5. DRIVER ACCEPTANCE (continued)

5.5.3 Perceived Value – FCW

The perceived value objective examined driver satisfaction with the FCW system, awareness of safety, compatibility with an individual’s mental model, and driving skill enhancement. This section includes both descriptive and quantitative discussion of perceived value, as well as driver comments provided during debriefing and focus group sessions, and a concluding summary. The focus of the discussion for perceived value of the FCW system is necessarily descriptive, as statistically significant differences among age groups were minimal.

5.5.3.1 FCW Statistical Findings

Nine survey items were used to measure perceived value for FCW. Table J-1, as found in Appendix J provides the intercorrelations among the perceived value measures and identifies the statistically significantly intercorrelated items. Two of the 9 items referred to manual driving situations and, as such, were used for comparative purposes only. The remaining 7 items were subjected to correlational analysis. Sixteen of the 21 resulted in significant findings in the expected direction, suggesting that the chosen measures were internally consistent in their assessment of the construct of perceived value.

Across the entire FOT sample, opinions of the perceived value of the FCW system were generally positive. Table 5-9 displays the measures of central tendency and standard deviations for each measure. Of the 9 items, the two assessing “overall” attitudes regarding satisfaction and the potential for increased driving safety were less positive than the others.

Table 5-9. Perceived Value Sub-objective Measure Descriptive Statistics

Sub-objective	Survey Item	Mean	Standard Deviation	Median	Mode
Overall
	Overall how satisfied were you with the FCW system?	4.8	1.9	5.0	7.0
	1 (very unsatisfied) - 7
C ompatibility with mental model
	Overall, how easy was it to remember how to use and operate FCW while driving?	6.7	0.6	7.0	7.0
	1 (not at all easy) - 7
Driving skill enhancement
	Did you feel more comfortable performing additional tasks, (e.g., adjusting the heater, operating the radio, talking on a cellular telephone, etc.) while using the FCW system as compared to manual driving?	5.2	1.3	5.0	4.0
	1 (less comfortable) - 7
Safety
	How safe did you feel while driving the car using FCW?	6.0	1.1	6.0	7.0
	1 (very unsafe) - 7
	How safe did you feel driving the car manually?	6.7	0.6	7.0	7.0
	1 (very unsafe) - 7
	How easy or difficult did you find it to maintain a safe distance to the preceding vehicle when using FCW?	6.2	1.1	6.0	7.0
	1 (very difficult) - 7
	How easy or difficult did you find it to maintain a safe distance to the preceding vehicle when driving manually?	6.4	0.9	7.0	7.0
	1 (very difficult) - 7
	When using FCW, do you feel you drove more or less safely than when driving manually?	5.1	1.4	5.0	4.0
	1 (less safe) - 7
	Overall, I think that FCW is going to increase my driving safety	4.6	1.9	5.0	7.0
	1 (strongly disagree) - 7

As depicted in Figure 5-31, drivers reported a range of scores for overall satisfaction with FCW. At the positive end of the scale, a total of 49 percent of drivers responded with satisfaction ratings of 6 or 7. By contrast, a smaller percentage (16%) were unsatisfied with the system (values 1 or 2). More than one third of the sample had a neutral opinion (values 3, 4, or 5). The neutral-to-positive level of overall satisfaction with the FCW is reflected in the mean score of 4.8 for this item.

Figure 5-31. Overall, How Satisfied Were You with the FCW System?

[d]

Drivers rated FCW as easy to remember how to use and operate while driving, as illustrated in Figure 5-32. Almost three-quarters of the drivers gave FCW the highest score, 7, “very easy,” and 94 percent rated FCW as 6 or 7, indicating that the FCW implementation was easy to comprehend.

Figure 5-32. Overall, How Easy Was it to Remember How to Use and Operate FCW
While Driving?

[d]

Because FCW provides an additional support to the driving task, individuals may have believed that they could perform additional actions safely while driving using the system. Drivers were asked if they felt more comfortable performing additional tasks, such as talking on the cell phone or adjusting the heater, when they drove using FCW. The mean score was 5.2 and the mode for responses was a neutral “4.” Neutral zone responses of 3 – 5 represented 59 percent of the sample, indicating that most drivers felt equally as comfortable performing additional tasks while driving with FCW as when driving manually (see Figure 5-33). Importantly, no drivers indicated that they felt less comfortable (rating of 1 or 2), as compared to manual operations.

Figure 5-33. Did You Feel More Comfortable Performing Additional Tasks While Using the FCW System, as Compared to Manual Driving?

[d]

Drivers were asked to rate how safe they felt when they drove with FCW. More than three quarters of the drivers (77%) rated themselves as feeling very safe driving using FCW, responding with scores of 6 or 7 (see Figure 5-34). In fact, the mean score for this item was 6.0. Only 3 percent of the FOT drivers indicated that they felt unsafe to some degree.

Figure 5-34. How Safe did You Feel while Driving the Car Using FCW?

[d]

Drivers were asked both how safe they felt driving the car during the first six days of the FOT (ACAS-disabled, manual driving) and using FCW. Figure 5-35 compares responses to these survey items, where, in each case, more the vast majority of the ratings show that drivers felt quite safe (scores of 6 or 7), though a greater percentage of respondents felt “very safe” driving manually versus with FCW (74% and 42%, respectively).

Figure 5-35. Comparison of How Safe Drivers Felt Driving the Car Driving Manually
Versus Using FCW

[d]

Drivers were asked to rate how easy or difficult they found it to maintain a safe distance to the preceding vehicle when they drove the car manually, prior to when FCW was enabled, and then subsequently, using FCW. A comparison of responses, as depicted in Figure 5-36, indicates that in both cases more than 80 percent of the drivers found it very easy (scores of 6 or 7) to maintain a safe distance. A small percentage (2%) of respondents responded that they found it very difficult to maintain a safe distance using FCW.

Figure 5-36. Comparison of Ease of Maintaining a Safe Distance to the Preceding Vehicle Driving Manually Versus Using FCW

[d]

A related survey item, used to compare attitudes regarding safety, asked drivers the degree to which they felt more or less safe using FCW, compared to manual driving. The mean score was 5.1, suggesting that drivers felt slightly safer driving with FCW. However most commonly, responses fell at neutral (4), indicating that there was no discernable difference. Figure 5-37 depicts the response distribution for this item, showing that a total of 7 percent of all drivers felt less safe (scores from 1 – 3) driving with the FCW system.

Figure 5-37. When Using FCW, Do You Feel You Drove More or Less Safely Than
Manual Driving?

[d]

As a final means of assessing FCW safety issues, drivers were questioned regarding the degree to which they agreed that prospective use of FCW would increase their driving safety. Responses were distributed over the entire range of the item scale as shown in Figure 5-38. Overall, attitudes were largely neutral, as 43 percent of drivers responded in the 3 – 5 range and the mean score for this measure was 4.6. A total of 39 percent of all responses fell on the positive end of the scale (scores of 6 or 7), while 18 percent of responses fell at the opposite end of the scale (scores of 1 or 2).

Figure 5-38. Overall, FCW Is Going to Increase My Driving Safety

[d]

Given the previously noted focus on potential age group differences, ANOVA was used to verify the nature of any statistical relationships. Table 5-10 reports the statistical relationships among driver age groups and attitudes regarding perceived value, as broken down by sub-objective, with any significant group differences noted briefly in the rightmost “Results” column and nonsignificant results denoted using “NS.”

Table 5-10. Perceived Value Sub-objective Measures by Driver Age Group

Sub-objective	Survey Item	Age Group	Mean	ANOVA? Results
Overall
	Overall how satisfied were you with the FCW system?	Younger	4.5	NS
1 (very unsatisfied) - 7		Middle	4.5
		Older	5.5
C ompatibility with mental model
	Overall, how easy was it to remember how to use and operate FCW while driving?	Younger	6.7	NS
1 (not at all easy) - 7		Middle	6.5
		Older	6.8
Driving skill enhancement
	Did you feel more comfortable performing additional tasks, (e.g., adjusting the heater, operating the radio, talking on a cellular telephone, etc.) while using the FCW system as compared to manual driving?	Younger	5.0	NS
	1 (less comfortable) - 7	Middle	5.0
		Older	5.5
Safety
	How safe did you feel while driving the car using FCW?	Younger	5.7	NS
	1 (very unsafe) - 7	Middle	6.1
		Older	6.2
	How safe did you feel driving the car manually?	Younger	6.5	NS
	1 (very unsafe) - 7	Middle	6.7
		Older	6.8
	How easy or difficult did you find it to maintain a safe distance to the preceding vehicle when using FCW?	Younger	6.5	Y found less difficult than M
	1 (very difficult) - 7	Middle	5.7
		Older	6.2
	How easy or difficult did you find it to maintain a safe distance to the preceding vehicle when driving manually?	Younger	6.4	NS
	1 (very difficult) - 7	Middle	6.2
		Older	6.5
	When using FCW, do you feel you drove more or less safely than when driving manually?	Younger	5.0	NS
	1 (less safe) - 7	Middle	5.1
		Older	5.2
	Overall, I think that FCW is going to increase my driving safety.	Younger	4.5	NS
	1 (strongly disagree) - 7	Middle	4.4
		Older	4.9

The single statistically significant finding for perceived value survey measures is depicted in Figure 5-39. Here, the younger driver age group reported finding it easier to maintain a safe distance to the preceding vehicle using FCW than those in the middle-age group (F(2, 63) = 3.11, p = .04). However, it may be noted from group means that the difference in attitudes is not striking.

Figure 5-39. Ease of Maintaining a Safe Distance to Lead Vehicle while Using FCW
by Age Group

[d]

5.5.3.2 Investigating Travel Behavior and FCW Perceived Value Measures

Follow-on analyses were conducted to determine the degree to which objective travel behavior measures were correlated with the perceived value survey measures. The objective measures encompassed ACAS-enabled driving and included FCW distance traveled (km) in valid trips, number of FCW alerts and mean number of FCW alerts per 100 km. The only statistically significant relationship that resulted (r = -.26, p < .05) indicated that the more total FCW alerts received, the less likely a driver was to feel comfortable performing additional tasks when driving with FCW.

5.5.3.3 Debriefing and Focus Group Comments Regarding Perceived Value of FCW

Drivers’ comments, excerpted from the focus group and debriefing sessions, provide a fuller understanding of the perceived value of FCW. Drivers had mixed responses when asked if they thought that use of FCW made them safer drivers. Some thought FCW use made no difference due to their years of driving experience while others said FCW made them less vulnerable to other drivers’ mistakes. The drivers who said that FCW made them safer drivers credited their increased safety to the ability of FCW to improve their alertness as well as reinforcing good driving habits.

When asked “Were there situations when you got an alert when you were not paying enough attention?”, drivers recalled instances when FCW alerted them to danger. Given the sharpness of their recall of these events, if FCW alerted drivers to a situation when they were at risk, due to driver distraction, they were likely to rate FCW positively. Because these kinds of events occur infrequently, many drivers did not experience them.

Some drivers said that they were troubled by FCW’s inconsistent threat detection. If drivers saw FCW fail to alert, or were not able to provoke an alert when they thought one was required, their confidence in FCW’s safety was undermined due to its perceived unreliability.

Focus group drivers described situations when they did not get an alert when they felt it was required such as merging traffic during the rush hour, approaching a truck, and motorcycle pulling out from a side road.

Focus group participants generally agreed that FCW will reduce the harm caused by rear-end crashes when it becomes a product. They identified reasons why FCW will be helpful to reduce rear-end crashes such as counteracting driver distraction and age-related slowing of reaction time.

Drivers were asked if they experienced situations when FCW operated in a way they did not understand or was opposite of what they expected. Some drivers mentioned situations when they did not understand FCW. Typically these situations involved false, late, and unexpected alarms; FCW failing to operate properly; and FCW malfunction. However, when asked this, some drivers responded that they felt safer driving using FCW.

Drivers described situations when they did not understand FCW information due to issues with FCW system characteristics, false messages, and driving feedback.

5.5.3.4 Summary

Many of the FCW perceived value measures were significantly intercorrelated, which indicates that the perceived value objective was largely internally consistent. Drivers provided generally positive ratings regarding FCW safety and understandability, but overall satisfaction was somewhat variable. Slightly less than half the drivers expressed high satisfaction with FCW (6 or 7), over one third of the drivers expressed a neutral-level of satisfaction, and 16 percent were dissatisfied.

Driver age did not statistically differentiate ratings on the perceived value measures, with the exception of younger drivers having found it less difficult than the middle-age group to maintain a safe distance to the preceding vehicle using FCW. Generally, the middle-age group of drivers expressed a wide range of ratings for the FCW perceived value measures, which may have been related to their varied experiences with the system during daily trips on local roads, in that they used ACC less than the older drivers did.

For this reason, it is recommended that future analyses explore how differential driver use of ACC is related to the evaluation of FCW. Drivers reported that FCW was valuable to the degree that it helped them to maintain alertness and counteract distraction. In fact, the more FCW alerts that were received, the less comfort drivers expressed regarding performing additional tasks while driving, including those that are potentially distracting, such as using a cell phone. Anecdotally, drivers who experienced an FCW alert while distracted appeared to clearly recognize FCW benefits.

5.5.4 Perceived Value – ACC

The perceived value objective examined driver satisfaction with the ACC system, awareness of safety, compatibility with an individual’s mental model, and driving skill enhancement. Both descriptive and quantitative discussion of perceived value is offered in this section, as well as driver comments as provided in debrief and focus group sessions and a concluding summary.

5.5.4.1 ACC Statistical Findings

Nine survey items were used to measure perceived value for ACC. Table K-1, as found in Appendix K, provides the correlations among the perceived value measures and identifies the significantly intercorrelated items. One of the nine items referred to manual driving in isolation and was therefore used for comparative purposes only. Of the remaining eight items, 25 of the 28 resulted in significant findings in the expected direction, suggesting that the chosen measures were internally consistent in their assessment of the construct of perceived value.

Across the entire FOT sample, opinions regarding the perceived value of the ACC system were generally positive. Table 5-11 displays the measures of central tendency and standard deviations for each measure. Of the nine items, the measure that addressed the degree of concern regarding the traffic behind the driver when using ACC resulted in the least positive overall attitude. The mean for this item was 4.0, with the most prevalent response, a score of 3, indicating a greater level of concern among the largest number of drivers in the sample.

Table 5-11. Perceived Value Sub-Objective Measure Descriptive Statistics

Sub-objective	Survey Item	Mean	Standard Deviation	Median	Mode
Overall
	Overall how satisfied were you with the ACC system?	6.0	1.1	6.0	7.0
	1 (very unsatisfied) - 7
C ompatibility with mental model
	Overall, I felt the operation of the ACC system was predictable.	5.9	1.1	6.0	6.0
	1 (strongly disagree) - 7
	When I was using ACC, I understood when I had to take control - either by accelerating or braking	6.5	0.9	7.0	7.0
	1 (strongly disagree) - 7
	How distracting did you find the ACC system operation (e.g., automatic acceleration and deceleration or warnings)?	5.4	1.7	6.0	7.0
	1 (very distracting) - 7
Driving skill enhancement
	Did you feel more comfortable performing additional tasks, (e.g., adjusting the heater, operating the radio, talking on a cellular telephone, etc.) while using the FCW system as compared to manual driving?	5.6	1.2	6.0	6.0
	1 (less comfortable) - 7
Safety
	How safe did you feel while driving the car using ACC?	6.0	1.3	6.0	7.0
	1 (very unsafe) - 7
	How safe did you feel driving the car manually?	6.7	0.6	7.0	7.0
	1 (very unsafe) - 7
	When using ACC, do you feel you drove more or less safely than when driving manually?	5.5	1.4	6.0	7.0
	1 (less safe) - 7
	Relative to manual driving, how concerned were you about the traffic behind you when using ACC?	4.0	1.9	4.0	3.0
	1 (much more concerned) - 7
	Overall, do you think that ACC is going to increase your driving safety?	5.5	1.5	6.0	6.0
	1 (strongly disagree) - 7

As depicted in Figure 5-40, the scores for overall satisfaction with ACC were mostly positive. More than three-quarters of the drivers gave satisfaction ratings of 6 or 7. By contrast, less than 25 percent expressed a more neutral opinion (values 3, 4, or 5) and no drivers reported being unsatisfied with the system (scores of 1 or 2). A positive level of overall satisfaction with the ACC is reflected in the mean score of 6.0 for this item and a mode of 7.

Figure 5-40. Overall, How Satisfied Were You with the ACC System?

[d]

With regard to the degree to which ACC operation matched drivers’ expectations (i.e., mental model) for how such a system should operate, responses indicated that, overall, drivers found ACC predictable (see Figure 5-41). Nearly three-quarters of respondents (71%) indicated that they strongly agreed (scores of 6 or 7) that ACC operation was predictable and no one strongly disagreed (scores of 1 or 2) with this statement.

Figure 5-41. Overall, I Felt the Operation of the ACC System was Predictable

[d]

Perhaps as a result of finding ACC operation predictable, there were cases when drivers understood when they had to take control and override the system (see Figure 5-42). The large majority, 90 percent, of the drivers indicated that it was quite obvious when a system override was necessary (scores of 6 or 7).

Figure 5-42. When using ACC I Understood When I Had to Take Control, Either by Accelerating or Braking

[d]

Drivers were asked to rate how distracting they found ACC operation in terms of its automatic acceleration and deceleration. As shown in Figure 5-43, close to two-thirds of the sample did not find this function distracting and responded at the positive end of the scale (scores of 6 or 7). However, it is important to note that approximately 22 percent of drivers responded in the negative range (scores of 2 or 3), suggesting that they found aspects of the automatic acceleration and braking distracting to their driving.

Figure 5-43. How Distracting Did You Find the ACC System Operation?

[d]

Because ACC provides support to the driving task, individuals may have believed that they could perform additional actions safely while driving using the system. Drivers were asked if they felt more comfortable performing additional tasks, such as talking on a cell phone or adjusting the heater, when they drove using ACC, compared to manually. The mean score was 5.6. Figure 5-44 shows that 61 percent of drivers responded at the positive end of the scale (rating of 6 or 7), indicating that they were more comfortable performing additional tasks while driving using ACC. Importantly, no drivers indicated that they felt less comfortable (rating of 1 or 2), as compared to manual operations. However, 26 percent responded neutrally (4) that they were neither more, nor less comfortable performing additional tasks using ACC.

Figure 5-44. Did You Feel more Comfortable Performing Additional Tasks While Using the ACC System, as Compared to Manual Driving?

[d]

With respect to a comparison of manual versus ACC driving safety, respondents were asked the same survey question for each driving situation. A comparison of responses as provided in Figure 5-45 indicates that in both cases, the vast majority (from 74% - 95%) of all drivers felt that they were quite safe (scores of 6 or 7) on the road. However, 27 percent more drivers felt “very safe” driving manually, compared to driving with ACC.

Figure 5-45. Comparison of How Safe Drivers Felt Driving the Car Using ACC versus Driving the car Manually

[d]

Using a different approach, a related survey item asked drivers the degree to which they felt more or less safe using ACC, as compared to manual driving. As depicted in Figure 5-46, over half of the sample (55%) responded at the positive end of the scale (scores of 6 or 7), indicating that they felt safer driving with ACC. However, the mean score was only somewhat positive, at 5.5, as 43 percent of drivers responded in the neutral range of the scale (scores of 3, 4, or 5). Nevertheless, only a very small percentage (2%) responded at the “less safe” end of the scale (scores of 1 or 2).

Figure 5-46. When Using ACC, Do You Feel You Drove More or Less Safely
than Manual Driving?

[d]

In another safety-related measure, drivers rated their concern about the traffic behind them when using ACC as compared to manual driving. As depicted in Figure 5-47, 45 percent of the sample expressed some level of concern about the traffic behind them when they were using ACC and 12 percent were much more concerned than when driving manually. It should be noted that one fifth of the respondents said that they were much less concerned about the traffic behind them than as compared to when they drove manually.

Figure 5-47. Relative to Manual Driving, How Concerned were You About the Traffic Behind You when Using ACC?

[d]

Finally, drivers were asked, overall, if they believed that use of ACC would increase their driving safety. Nearly two-thirds of the drivers responded using the positive end of the scale (scores of 6 or 7) as shown in Figure 5-48; however, the mean was quite neutral at 4.6. In fact, almost one-third of the sample responded in the neutral range (scores of 3, 4, or 5). Five percent of drivers did not agree that using ACC would improve their driving safety (scores of 1 or 2).

Figure 5-48. Overall, do You Think that ACC is Going to Increase Your Driving Safety?

[d]

As previously noted, driver acceptance analyses were narrowed to a focus on potential age group differences. ANOVA was used to verify the nature of any statistical relationships. Table 5-12 reports the statistical relationships among driver age groups and attitudes regarding perceived value, as broken down by sub-objective, with any significant group differences noted briefly in the rightmost “Results” column and nonsignificant results denoted using “NS.”

Driver age group was significantly related to five of the ACC perceived value measures; i.e., overall satisfaction and safety, the degree of system predictability and level of distraction related to ACC features, such as automatic acceleration and deceleration, and a comparison of ACC versus manual driving with regard to level of driving safety. Each statistical finding is depicted graphically and discussed below.

Table 5-12. Perceived Value Sub-Objective Measures by Driver Age Group

Sub-objective	Survey Item	Age Group	Mean	ANOVA Results
Overall
	Overall how satisfied were you with the ACC system?	Younger	5.7	O more satisfied than Y and M
1 (very unsatisfied) - 7		Middle	5.6
		Older	6.6
C ompatibility with mental model
	Overall, I felt the operation of the ACC system was predictable.	Younger	5.5	O agreed more so than Y
	1 (strongly disagree) - 7	Middle	5.9
		Older	6.5
	When I was using ACC, I understood when I had to take control - either by accelerating or braking	Younger	6.6	NS
	1 (strongly disagree) - 7	Middle	6.3
		Older	6.8
	How distracting did you find the ACC system operation (e.g., automatic acceleration and deceleration or warnings)?	Younger	4.8	O found less distracting than Y
1 (very distracting) - 7		Middle	5.1
		Older	6.2
Driving skill enhancement
	Did you feel more comfortable performing additional tasks, (e.g., adjusting the heater, operating the radio, talking on a cellular telephone, etc.) while using the FCW system as compared to manual driving?	Younger	5.8	NS
		Middle	5.4
		Older	5.7
Safety
	How safe did you feel while driving the car using ACC?	Younger	5.6	NS
	1 (very unsafe) - 7	Middle	5.8
		Older	6.5
	How safe did you feel driving the car manually?	Younger	6.5	NS
	1 (very unsafe) - 7	Middle	6.7
		Older	6.8
	When using ACC, do you feel you drove more or less safely than when driving manually?	Younger	5.3	O felt drove more safely than M, when using ACC
	1 (less safe) -7	Middle	5.0
		Older	6.1
	Relative to manual driving, how concerned were you about the traffic behind you when using ACC?	Younger	4.2	NS
	1 (much more concerned) - 7	Middle	3.8
		Older	4.0
	Overall, do you think that ACC is going to increase your driving safety?	Younger	5.3	O agreed more strongly than M
	1 (strongly disagree) - 7	Middle	5.0
		Older	6.2

Figure 5-49 depicts the overall satisfaction levels for ACC by age group. Drivers in the older age group reported greater satisfaction with the ACC system than drivers in the middle and younger age groups (F(2, 63) = 5.49, p = .01; and F(2, 63) = 5.49, p = .02, respectively). In fact, 73 percent indicated that they were very satisfied with the system (7).

Figure 5-49. Overall Satisfaction with the ACC System by Driver Age Group

[d]

As presented in Figure 5-50, drivers in the older age group were significantly more likely to agree that the ACC operated in a predictable manner than drivers in the younger age group (F(2, 55) = 3.45, p = .04), suggesting that the system matched the mental model of the older drivers more so than those who were younger.

Figure 5-50. Overall ACC System Predictability by Driver Age Group

[d]

Results pertaining to the level of distraction experienced by drivers as related to ACC features, such as automatic acceleration and deceleration, are depicted in Figure 5-51. This figure indicates that older drivers found ACC system operation significantly less distracting than drivers in the younger age group (F(2, 55) = 3.45, p = .03). Older drivers who found that ACC was not at all distracting in its operation (59% with a score of 7) made up the largest percentage of responses.

Figure 5-51. How Distracting Did You Find the ACC System Operation by Driver Age Group

[d]

Figure 5-52 presents results for the survey item that asked respondents to compare how safe they felt driving using ACC, compared to manual driving. A significant difference existed, in that older drivers reported feeling more safe than those in the middle-age group (F(2, 63) = 3.41, p = .04). More than half of the sample of older drivers (55%) reported feeling safer (7), compared to 9 percent of the middle-age group who responded at the most positive end of the scale.

Figure 5-52. Comparison of How Safe Drivers Felt Driving the Car Using ACC versus Driving the Car Manually by Driver Age Group

[d]

In an additional item that addressed overall ACC system safety, Figure 5-53 depicts the significant finding that older drivers were more likely to agree than those in the middle-age group that using ACC would increase their driving safety (F(2, 63) = 3.56, p = .03). Older drivers represented the largest percentage of responses for any scale value, with 55 percent indicating that they strongly agreed (score of 7) that the ACC system would improve their driving safety.

Figure 5-53. Overall Belief that ACC would Increase Driving Safety by Driver Age Group

[d]

5.5.4.2 Investigating Travel Behavior and ACC Perceived Value Measures

Follow-on analyses were conducted to determine the degree to which objective travel behavior measures were correlated with the perceived value survey measures. The objective measures that encompassed ACAS-enabled driving and ACC use included ACC distance traveled (km) in valid trips and number of alerts with ACC engaged. However, as the distribution for number of alerts with ACC engaged contained over 50 percent of drivers with zero alerts, this correlation was not statistically sound. Using a related measure in its stead, there was no relationship between the mean number of alerts per 100 km during all ACAS-enabled driving and the ACC perceived value measures.

Significant, positive correlations existed between total distance traveled in ACC mode and the following variables: the degree to which individuals found ACC predictable (r = .26, p = .03), level of safety felt while driving using ACC (r = .32, p = .01), and overall attitudes regarding the degree to which ACC would increase driving safety (r = .29, p = .02). Each relationship was in the expected direction, in that greater distances traveled using ACC were associated with a greater degree of feeling that the system was predictable and safe.

5.5.4.3 Debriefing and Focus Group Comments Regarding Perceived Value of ACC

Drivers’ comments excerpted from focus groups and debriefing sessions provide a fuller understanding of the perceived value of ACC. Drivers were almost unanimous in endorsing ACC use.

Drivers liked the way that ACC made their vehicle resume its speed when the preceding car was out of the way.

Another feature drivers liked about ACC was its ability to maintain the desired speed.

The enthusiasm extended to drivers who admitted they had made little prior use of conventional cruise control. When asked to compare ACC with conventional cruise control, drivers preferred ACC to CCC.

Drivers were asked in what traffic conditions they would use ACC. Some drivers said they would use, or try to use, ACC in almost all traffic conditions. Other drivers said that they did not want to use ACC in heavy congested traffic or inclement weather. Some drivers mentioned special uses for ACC such as maintaining speed in areas where there are police traffic enforcement traps.

Drivers were asked if they thought using ACC made them safer drivers. Several drivers anticipated that ACC use would reduce road rage incidents in heavy traffic. Overall, drivers agreed that ACC made them safer drivers.

Drivers had conflicting opinions as to whether they thought that ACC use would reduce harm. Some drivers remembered that they were instructed not to use ACC in congested local traffic yet thought that rear-end crashes are more likely in these conditions. Drivers were confused as to where there are risks of rear-end crashes versus the appropriate locations using ACC.

5.5.4.4 Summary

Almost all of the ACC perceived value measures were significantly intercorrelated, which indicates that the objective was largely internally consistent. Drivers were quite satisfied with ACC overall and reported that they felt quite safe using it, though were somewhat less convinced that they drove more safely using ACC as compared to manually. As a whole, drivers also rated ACC highly with regard to understanding when overriding system acceleration and/or deceleration was necessary. While still positive, drivers did rate ACC slightly lower on predictability, and the degree to which system functioning was distracting and would increase driving safety.

Almost half of the sample reported some concern as to the degree to which the traffic behind would understand the operation of an ACC-equipped vehicle. This points to the issue of driver expectations regarding vehicle actions, as road safety requires a common set of behavioral expectations that form over time in the case of implementing new or emerging vehicle technologies.

Driver age was related to ratings on some of the perceived value measures. The older driver age group reported a greater degree of overall satisfaction with ACC than both the younger and middle-age groups. Additionally, older drivers agreed that ACC operation was more predictable and were less distracted by ACC than the younger driver group. Finally, in comparison to the middle-age group, older drivers felt that they drove more safely using ACC, as compared to manually, and agreed more strongly that ACC use would increase their driving safety.

With regard to trip behaviors, travel using ACC was related to the level of safety felt while driving using ACC, the degree to which individuals felt the system was predictable, and would increase driving safety. Generally, with increased use, it appears that drivers tended to become more positive toward the value of various aspects of the ACC system.

5.5.5 Ease of Use – FCW and ACC

Examination of the ease of ACAS use explored the degree to which drivers found the system easy to set up, understand, adjust, and use. Ease of use considerations were specified with regard to several sub-objectives. These included how drivers rated ACAS in comparison to conventional in-vehicle systems, what demands system use placed on drivers, how drivers used ACAS, their understanding and regard for warnings and nuisance alerts (FCW), as well as overall usability, including the HUD.

5.5.5.1 FCW Statistical Findings

This section presents the results associated with analyses performed for the FCW ease of use measures. It includes a descriptive and quantitative discussion of the ease of use survey items. Driver comments as obtained during focus group and driver debriefing sessions are provided to give a fuller understanding of how drivers assessed FCW ease of use. This section concludes with a summary of findings.

As appropriate, correlations were calculated among the FCW ease of use measures by sub-objective and are found in Appendix L. Relationships for Likert scale items, where significant, were in the expected direction.

Drivers responded to survey items that assessed attitudes regarding FCW ease of use. Appendix M includes tables with descriptive statistics for responses to each of the survey measures, broken down by sub-objective. As responses were not always normally distributed, measures of central tendency, in addition to the mean and standard deviation, are provided for each measure, where appropriate. For items that elicited dichotomous responses, the percent of “yes” replies to the measure is provided.

As indicated in the methods section of this report, driver acceptance analyses were targeted where there were differences among age groups and/or age and gender groups, as appropriate. Where statistically significant, meaningful findings that differentiated groups on FCW ease of use are discussed. For reference purposes, tables containing mean responses by age group are presented in parallel with the overall descriptive statistics for all survey measures in Appendix M. Significant group differences are noted briefly in the rightmost “Results” column, while nonsignificant results are denoted using “NS.” Analysis of Variance was performed for Likert-type survey measures and Chi square analyses were used for dichotomous measures.

Compared to conventional safety systems, such as ABS and airbags, ratings for FCW were neutral (i.e., neither better, nor worse; mean score = 3.9). Moreover, there were no significant between group differences among the age groups.

Additional demands on drivers, necessitated by interacting with the FCW system, were assessed. A survey item asked participants how much stress they felt while using FCW compared to manual driving. The mean score for all drivers was 4.7, indicating a slight tendency toward less stress using FCW over manual driving. Figure 5-54 depicts a percent distribution of scores for this item, overall and by age group. Almost one half of the drivers, 49 percent, reported less stress driving with FCW (scores of 5 – 7), while one quarter of the drivers reported more stress (scores of 1 – 3). ANOVA results indicated that the younger driver age group was more stressed than the older drivers by FCW, compared to manual driving (F(2, 55) = 4.66, p = .01).

Figure 5-54. Did You Experience More or Less Stress When Driving With FCW as Compared to Manual Driving by Driver Age Group?

[d]

Additionally, drivers were asked if the visual display of FCW cautionary alerts was distracting. The mean was 5.4, indicating that the visual cautionary alerts were not considered extremely distracting, however there were a wide range of responses to this item.

Figure 5-55 shows the percent age distribution of drivers’ scores overall and by age group. Sixty-nine percent replied that the cautionary visual alerts were toward the “not at all distracting” end of the item response scale (scores of 5 – 7), while 19 percent of respondents fell at the “distracting” end of the scale (scores of 1 – 3). In a comparison of the age groups, younger drivers reported finding the visual alerts more distracting than did the older driver age group (F(2, 55) = 4.99, p < .01). However, it is important to note that the mean for the younger group (4.5) was quite neutral. The mean score for the older driver group (6.2) suggests that older drivers did not view the visual FCW cautionary alerts as very distracting.

Figure 5-55. How Distracting were the Visual Alerts that Signaled a Cautionary Situation by Driver Age Group?

[d]

Drivers were asked a series of questions to identify what types of driving conditions precipitated adjustments to the FCW alert timing. Traffic conditions were a reason for changing FCW settings by the largest percentage of “yes” responses by drivers, at 77 percent (see Figure 5-56). Weather conditions were the next most frequent reason for FCW alert timing adjustments, 41 percent, with smaller percentages of respondents indicating that being in a rush, tired, or concerned about alertness warranted a change.

Figure 5-56. Percent “Yes” Responses to FCW Alert Timing Changes by Conditions, Overall and by Driver Age Group

[d]

Use patterns for the FCW system were assessed by measuring driver comfort utilizing FCW in adverse weather conditions. More than 10 percent of drivers never experienced system operations in poor weather. However, for those who did drive in adverse weather, the mean response to a seven-point scale (1 = very uncomfortable) was 5.5, suggesting that drivers felt moderately comfortable using FCW. Age was not found to significantly differentiate between groups on this measure.

With the knowledge that FCW issued false imminent warnings, it was important to address the level of tolerance drivers exhibited toward what could be considered “nuisance” alerts. Figure 5-57 depicts response distributions for the entire sample and by age group for the measure assessing overall annoyance regarding alerts that were deemed “unnecessary.” Overall, 27 percent of the drivers reported that they were “not at all” annoyed by unnecessary FCW alerts, while slightly more than one third, 35 percent, of the drivers reported marked annoyance. The mean response for this item was 3.4, indicating that, on average, attitudes toward unnecessary alerts fell between “tolerable” and “slightly annoying.” Using ANOVA to differentiate among the age groups, results indicated that the younger and middle-age drivers reported being more annoyed by what they deemed as unnecessary FCW alerts, compared to the older drivers (F(2, 60) = 17.25, p < .01; F(2, 60) = 17.25, p < .01). Mean scores by group were 2.6 for younger drivers, 3.1 for middle-age drivers, and 4.4 for older drivers.

Figure 5-57. Overall, Indicate the Annoyance Level Associated with Unnecessary FCW Alerts Overall and by Driver Age Group

[d]

The degree to which drivers felt “annoyed” by various driving situations that could have resulted in unnecessary FCW alerts was also investigated (1 = unacceptably annoying – 5). Drivers rated as “only slightly annoying” unnecessary FCW alerts resulting from when they cut in behind another vehicle (mean = 4.0) or changed lanes (mean = 4.0). The highest annoyance ratings were associated with passing a sign, light post, or guardrail, though the mean score of 3.1 indicated that, in actuality, drivers as a whole found even such alerts “tolerable.”

Analyses were also performed to determine whether age differentiated among annoyance ratings for the various driving situations. There was no significant between-age group difference in responses to passing a parked vehicle. However, age did differentiate attitudes with regard to the remaining seven scenarios. Figure 5-58 and Figure 5-59 depict mean annoyance ratings by age group where there were significant differences. In four of the scenarios, both younger and middle-age driver groups were significantly more annoyed than the older drivers. These included “when a vehicle ahead of me turned” (F(2, 60) = 11.62, p < .01); “when a vehicle ahead changed lanes” (F(2, 60) = 10.16, p < .01); “when a vehicle cut in front of me” (F(2, 60) = 5.70, p < .05); and “when my vehicle changed lanes” (F(2, 60) = 6.71, p < .05). For two of the remaining scenarios, younger drivers were significantly more annoyed than the older drivers. These included “when I passed a moving vehicle” (F(2, 60) = 6.49, p < .01) and “ when I cut in behind another vehicle” (F(2, 60) = 5.85, p < .01). Finally, the middle-age group reported greater annoyance than older drivers with regard to false FCW alerts associated with signs, light posts, and guard rails (F(2, 60) = 3.89, p < .05).

Figure 5-58. Mean Annoyance with Unnecessary FCW Alerts Overall and for Lead Vehicle Scenarios by Driver Age Group

[d]

Figure 5-59. Mean Annoyance With Unnecessary FCW Alerts for Host Vehicle Scenarios by Driver Age Group

[d]

The understanding of warnings sub-objective addressed driver comprehension of FCW warnings. The goal of this objective was to assess driver self-reported ability to recognize and discriminate various features of the FCW alerts while driving. Data pertaining to objective components of the alerts are addressed in the safety benefits section of this report.

Briefly, the FCW system functioned by generating imminent crash-warning icons on the HUD, accompanied by an auditory alert to warn drivers that they were too close to a lead vehicle and should apply the brake. As further elucidated in the section on ease of learning, there is evidence that some drivers did not understand how FCW functioned, even at the culmination of their FOT participation. Given the fact that the system was not always fully comprehended or intuitive for drivers to use, it is important to determine the degree to which various aspects of the warnings were acceptable to and understood by the user.

In the case of visual alerts, drivers were asked how well they could identify a warning for a cautionary situation versus an imminent threat. The mean response to this item was 6.3 on a scale of 1 (not well at all) – 7, suggesting that drivers did not have a problem discriminating between cautionary and imminent alerts. There was no significant difference between age groups for this measure.

With regard to alert triggers, drivers were asked how often FCW provided an alert where the source could not be determined. As depicted in Figure 5-60, nearly 40 percent of drivers reported that they could not identify the source of a FCW alert once or twice, while 29 percent reported receiving from three to twenty such alerts. One-third of the drivers indicated that they felt that they were always able to identify the source of the FCW alert.

Figure 5-60. Rating of How Often Drivers Could not Identify Source of FCW Alert

[d]

Disaggregating these data by age and gender, as shown in Figure 5-61, indicates that the majority of the older males, 55 percent, and the older females, 64 percent, felt that they could always identify the source of an FCW alert. In contrast, only 18 percent of the younger and middle-age males and the middle-age females responded that they felt as though they could always identify the source of the FCW alert.

Figure 5-61. Rating of How Often Could Not Identify Source of FCW Alert by Driver Age Group and Gender

[d]

Drivers were also questioned regarding the degree to which they found the FCW auditory alert startling. Although the mean response for this item was 4.6, slightly less startling than neutral on the 7-point scale, Figure 5-62 depicts a wide distribution of scale scores. Over one-quarter of the drivers, 26 percent, responded that the auditory alert was not at all startling, whereas 7 percent indicated that the alert was very startling. Driver age group differentiated these responses statistically. Younger drivers rated the auditory alert significantly more startling (mean = 3.6) than the older drivers (mean = 5.4; F(2, 55) = 3.92, p = .02).

Figure 5-62. How Startling did You Find the Auditory Alert when it Occurred Overall and by Driver Age Group?

[d]

The remaining measures that comprised the understanding of warnings sub-objective addressed the effectiveness of the visual and auditory alerts, as well as the use of color for the alert icons. None of these items were differentiated statistically among driver age groups. For drivers as a whole, with regard to the degree to which using color improved the understanding of the FCW information presented in the HUD, the overall mean response was 6.2 (7 = strongly agree), indicating that the use of color was deemed beneficial. In terms of the effectiveness of the alerts (7 = very effective), overall mean responses were each nearing the positive scale anchor. As it pertained to the visual component, drivers indicated that the warnings were effective at getting their attention quickly (mean = 6.2). Regarding the audio component, drivers responded similarly as to the effectiveness of the audio alert in communicating imminent threats (mean = 6.2) and getting their attention quickly (mean = 6.5).

The usability sub-objective gauged aspects of driver comfort level and ease of adjusting and driving with the FCW system. For two items, responses differed statistically by age group. These pertained to the degree of annoyance associated with the imminent visual and auditory alerts (see Figure 5-63 and Figure 5-64). Overall, the imminent visual and auditory alerts were rated as tolerable, slightly annoying, or not at all annoying by 76 percent and 71 percent of drivers, respectively. As depicted in Figure 5-63, among the age groups, younger and middle-age drivers reported being significantly more annoyed by the imminent visual alert than older drivers (F(2, 35) = 4.77, p < .05), though their degree of annoyance was moderate (mean = 3.2 for both younger and middle-age groups, compared to the older group mean = 4.8).

Figure 5-63. “How Annoying Was the Visual Alert That Signaled a Situation in Which You May Be About to Crash?” Overall and by Driver Age Group

[d]

As shown in Figure 5-64, when drivers were questioned regarding the degree to which they found the auditory alert indicating an imminent crash risk situation annoying, nearly 30 percent responded that the alert was either somewhat or unacceptably annoying. Again in this case, younger (mean = 2.9) and middle (mean = 3.4) age drivers were significantly more annoyed by the FCW auditory alert for imminent crash situations, finding them in the range of “tolerable”, compared to the older drivers (mean = 5.0; F(2, 35) = 6.06, p < .05).

Figure 5-64. “How Annoying Was the Auditory Alert That Signaled a Situation in Which You May Be About to Crash?” Overall and by Driver Age Group

[d]

Additional usability measures that did not differentiate age groups, but were nevertheless of interest, included an item assessing driver comfort level using FCW (1 = very uncomfortable – 7), which resulted in an overall mean score of 5.5, indicating a moderate level of comfort with FCW use across the sample. Regarding the amount of time it took for participants to reach this level of comfort, the mean response for the sample was 2.0 indicating that, on average, drivers took 2-3 days. Finally, with regard to the usability of the alert timing adjustment, on average, drivers reported that it was easy to understand and use (7 = very easy; sample mean = 6.3). Additionally, drivers indicated that they changed the alert timing adjustment on average a bit more frequently than 2-3 times over the three weeks FCW was enabled (5 = I changed the setting every day; sample mean = 3.3).

5.5.5.2 HUD Statistical Findings

Drivers were asked which aspects of the HUD they would prefer to see moved to the head-down instrument panel, or have visible only when adjustments were made, rather than being continually displayed. The majority of drivers did not recommend any change to the HUD in terms of removing information and displaying it in the head-down instrument panel. As it pertained to displaying information only while adjustments were being made, 46 percent suggested displaying the ACC gap setting/headway only during adjustments, 26 percent would have preferred that the FCW alert timing setting was visible only while being adjusted, and 14 percent indicated that only displaying the ACC set speed during adjustments would have been adequate. Additionally, drivers were asked how frequently they intentionally adjusted the HUD location to hide the information display and drove with it in that position for an extended period. The majority of drivers, 83 percent, stated that they never adjusted the HUD in this way, while 3 percent of the sample responded that they adjusted the HUD “very frequently” in this manner.

5.5.5.3 Investigating Travel Behavior and FCW Ease of Use Measures

Follow-on analyses were conducted to see if objective travel behavior measures were correlated with the subjective Likert-type ease of use survey measures. The objective measures encompassed ACAS-enabled driving and included FCW distance traveled (km) in valid trips, number of FCW alerts and mean number of FCW alerts per 100 Km. No meaningful statistical relationships were obtained among the survey and objective travel behavior measures.

5.5.5.4 Interpretation of Debriefing and Focus Group Comments

Drivers’ ease of use related comments during the focus groups and driver debriefings are discussed to enhance the understanding of their rating of FCW’s ease of use.

Drivers, participating in focus groups after completing the FOT, discussed, “When you got the (FCW) imminent alert, what did you typically do in response to that? Did you apply the brakes?” They framed their answers in several ways. Some described how they handled the car mechanically, others described how their reaction to FCW evolved through time, and others classified their responses depending on whether they were attending to the forward scene or not.

Drivers associated the FCW crash-imminent alert with the audible sound and some, but not all, associated it with the large visual icon. A few drivers noticed that the audible alert is not unique and that other infrequently occurring alerts use the same sound, i.e., when ACC is no longer available because the vehicle speed is too slow. Using the same sound for multiple conditions may require the driver to devote more attention to discriminating the meaning of the sound as well as respond to it.

Some drivers used the color and size shifts in the icons to signal a change in risk. Drivers said that it took time to understand the imminent FCW alert because it happened infrequently, suggesting that they had a learning curve for FCW.

The comments raise a question about the utility of cautionary alerts as some drivers said they learned to ignore the “early ones.” Typically, most drivers hover their foot over the brake while assessing the forward scene to react to an imminent alert.

Some drivers mentioned that they needed time to learn how to use FCW.

Comments recorded during the focus groups and in the debriefings convey that some drivers felt shock when they received a crash-imminent alert because it happened so infrequently.

Drivers were asked how their response to imminent alerts changed through time, “do you think the way you responded differed depending on the scenario? or change with more experience?”. They said that they became more comfortable using FCW, they initially did not know what to expect and overcame this feeling, they grew to like and use the slow warning system more than the imminent alert itself, and they learned that to do something is response to an imminent alert. i.e. checking the forward scene, etc., and made more adjustments to the FCW settings for weather etc. over time.

Other drivers mentioned how they used FCW according to their situation. For example, some drivers said that, if they were distracted and FCW alerted them, they immediately applied the brake. However, if they were attending to the forward scene and saw a lead vehicle make a right turn and got an imminent FCW alert in response, they learned to pay no attention because the situation was safe.

Some drivers used FCW as instruction about their driving, commenting that FCW modified their driving. They viewed an alert as beneficial because it reminded them to monitor the road situation as well as remedy distraction.

If a driver disagreed with the threat being warned about, he/she had to identify the risk, if any, about which they were being warned. They found having to look for the source of the alert to be annoying because it required added effort.

During their debriefing, drivers made comments about their comfort using FCW. Drivers’ comfort was improved because FCW reinforced their need to maintain a forward view. Their comfort was also impaired by false warnings, FCW’s failure to detect threats, the time needed to get used to FCW, difficulty in testing FCW and finding a suitable FCW setting, obscure meaning of each setting option, distraction caused by the icons, misleading detection of non-threatening vehicles, and the late onset of FCW alerts at higher speeds.

During the debriefings, drivers were asked what annoyed them about FCW. Drivers said FCW warnings were annoying because they occurred in conjunction with false warnings. They were interrupted and had to identify a source for the alert and then regain their driving composure. Drivers found the combination of needless interruption, with its insistent intrusive sensory characteristics, annoying.

When drivers were asked what annoyed them about FCW using an open-ended response format, their answers can be categorized as FCW’s auditory tone; the size and color of the FCW icons; the unexpected, distracting, or startling character of the FCW alert; and the FCW timing, in terms of the lateness of the alert because, often, they were already reacting to the threat situation.

When asked about the HUD, drivers said that they liked seeing their vehicle speed on the HUD, even though the changing digits could be distracting. They also mentioned that the HUD was useful for night driving. Some commented that bright sunlight could wash out the HUD display. Other drivers mentioned being annoyed initially by the icons on the HUD but learned to overlook them with time.

5.5.5.5 Summary

Although ease of use ratings for FCW ranged from neutral to positive overall, the distributions of the ratings and their association with driver age are informative. Compared to manual driving, mean stress levels when using FCW were reported as neutral to slightly positive, however the ratings were widely distributed. One quarter of the drivers indicated that FCW use led them to experience more stress, compared to 49 percent reporting less stress.

With regard to the cautionary visual icons, the distribution of scores is also quite revealing. While, overall, drivers did not rate the alerts as very distracting, those who did (scores of 1 or 2) were part of the younger age group (20%, compared to 0% of middle and 5% of older drivers). Pertaining to the FCW alert timing settings, drivers reported, overall and with no significant age group differences, that traffic and weather conditions were the most frequent reasons for changing the setting. Age group did differentiate responses regarding level of annoyance with the incidence of false FCW alerts, however. While, as a group, over one quarter of the drivers reported being not at all annoyed by the false alerts, slightly more than one-third experienced marked annoyance. Further, the younger and middle-age groups expressed greater annoyance than the older drivers regarding alerts that were deemed unnecessary. Similarly, younger drivers tended to report greater annoyance by false FCW alerts in most driving situations, while older drivers tended to report being less annoyed. Overall, one third of the drivers said they were always able to identify the reason for the FCW imminent alert and older drivers were over-represented among those able to identify the source of the alert. Three fifths of the older drivers said they were always able to identify the source of the FCW imminent alert compared to just under one quarter of the younger drivers and less than one fifth of the middle-age drivers. With regard to the auditory component of the FCW alert, as a group, drivers did not find it to be overly startling, though younger drivers rated the alert more startling and also more annoying than did older drivers. Additionally, the younger and middle-age groups expressed greater annoyance by the imminent visual alert than older drivers.

Other findings included reports that drivers were generally in favor of the HUD implementation and the vast majority, 83 percent, never intentionally adjusted and drove with the HUD in a way that hid information in the display. Changes to the HUD that did receive some element of support included displaying the ACC gap/headway and FCW alert timing settings only when adjustments were being made.

Any nascent anecdotal reports of annoyance with FCW alerts appeared to be aggravated to the degree that they turned out to be false. These alerts interrupted drivers, obligated them to identify a source for the alert, and to subsequently regain their driving composure. Some drivers expressed that the combination of needless interruption, especially with its insistent, intrusive, sensory characteristics, was annoying.

5.5.5.6 Ease of Use – ACC

The ACC ease of use objective, similar to that for FCW, assessed the degree to which drivers found FCW easy to set up, understand, adjust, and use in various circumstances. For reference purposes, Appendix N – Appendix O present results for ACC ease of use measures in a form parallel to that used for FCW, where possible. This includes survey item intercorrelations, descriptive statistics, and ANOVA results for analyses investigating potential age group attitudinal differences. Data about drivers’ assessment of the ease of use of ACC are provided in the appendices cited. No detailed analyses are provided due to scope limitations.

5.5.6 Ease of Learning – FCW and ACC

The ease of learning objective assessed whether drivers were able to learn and retain knowledge regarding ACAS use. Ease of learning is an important aspect of driver acceptance of vehicle technologies, since a feature that is easy to learn and understand is more likely to be used appropriately and frequently over time. This objective encompassed both the effectiveness of the instructions and the time required to understand and become comfortable with its use.

5.5.6.1 FCW Statistical Findings

This section presents the results of the FCW ease of learning measures. It includes both descriptive and quantitative discussion of the subjective measures from the FOT surveys, as well as driver anecdotes supplied during debriefing and focus group sessions. Finally, a summary of the analysis of ease of learning FCW is offered.

Correlations were calculated among the ease of learning measures²⁰. The items, “How long did it take before you became comfortable with the operations of FCW?” and “How long did it take before your understood the operations of FCW?” were significantly intercorrelated in the expected direction (see Table 5-13), suggesting that drivers who more quickly understood the operation of FCW were also more likely to feel comfortable with the operation of FCW in less time.

Table 5-13. Ease of Learning Sub-Objective Survey Measure Intercorrelations
(Spearman’s rho)

	Sub-objective	Survey Item
			1.	2.	3.
1.	Time to learn	How long did it take before you became comfortable with the operations of FCW?	¾	.46	NS
		1 (comfortable with FCW within 1st day) - 5
2.		How long did it take before you understood the operation of FCW?		¾	NS
		1 (understood operations of FCW within 1st day) - 5
3.	Utility of instructions/training	How useful was the training video in understanding how to use ACC and FCW?			¾
		1 (not at all useful) - 7
Note: All correlations significant at p ≤ .05 except where denoted as nonsignificant (NS)

Table 5-14 presents descriptive statistics for responses to each of the survey measures, broken down by sub-objective. As responses were not always normally distributed, measures of central tendency, in addition to the mean and standard deviation, are provided for each measure.

Table 5-14. Ease of Learning Sub-Objective Survey Measure Descriptive Statistics

Sub-objective	Survey Item	Mean	Standard Deviation	Median	Mode
Time to learn
	How long did it take before you became comfortable with the operations of FCW?	2.0	1.0	2.0	2.0
	1 (comfortable with FCW within the first day) - 5
	How long did it take before you understood the operation of FCW?	1.5	0.8	1.0	1.0
	1 (understood operations of FCW within 1st day) - 5
Utility of instructions/ training
	How useful was the training video in understanding how to use ACC and FCW?	6.6	0.7	7.0	7.0
	1 (not at all useful) - 7

 

As indicated previously in the methods section of this report, driver acceptance analyses were targeted at existing differences among age groups and/or age and gender groups, as appropriate. Along those lines, Table 5-15 shows statistical relationships between driver age group and FCW ease of learning by sub-objective, with any significant group differences noted briefly in the rightmost “Results” column and nonsignificant findings denoted using “NS.”²¹

Table 5-15. Statistical Comparison of FCW Ease of Learning Sub-Objective Measures by Driver Age Group

Sub-objective	Survey Item	Age Group	Mean	ANOVA Results
Time to learn
	How long did it take before you became comfortable with the operations of FCW?	Younger Middle	1.8 2.0	NS
	1 (comfortable with FCW within 1st day) - 5	Older	2.1
	How long did it take before you understood the operation of FCW?	Younger Middle	1.6 1.3	NS
	1 (understood operations of FCW within 1st day) - 5	Older	1.6
Utility of instructions/ training
	How useful was the training video in understanding how to use ACC and FCW?	Younger Middle	6.5 6.5	NS
	1 (not at all useful) - 7	Older	6.7

ANOVA results of the survey items suggested that driver age group was not a factor with regard to the time required to become comfortable with the operation of FCW, nor with the time needed to understand FCW operation. Parallel analyses indicated that driver gender also did not yield significantly different responses on any of these measures.

Descriptively, as it pertains to the ease of use survey items above, it is interesting to note that within the first day of use, 60 percent of the drivers reported that they understood FCW. A smaller percentage of the sample, 35 percent, became comfortable with its operation within that same timeframe. Within 2 to 3 days, over three quarters of the sample, 76 percent, reported that they felt comfortable using FCW, however, 3 percent indicated that they never became comfortable with the operation of FCW, and one young male driver reported that he never understood FCW. Overall, drivers felt that the instructional video was very useful, as 89 percent rated this item with a score of 6 or 7.

5.5.6.2 Investigating Travel Behavior and FCW Ease of Learning Measures

Follow-on analyses were conducted to see if relevant objective FCW travel behavior measures were correlated with the ease of learning measures. The objective measures that encompassed ACAS-enabled driving and FCW use included FCW distance traveled (km) in valid trips and number of alerts with FCW engaged. A significant negative correlation existed between FCW distance traveled (km) in valid trips and the survey item, “How long did it take you before you felt comfortable with the operations of FCW?” (r = -.30, p < .05). A significant positive correlation existed between the number of alerts with FCW engaged and the same survey item (r = .28, p < .05). These relationships suggest that the greater overall distance traveled in FCW mode, the more quickly a driver felt comfortable using the FCW system. Additionally, higher numbers of FCW alerts, normalized for distance traveled, were associated with drivers needing more time to feel comfortable using FCW.

5.5.6.3 Debriefing and Focus Group Comments Regarding Ease of Learning of FCW

As a result of examining the qualitative data acquired from the debriefing and focus groups, it is evident that ease of learning of FCW may only be fully explained by incorporating the results of the survey measures with anecdotal findings.

It became apparent from comments during debriefings that, although the vast majority of drivers responded in the surveys that they understood how to use FCW in a short amount of time, many did not truly comprehend how FCW functioned. Drivers were asked during the debriefing about their understanding of the timing of the FCW imminent alert. Of the 49 drivers who answered this question, 41 percent replied incorrectly, stating that changing the FCW settings altered the timing of the imminent alert. Examining the responses by driver indicates that the frequency of erroneous interpretations of FCW imminent alert timing decreased among the later subjects, although it still occurred. This is most likely a result of the fact that FOT administrators revised their instructions to subjects during the FOT to emphasize that the imminent alert timing was fixed. Some comments by later drivers mentioned that this point was stressed to them, yet earlier participants did not have the same understanding of the system, which may have affected overall driver acceptance of FCW. Given that two-fifths of drivers’ responses indicated that they did not understand how FCW imminent alerts were triggered, individuals were likely to experience frustration resulting from the assumption that they were affecting the FCW imminent alert timing by altering FCW sensitivity settings.

As a result of using the FCW system, drivers reported anecdotes explaining in what ways they also learned about their driving behavior. Instances that referred to proper spacing in between the FOT and lead vehicles, in addition to a better understanding of the various stopping distances required to be safe given different travel speeds, and indications regarding how often, as drivers, they let their attention wander were all highlighted as important contributions that the FCW system made to participants’ driving behavior. FOT participants also described ways in which the FCW system fostered good driving behavior in terms of learning how to use the sensitivity settings by going through a process to find the most suitable setting for various driving conditions and their individual driving style. Participants also reported gaining a more complete understanding of what FCW alert feedbacks they could or should ignore and which they should attend to.

As a final indication of the learning drivers needed to understand the full ACAS system, it was not uncommon in the debriefings and focus groups for drivers to confuse FCW with ACC. Typically, such errors were corrected by the experimenter during a debriefing or focus group session; however, there is at least one documented case of a discussion that referred to FCW, where ACC function was being described.

5.5.6.4 Summary

Most drivers reported that they learned to use, and felt comfortable using, FCW very quickly. Driver age and gender were not related to the ease of learning measures. Travel behavior variables were related to how long it took drivers to become comfortable with FCW use. The more participants drove with FCW engaged, the sooner they reported feeling comfortable using FCW. Additionally, the more FCW alerts drivers received, as normalized by distance driven, the longer it took them to feel comfortable using FCW.

Anecdotal evidence provided by drivers suggested that it is important to distinguish between learning to use a system versus understanding a system. Debriefing comments indicated that 41% of drivers did not understand the FCW crash-imminent alert timing. This misunderstanding could have contributed to dissatisfaction with FCW to the extent that the system did not meet expectations, in that some drivers were not able to set the imminent alert timing as they believed they could. Additionally, some drivers reported gaining an unexpected benefit from their FCW use, as they felt that it provided them with an opportunity to learn about their driving.

5.5.6.5 Ease of Learning – ACC

The ACC ease of learning objective, similar to FCW, assessed the degree to which drivers were able to easily learn and retain knowledge regarding how to use the ACC system. For reference purposes, Table P-1 in Appendix P presents results for ACC ease of learning measures in a form parallel to that used for FCW, where possible. This includes survey item intercorrelations, descriptive statistics, and ANOVA results for analyses investigating potential age group attitudinal difference. Data about drivers’ assessment of the ease of learning to operate ACC are provided in the appendices cited. No detailed analyses are provided due to scope limitations.

5.5.7 Driving Performance – FCW and ACC

The driving performance objective assessed to what degree, and how, drivers adjusted their driving with respect to ACAS. Driver performance considerations were specified with regard to four sub-objectives²². These included awareness, which addressed driver vigilance; vehicle control inputs, to examine driver behavior with regard to adjusting ACAS settings; and trip patterns, to evaluate potential changes in travel behavior associated with ACAS-enabled driving.

Changes in driving performance were expected across the duration of the FOT, given incremental exposure to ACAS. Initial driving occurred with ACAS disabled and was segmented by the independent evaluation into two periods, P1 and P2 (median split of distance traveled per driver). Subsequent ACAS-enabled driving was also divided into two periods (P3 and P4) using a median split of distance traveled on a per-driver basis. Analysis of driving parameters given this breakdown is performed as a means of investigating changes in driving performance associated with ACAS exposure.

5.5.7.1 FCW Statistical Findings

This section presents the results of the ACAS driving performance measures. Driving performance was gauged using subjective measures from the FOT surveys, as well as objective data from the data acquisition system (DAS) and driver anecdotes supplied during debriefing and focus group sessions. This section concludes with a summary of the analysis of driving performance.

5.5.7.1.1 Awareness

Correlations were calculated for the subjective measures of the awareness sub-objective²³. Table Q-1 in Appendix Q contains the correlation matrix for the awareness sub-objective survey items. The significant associations revealed that drivers who, when using FCW, considered themselves more responsive to the actions of other vehicles, also assessed themselves as more aware of their driving situation and felt slightly more comfortable performing additional tasks while driving. Additionally, drivers who felt that they did not over-rely on FCW also deemed themselves more aware of the surrounding driving situation and more comfortable performing additional tasks while driving. The three items referring to manual driving were included for comparison with the measures of awareness while driving with FCW.

Table 5-16 presents descriptive statistics for responses to survey measures of the awareness sub-objective. As responses were not always normally distributed, measures of central tendency, in addition to the mean and standard deviation, are provided for each measure.

Table 5-16. Driving Performance Sub-Objective Survey Measure Descriptive Statistics

Survey Item	Mean	Standard Deviation	Median	Mode
Awareness
When using FCW, how responsive were you to the actions of other vehicles around you?	6.4	0.8	7.0	7.0
1 (very unresponsive) - 7
When driving manually, how responsive were you to the actions of vehicles around you?	6.3	0.8	6.0	7.0
1 (very unresponsive)- 7
Overall, I found myself relying too much on the FCW system	2.0	1.3	1.0	1.0
1 (strongly disagree) - 7
When using FCW, how aware were you of the driving situation (surrounding traffic, posted speed, traffic signals, etc)?	6.5	0.7	7.0	7.0
1 (very unaware) - 7
When driving manually, how aware were you of the driving situation (surrounding traffic, posted speed, traffic signals)	6.2	0.8	6.0	7.0
1 (very unaware) - 7
While using FCW, please tell us the number of times, if ever, you came close to experiencing a rear-end collision?	0.8	1.3	0.0	0.0
open-ended # response
While driving manually, please tell us the number of times, if ever, you came close to experiencing a rear-end collision?	0.4	1.3	0.0	0.0
open-ended # response
Did you feel more comfortable performing additional tasks while using the FCW system as compared to manual driving?	5.2	1.3	5.0	4.0
1 (less comfortable) - 7

As indicated previously in the methods section of this report, driver acceptance analyses were targeted at differences among age groups and/or age and gender groups, as appropriate. For reference purposes, Table 5-17 depicts statistical relationships between driver age group and FCW driving performance, by sub-objective, with significant group differences noted briefly in the rightmost Results column and nonsignificant findings denoted using NS. Analysis of variance²⁴ was performed for Likert-type survey measures were statistically significant, meaningful findings that differentiated groups on FCW driver performance are discussed in the below text.

Table 5-17. Statistical Comparison of FCW Driving Performance Sub-Objective Measures by Driver Age Group

Sub-objective	Survey Item	Age Group	Mean	ANOVA Results
A wareness
	When using FCW, how responsive were you to the actions of other vehicles around you?	Younger	6.2	O more responsive than Y and M
	1 (very unresponsive) - 7	Middle	6.1
		Older	6.9
	When driving manually, how responsive were you to the action of vehicles around you?	Younger	6.0	NS
	1 (very unresponsive)- 7	Middle	6.3
		Older	6.5
	Overall, I found myself relying too much on the FCW system	Younger	2.0	NS
	1 (strongly disagree) - 7	Middle	2.2
		Older	1.9
	When using FCW, how aware were you of the driving situation (surrounding traffic, posted speed, traffic signals, etc)?	Younger	6.2	O more aware than Y and M
	1 (very unaware) - 7	Middle	6.3
		Older	6.9
	When driving manually, how aware were you of the driving situation (surrounding traffic, posted speed, traffic signals)	Younger	6.0	NS
	1 (very unaware) - 7	Middle	6.1
		Older	6.5
	While using FCW, please tell us the number of times, if ever, you came close to experiencing a rear-end collision?	Younger	1.1	NS
	open-ended # response	Middle	0.9
		Older	0.5
	While driving manually, please tell us the number of times, if ever, you came close to experiencing a rear-end collision?	Younger	0.5	NS
	open-ended # response	Middle	0.6
		Older	0.1
	Did you feel more comfortable performing additional tasks while using the FCW system as compared to manual driving?	Younger	5.0	NS
	1 (less comfortable) - 7	Middle	5.0
		Older	5.5

The awareness sub-objective addressed driver responsiveness to the actions of surrounding vehicles in cases of FCW and also manual driving. In both situations, overall mean responses suggested that drivers felt quite responsive and no scores on this measure fell below the scale midpoint (see Figure 5-65). When driving using FCW, 88 percent of the sample rated themselves toward the “very responsive” end of the rating scale (score of 6 or 7). Similarly, 85 percent of drivers responded with scores of 6 or 7 for manual driving.

Figure 5-65. “How Responsive Were You to the Actions of Other Vehicles Around You?” Using FCW and During Manual Driving

[d]

Figure 5-66 depicts the distribution of scores and significant differences among age groups regarding responsiveness to surrounding vehicles while using FCW. Older drivers rated themselves as significantly more responsive to the actions of other vehicles than the younger or middle-age groups (F(2, 63) = 9.02, p < .01). The means for the three age groups were all at the positive end of the scale; the average score for the older driver age group was 6.9, middle-age drivers, 6.1, and younger drivers, 6.2.

Figure 5-66. “When Using FCW, How Responsive Were You to the Actions of Other Vehicles Around You?” by Driver Age Group

[d]

Drivers rated how aware they felt driving with FCW and manually, as shown in Figure 5-67. Mean responses to the two parallel items were similar, in that drivers regarded themselves as quite aware in both driving conditions and no responses fell below the scale midpoint.

Figure 5-67. “…How Aware Were You of the Driving Situation…?” Using FCW and During Manual Driving?

[d]

As illustrated in Figure 5-68, among age groups, older drivers considered themselves to be significantly more aware of the traffic situation using FCW, compared to younger and middle-age drivers (F(2, 63) = 7.93, p < .01). The mean score on this measure for the older driver age group was 6.9, compared to a mean of 6.3 for the middle-age group and 6.2 for the younger driver age group.

Figure 5-68. “When Using FCW, How Aware Were You of the Driving Situation?”
by Driver Age Group

[d]

Figure 5-69 shows the percent distribution of responses to parallel survey items asking drivers to estimate the number of times they felt that they came near to experiencing a rear-end collision driving with FCW and also when driving manually. Using FCW, the mean number of reported near rear-end collisions was 0.84 (SD = 1.3), compared to a mean of 0.41 (SD = 1.3) for manual driving. Age differences were nonsignificant, but were inversely related; the younger driver age group reported a mean of 1.1 near-collisions, compared to 0.9 for the middle-age group and 0.5 for the older age group.

Figure 5-69. Estimated Number of Times Drivers Came Close to Experiencing a Rear-End Collision Using FCW and During Manual Driving

[d]

Finally, while there was no significant difference in responses by age group to the item assessing the degree to which drivers found themselves relying too much on the FCW system, it is interesting to note that the sample as a whole felt quite strongly that they did not over-rely on FCW (mean score = 2.0; 1 = strongly disagree – 7).

5.5.7.1.2 Vehicle Control Inputs

Another aspect of driver performance is the manipulation of vehicle controls that were associated with ACAS. For example, frequency of use of the HUD position adjustment and brightness controls, in addition to frequency of manipulation of FCW sensitivity settings were obtained from the DAS in the FOT vehicle.

HUD Position Adjustments

As illustrated in Figure 5-70, the mean number of position adjustments to the HUD per 1,000 km decreased over time during ACAS-enabled driving. Collapsing across light and dark driving, comparing P3 and P4, HUD adjustments per 1,000 km traveled decreased from 6.5 to 1.5. A comparison of P3 and P4 driving in conjunction with time of day evidenced that the frequency of adjustments made during daylight decreased 80 percent, while adjustments during darkness decreased 78 percent. During ACAS-enabled P3 driving, individuals tended to adjust the HUD position nearly twice as frequently while driving at night, compared to day, and maintained the day-night differential in P4.

Figure 5-70. HUD Position Adjustments per 1,000 Km Traveled by FOT Segment and
Time of Day

[d]

Figure 5-71 and Figure 5-72 show that during P3 and P4 driving and light and dark periods, older and female drivers appeared more likely to adjust the HUD position. For each category of drivers, proportionally, the largest decrease in HUD position manipulation occurred for driving in the dark. Older drivers made 24.8 changes to the HUD position per 1,000 km driven in darkness, whereas the incidence decreased to 2.6 in P4.

Figure 5-71. HUD Position Adjustments per 1,000 Km Traveled by FOT Segment and Time of Day by Driver Age Group

[d]

As depicted in Figure 5-72, female drivers also markedly reduced the number of changes they made to the position of the HUD, particularly while driving in darkness, from 16 to 3.6 per 1,000 km traveled.

Figure 5-72. HUD Position Adjustments per 1,000 Km Traveled by FOT Segment and Time of Day by Gender

[d]

HUD Brightness Adjustments

As illustrated in Figure 5-73, descriptive comparison indicates that HUD brightness adjustments were made more frequently during night driving than during the day and that the mean number of manipulations to HUD brightness settings decreased over time during ACAS-enabled driving. Overall, per 1,000 km, drivers made an average of 15.3 brightness adjustments in P3, evidencing a reduction in frequency to 11 in P4, a 28 percent decrease. The frequency of brightness adjustments made during day and night driving from P3 to P4 also decreased. The daytime rate of brightness adjustment decreased from 12.5 to 8.9 and the nighttime brightness adjustments decreased from 23.3 to 17 per 1,000 km.

Figure 5-73. HUD Brightness Adjustments by FOT Period and Time of Day

[d]

Brightness adjustments by time of day and FOT segment are further broken down by gender in Figure 5-74. Descriptive comparison evidences that female drivers adjusted the HUD brightness controls more frequently per 1,000 km for all segments examined, with the exception of daytime driving in P3.

Figure 5-74. HUD Brightness Adjustment by FOT Period, Time of Day, and Gender

[d]

FCW Sensitivity Setting Adjustments

FCW provided six sensitivity settings, with setting 1 (S1) representing the least sensitive setting and S6 the most sensitive setting. In S1, the cautionary icons were suppressed altogether, providing only the imminent alerts (for additional detail, see General Motors Corporation, 2005). By contrast, S6 was most sensitive, where icons changed in size and color as the distance to the lead vehicle decreased, culminating in an imminent auditory alert. The ACAS vehicle was set at S4 by default when the driver received the car.

During the twenty days of ACAS-enabled driving, participants adjusted the FCW sensitivity settings frequently, an average of 10 times per day per 1,000 klm. Descriptive comparison of “all” FCW sensitivity-setting adjustments per 1,000 km in Figure 5-75 indicates that the overall number of adjustments decreased 42 percent between P3 and P4, where younger and older drivers made the most frequent adjustments in P3, while older drivers made the most changes in P4. The most marked reduction between P3 and P4 was for younger drivers, at 52 percent. As illustrated in Figure 5-75, when considering road type, during both P3 and P4, drivers made approximately four times as many changes to FCW sensitivity settings on arterial roads compared to highway driving.

Figure 5-75. Changes to Sensitivity Settings by FOT Period and Road Type

[d]

In Figure 5-76, the percent distribution of kilometers driven by setting during P3 and P4 depicts how drivers allocated their preferred sensitivity settings over the duration of the FOT. In P3, the two most sensitive settings accounted for slightly more than one third, 35 percent of km traveled, compared to 28 percent in P4. Similarly, the two least sensitive settings, S1 and S2, accounted for 30 percent of km traveled in P3 and increased to just over one third of travel distances, 34 percent, in P4.

Figure 5-76. Percent Distribution of Km Driven by FCW Sensitivity Settings and FOT Period

[d]

With regard to gender, Figure 5-77 illustrates that use patterns for each of the sensitivity settings were similar for both men and women. Over time, the tendency was to move away from the most sensitive setting, S6. Additionally, there was a slight increase in the use of S2 and S4 across P3 and P4 by both men and women, with women preferring this setting over men in general. Further, whereas men sharply decreased their use of S1, women did not alter their use of this setting over time.

Figure 5-77. Percent Distribution of Km Driven by FCW Sensitivity Settings, FOT Period
and Gender

[d]

In Figure 5-78, the percent distribution of kilometers traveled using each sensitivity setting is presented descriptively for age groups and overall, by FOT period. Older drivers drove a mean of 41 percent of their travel distances during P3 in S6, reducing travel in this setting to 34 percent for P4. Similarly, middle-age drivers reduced the percent of their driving in setting 6 from 23 percent to 16 percent; younger drivers evidenced the largest reduction, dropping to 12 percent from 22 percent. For the least sensitive setting, S1, younger drivers drove the greatest percentage of their P3 travel distances in this setting, at 29 percent. There was a slight reduction to 26 percent for P4. In contrast, older drivers rarely used this setting, a mere 3 percent of their travel over P3 and only 4 percent in P4. Middle-age drivers reduced their use of S1 from 26 percent to 21 percent over the FOT duration.

Figure 5-78. Percent Distribution of Km Driven for FCW Sensitivity Settings by FOT Period and Age Group

[d]

5.5.7.1.3 Trip Patterns

Patterns of travel were examined to evaluate changes associated with the availability of ACAS during the twenty days it was available. Figure 5-79 compares the percent distribution of distance traveled by age and gender groups for the time when ACAS was disabled (P1 and P2) with the time when ACAS was enabled (P3 and P4). There is not a large amount of variability overall, however, older males traveled a greater percent of FOT distances in the ACAS-disabled period, while older females traveled a larger percent of the distance during ACAS-enabled driving.

Figure 5-79. Percent Km Driven for P1-P2 and P3-P4 by Age Group and Gender

[d]

Additionally, Table 5-18 compares the mean number of trips per day and distance traveled per day by driver and ACAS-disabled versus ACAS-enabled periods. No differences resulted in the mean number of trips or distance traveled per day using ACAS as compared to ACAS-disabled driving.

Table 5-18. Number of Valid Trips and Distance (Km) per Day ACAS-Disabled Versus Enabled

	Distance (km) /day/driver	Trips
ACAS disabled (P1+P2)	91.5	4.8
ACAS enabled (P3+P4)	92.0	4.6

5.5.7.2 Investigating Travel Behavior and FCW Driving Performance Measures

Travel behavior was significantly intercorrelated with the measure of driving performance that assessed the degree to which drivers felt comfortable performing additional tasks while using the FCW system as compared to manual driving (r = -.26, p < .05). This suggests that drivers who received more FCW imminent alerts were less comfortable performing additional tasks while driving, and may have been related to concerns regarding receiving additional alerts or a need to respond to alerts.

5.5.7.3 Debriefing and Focus Group Comments Regarding Driving Performance with FCW

Drivers made comments when they were asked during debriefings to describe situations in which they came close to having a rear-end collision. Their comments suggest that they became more aware of threats due to FCW. In many of these comments, drivers described the utility of FCW, most typically in terms of gaining their attention when they were distracted.

Although the duration of the FOT was not long enough to establish if FCW use changed drivers’ behavior, a comment made during one of the focus groups suggests that this occurred.

5.5.7.4 Summary

Survey responses indicated that drivers felt very responsive to, and aware of, traffic when they used FCW. This may have been attributed to the way FCW operates, in that it explicitly called their attention to potential threats. Among the age groups, older drivers rated themselves as more responsive and aware than the younger and middle-age drivers. On the other hand, participants did not describe themselves as relying too much on the FCW system.

On average, drivers estimated they came close to rear-end collisions 0.84 times using FCW, compared to an estimated incidence of 0.41 when driving the FOT vehicle in the ACAS-disabled, or manual, mode. The increased estimate for the incidence of close calls may, in part, be related to increased awareness of traffic and/or experimentation with the FCW system. In neither ACAS-enabled, nor -disabled driving, did age group differentiate responses to these survey items.

Particularly when first experiencing the FCW system, drivers made more frequent adjustments to the HUD interface. The incidence of adjustments tapered off during the three weeks of ACAS-enabled driving. On a descriptive basis, it was evidenced that older and female drivers appeared to adjust the vehicle controls most frequently. Additionally, time of day appeared to be a factor with regard to variations in frequency of HUD manipulations, in that drivers made more adjustments to the HUD position and brightness when they drove in the dark. Factors such as an individual’s height and eyesight may have contributed to the occurrence of these adjustments. Drivers also made frequent changes to the FCW sensitivity settings, however this activity decreased markedly over time, as participants became accustomed to driving with the activated system. It also appeared as though drivers as a group migrated away from the least and most sensitive settings over the duration of the FOT. Initially, older drivers selected the most sensitive setting with the greatest frequency, and while a decrease was evidenced over time, the trend toward driving in S6 was nevertheless maintained in P4.

Driver comments suggested that they often felt that they became more aware of traffic threats while using FCW. In cases where individuals realized that they were not sufficiently attentive to the driving task, they tended to express appreciation with regard to the FCW alert. In addition, some drivers viewed FCW providing feedback on the safety of their driving practices, enabling them to learn more about and improve their driving.

5.5.7.5 Driving Performance – ACC

The ACC driving performance objective, similar to that for FCW, assessed to what degree, and how, drivers adjusted their driving with respect to ACC. Driver performance considerations were specified with regard to four sub-objectives²⁵. These included awareness, which addressed driver vigilance; vehicle control inputs, to examine driver behavior with regard to adjusting ACC settings; and trip patterns, to evaluate potential changes in travel behavior associated with ACAS-enabled driving. For reference purposes, Appendix R – Appendix T present results for ACC driving performance measures in a form parallel to that used for FCW, where possible. This includes survey item intercorrelations, descriptive statistics, and ANOVA results for analyses investigating potential age group attitudinal differences. Additionally, Appendix U presents figures that support the vehicle control inputs and trip patterns sub-objectives. Data about drivers’ assessment of their driving performance with ACC are provided in the appendices cited. No detailed analyses are provided due to scope limitations.

5.6 CONCLUSIONS

This chapter assesses the subjective opinions of FOT participants with regard to overall acceptance of ACAS and its system components: FCW, to a greater extent, and ACC. Because there is no recognized approach to driver acceptance, the independent evaluation developed a framework, building on previous research, to guide the driver acceptance analyses. The five objectives of this framework structured the driver acceptance assessment.

5.6.1 Advocacy

Most generally, analysis of the advocacy survey data suggested that driver attitudes, overall, regarding acceptance of the ACC system were positive and that those regarding FCW were somewhat less so. However, it is important to consider that there was often variability in responses that was masked by summary statistics and borne out in analyses investigating age group differences, in particular.

While the majority of drivers reported moderately positive attitudes toward FCW, anywhere from 14 percent - 36 percent of younger and middle-age drivers expressed negative opinions regarding their intent to purchase FCW. Older drivers were more likely to consider purchase, on average, than younger drivers. However, it is relevant to note that older drivers made significantly greater use of the ACC system, which served to reduce the number of imminent alerts received, and in turn may have an impact on attitudes.

Drivers’ reported experiences with FCW alerts appeared to be associated with attitudes regarding the system, as expressed in focus groups and during debriefings. For instance, imminent FCW alerts that served to refocus the driver on the road, or alerted him/her to “actual” perceived threats, tended to garner positive regard for the system. Additionally, the subset of drivers who experienced situations where they rated at least half of their FCW alerts as useful tended to be more consistently positive with regard to their degree of advocacy. However, participants did express concern regarding possible reactions from other drivers if, based on an alert, they responded to something not typically viewed as a threat.

Driver acceptance of ACC, as indicated by the advocacy survey measures, is best reflected in the dearth of negative Driver Acceptance Scale scores. Although drivers varied in the extent of their advocacy regarding ACC, responses were overall quite positive. A main concern expressed during focus groups and debriefings pertained to how other drivers might react to a vehicle exhibiting unexpected deceleration, acceleration, and/or activated brake lights in the midst of free flowing expressway traffic. Participants were concerned about how to warn other drivers about non-normative or unexpected vehicle actions, even to the point of suggesting placing a placard on the FOT vehicle’s roof.

A head-to-head comparison of driver acceptance regarding FCW and ACC is inequitable and should be guarded against. ACC is best classified as an “incremental” innovation, building on the familiarity that most drivers now have with CCC, is used when convenient, and at the driver’s discretion. However, FCW, as a “preventive” innovation, is less familiar, unable to be deactivated and, given the relative rarity of collisions it may mitigate, called on only infrequently. Rodgers (1995) noted that preventive innovations, such as FCW, are more difficult to introduce because the time scale required in order to see benefits is much longer as compared to innovations that are considered incremental.

5.6.2 Perceived Value

Drivers offered generally positive ratings pertaining to FCW safety and understandability. The more alerts drivers received, the less comfortable they felt performing additional tasks. Anecdotally, drivers who received imminent alerts while distracted appeared to recognize the benefits of such a system. With regard to ACC, measures of perceived value were consistently positive. Among the age groups, older drivers tended to be more positive in their attitudes concerning the predictability, distraction, and safety associated with using ACC. It is challenging to ask drivers if a system such as ACC will improve their safety because drivers identify themselves as safe drivers prior to acquiring enhancements such as ACC.

5.6.3 Ease Of Use

Drivers rated the ACAS implementation easy to use in terms of its settings and controls. They reacted positively to the HUD and its display of the FCW and ACC visual elements. Their suggestions to improve ACAS included the following: reducing the number of false alarms, simplifying the visual display and refining the color palette and icons, improving FCW’s reliability (including during bad weather conditions), providing user-adjustable options, such as an on-off switch and adjustments by traffic state, and altering the imminent alert timing.

When drivers received FCW alerts that were not useful, considered false, or called their attention to obvious and expected actions of other vehicles, they tended to express less positive opinions. In particular, drivers did not like to receive what they viewed as “unnecessary” FCW crash-imminent alerts and were especially annoyed when they received alerts triggered by stationary objects on the side of the road or for no obvious reason. False imminent alerts were viewed as more annoying if they occurred repeatedly. Some FOT participants drove the same roads every day and expressed annoyance at receiving recurring false alerts from the same non-threatening objects. Middle-age and younger drivers were more likely to report this type of annoyance due to their work trips.

FCW false alerts deemed to have been triggered by benign inanimate sources tended to undermine the credibility of FCW. This outcome is consistent with research on trust in automation. Madhavan (2003) reported that if an automated aid makes errors on easy tasks, people are less willing to trust and rely on it than an aid that makes errors on difficult tasks and performs easy ones reliably. One driver used the “cry wolf” analogy to describe his reaction to repeated false alerts, while agreeing that, overall, FCW had safety benefits.

5.6.4 Ease of Learning

Although drivers rated FCW as easy to learn to use in a short amount of time, many did not appear to understand how it worked, as was evident from remarks made during focus groups and debriefings. For instance, when asked during the debriefing whether changing the sensitivity settings affected the timing of crash-imminent alerts, 41 percent of the drivers responded that manipulating the FCW sensitivity settings altered the imminent alert timing, which, in reality, was fixed.

Misunderstanding turned to frustration when drivers attempted to purposefully trigger an imminent FCW alert and were not able to do so. In the process, individuals sometimes pushed themselves to close the distance gap to the vehicle ahead, without realizing that relative velocity was also a factor. Consequently, some drivers were uncertain about exactly what activated their FCW alerts. They felt that the system alerted too late, because their ineffective attempts at testing FCW resulted in their vehicle ending up too close to the lead vehicle for their own comfort. The assumption made by some drivers, that FCW alerted based solely on distance to the vehicle ahead, may have been inadvertently and partially fostered by the system’s visual representation of distance to the lead vehicle, a set of waves that some drivers referred to as “car lengths.”

Nevertheless, many drivers volunteered that using FCW helped them to learn about their driving and reinforce good habits. Drivers realized that FCW was intended to mitigate driving risks that, on an individual basis, occur infrequently. As a result, when asked, drivers tended to identify feedback about their driving as a safety benefit.

5.6.5 Driving Performance

Drivers adjusted FCW controls fairly frequently, particularly when they first interacted with the system; however, the incidence of adjustments tapered off during the three weeks of ACAS-enabled driving, indicating a learning effect. Time of day appeared to be a factor, in that more frequent HUD adjustments were made during night driving.

Drivers evaluated themselves as more responsive to, and aware of, traffic when they used FCW. This may be attributed to the way FCW operates, as it explicitly called their attention to potential threats. However, drivers did not describe themselves as relying too much on FCW. Comments provided during focus groups and debriefings also supported the notion that drivers were not over-reliant, rather that their awareness of traffic threats increased when FCW was operating.

In conclusion, driver acceptance findings suggest a mixed response to the FCW system by FOT participants as a group. The data indicate that, when FCW alerted drivers to actual threats, their opinion of the FCW system was more positive. However, drivers did not experience many actual threats. The more tentative opinions may result from receiving false alerts that were deemed excessive and/or recurring. In general, drivers viewed ACC very positively, despite expressing concerns about its ungainly acceleration and braking, as well as some degree of uncertainty about brake light activation.

---

²⁰Correlations were performed using Spearman's rho for nonparametric data.

²¹Parametric ANOVA results are reported here and throughout this report as justified in the section on Data analysis (5.3.4).

²²The driving style/risk compensation sub-objective was proposed to address the possibility that driver behavior was affected in a way that was not consistent with the goals the ACAS system. At present, the proposed analysis of variables including headway distance and driver distraction require an additional effort.

²³Correlations were performed using Spearman's rho for nonparametric data.

²⁴Parametric ANOVA results are reported here and throughout this report as justified in the section on data analysis (5.3.4).

²⁵The driving style/risk compensation sub-objective was proposed to address the possibility that driver behavior was affected in a way that was not consistent with the goals the ACAS system. At present, the proposed analysis of variables including headway distance and driver distraction require an additional effort.

Previous | Next | Top | Table of Contents

ITS Home | US DOT Home | Privacy | Feedback

United States Department of Transportation - ITS