than or equal to 6 mi/h section, the second complete stops bar increases slightly from the first, and then the third and fourth complete stops bars both decrease. The second rolling stops bar increases from the first, the third rolling stops bar decreases, and the fourth rolling stops bar is roughly equivalent to the third one. All of the bars have narrow error bars.