1 Introduction

Instrument is an important component that shows the normal operation of the car or the fault alarm. Some research works are attracted by the distraction and distraction caused by the in-vehicle information system to the driver [1]. As an important component of the car [2], its reasonable design layout plays an important role in the driver’s reading efficiency. For the overall design of the instrument, the problem of encoding the instrument information such as shape coding, character coding, color coding, position coding should also be considered. Scholars of China also have relevant conclusions and explanations on the design requirements of manipulator layout settings, color coding, warning characters, etc. [3,4,5]. From the perspective of information coding, the outer frame structure of the instrument and its size and color, symbol size and its color have a great effect on the driver’s reading [6]. At present, scholars mainly tend to study the shape and character encoding of the instrument panel [7, 8], and there are relatively few considerations for the color and position coding of the instrument panel.

Zhang [9] used eye movement technology to study the design of human-machine interface in aircraft cockpit. Liu [10] simulated aircraft in the cockpit with eye tracker and touch screen to obtain eye movement and manual data to optimize the ergonomic design of the aircraft cockpit panel. Sun [11] use eye tracking technology to read the correct rate and task duration of the vehicle dashboard Indicators such as task search duration were analyzed.

Therefore, ergonomics test is used to explore the human-computer interaction design of the instrument panel in order to propose reasonable human-machine interface design requirements and help drivers to obtain information efficiency and reduce driver distraction and inattention.

2 Method

2.1 Experimental Design

This experiment adopted the two-factor four-level design method, that is, the independent variable - “dashboard layout design × color design”, the dependent variable - “reading speed × eye movement indexes”, and analyzed the influence of the independent variable factor on the dependent variable. After investigation on the common car dashboard, the classification according to the frequency of each types of dashboards were summarized. The dashboards including 4 different layout designs and 4 different color schemes were obtained, and Adobe Illustrator was used to design the dial so that 35 dashboard analog pictures - experimental stimulus materials were obtained and readings were set on each dashboard. The subjects were required to complete the reading of the pointer on the instrument panel independently. The task completion time and visual search duration, reading time, eye movement data were collected in the process and the subjects were not be disturbed. The screen brightness and indoor illumination were stable and the distance between the subject and the screen was 0.55–0.65 m.

2.2 Experimental Equipment

The equipment for experiment includes hardware and software. The hardware is the Tobii X2-30 eye tracker, and the software includes a human-machine-environment synchronization platform system and a data analysis system. The eye tracker was used to capture the eye movement trajectory of the subject. The eye movement data of the subject was recorded in conjunction with the human-machine-environment synchronization platform system.

2.3 Experimental Subjects

32 college students and teachers were invited to participate in this experiment, and 2 of them were unable to take part in due to severe astigmatism. Therefore, 30 subjects (23 students and 7 teachers) were effective. The ratio of male to female was 1:1, and the overall age was between 20 and 45 years old. There were 12 drivers with driver’s license and driving experience, 8 with driver’s license but no driving experience, 10 without driver’s license. The experiment requires a right hand, normal vision or corrected vision, no serious astigmatism, and good health. Since most of the subjects are college students and young teachers, the results are more suitable for young drivers.

2.4 Experimental Materials

In the current car dashboard design, there are four types of instruments: speedometers, engine tachometers, fuel gauges and water temperature gauges. According to the position of the instrument, four different typical layout schemes were summarized. In order to facilitate the distinction and description, the different dashboard layouts were named A, B, C, D class design (see Fig. 1). Adobe Illustrator software was used to simplify the processing of the instrument panel, and 35 (from 1–35 sequentially numbered) simulated images were obtained as test stimulus materials.

Fig. 1.
figure 1

Simplified dashboard design

Full size image

In order to eliminate the design elements unrelated to the test variables, the size, line width, division value, scale and pointer of dial were unified. The background color of the instrument panel was uniformly set to black. Except the pointer of picture No. 34 was designed to be blue, others are uniformly set to red. The detailed settings of 35 stimulating material are shown in Table 1.

Table 1. Instructions of layout, color and reading about dashboard
Full size table

2.5 Procedure

The test includes three stages: 1. Before the test, explain the test content and precautions to ensure that the subjects understand the purpose and process of the experiment. 2. In the test, each subject only reads one type of instrument at a time; for each reading, subject can rest for 30 s to eliminate the error caused by the fatigue effect; reading is not limited by time and accuracy, the subject recognizes at normal speed. 3. After the test, the participants completed the questionnaire survey, and the main test personnel saved the data and organized the test site.

2.6 Data Processing

Data processing mainly includes two parts: analysis of the reading duration and eye movement data.

Search time is the time from scanning dial to find the target dial; reading duration is the time from the beginning of reading to recognize the dial. The sum of the two is the total duration of the task. The duration of the search, the reading duration, the total duration of the task and the number of times of misreading are recorded by the eye-tracker. According to the results of the experimental records, the data is filtered out, and the unreliable data is eliminated: the extreme data is eliminated according to the principle of adding or subtracting 3 standard deviations (M-3SD ≤ Xi ≤ M + 3SD) [12], where Xi represents the actual data, M(Mean) indicates the mean, and SD indicates the standard deviation. SPSS V21.0 is used to analyze the data.

For eye movement data analyzing, it includes dividing the Area of Interest (AOI), extracting the number of gaze points in the interest of area, calculating the average number of gaze times in different areas, and exploring the influence of different areas of the dashboard on visual search. The visualization function of the data analysis system obtains the eye movement trajectory of the subject and assists in analyzing the research conclusions.

3 Experiment Analysis

3.1 Influence of Different Types of Dashboard Design on Reading Duration

3.1.1 Analysis of Task Duration

Figure 2 shows the search time, reading time and total time of the speedometer in different layouts. From Fig. 2, the total time and average search time of the C-type layout design are the shortest, and the total time and average reading time of the B-type layout design are the longest.

Fig. 2.
figure 2

Reading time comparison of speedometers

Full size image

Table 2 shows the reading time of the speedometers for different color schemes. Among them, the F-type color matching design has a longer reading time, which indicated that the special color matching design (orange, blue, etc.) has a greater influence on the reading speed of the test.

Table 2. Reading time order of speedometers
Full size table

Similarly, the statistical analysis of the tachometer, fuel gauge, thermometer reading and reading time, the statistical results are shown in Table 3, which shows the average search time of the class A layout is the shortest and the total task time is also the shortest. To the fuel gauge and water temperature gauge, the average search time of the B-type layout is the shortest, while the A-layout has the shortest reading time. A-class design is the shortest in terms of total mission time.

Table 3. Reading time of different dashboards
Full size table

In summary, when the speedometer is the main target, the C-class layout is the best, but it is not conducive to the reading of the other three instruments. Considering the other three instruments, the A or B layout can be chosen.

3.1.2 Difference Analysis of the Total Task Time

Significant difference analysis is carried out for the average task time under different layouts, different color combinations and special instruments, and the significant level α = 0.05 was taken.

  1. (1)

    Different layout of the meter. The average test time of each instrument corresponding to different layout schemes is analyzed. The significance test results are shown in Table 4.

    Table 4. Differential analysis in different layout
    Full size table

For the tachometer, water temperature meter, the total task duration of C-class design is significantly different from the other three layouts. The differences between the three types of design, such as A, B, and D, are not significant, and C-class design and the other three types of design are the largest. The difference is that the speedometer of C-class design is placed in the middle and highlighted. It shows that the position of the speedometer has a significant impact on the reading of the tachometer and the water temperature meter, but has no significant effect on the reading of the fuel gauge. For the speedometer, there is a significant difference between class B and C instruments, class A and C. The differences between classes, D and C are not significant. In the previous analysis of the duration of various types of instrument tasks, when the speedometer is the main target, the C-class layout design is the best, but it is not conducive to the reading of the tachometer, the water temperature meter, the recognition of the tachometer and the water temperature meter. Reading time has a significant impact, considering the other three instruments for consideration, A layout is recommended to adopt.

  1. (2)

    Different color schemes of the instrument. The results of the significance test of the average task time of the instrument under different color schemes are shown in Table 5.

    Table 5. Significant test results in different color matching
    Full size table

The results show that different color matching designs only have an impact on the fuel gauge, and there is no significant difference in the reading of other instruments. The specific performance is the color matching design of B1 instruments (white on black) and B2 instruments (black and gray). There is a significant difference in the reading of the fuel gauge. With the previous analysis it can be seen that the use of complex color matching should be reduced in the design, and the easy-to-read color matching design is selected to improve the driver’s reading speed.

  1. (3)

    Special instrument analysis. Comparing the instruments with the scale red and the special color matching with the original image, the significance test results of the average task total time are shown in Table 6. The water temperature table is not set in No. 7, 34, 10, and 35 of the special color matching.

    Table 6. Significant test results for special dashboard
    Full size table

It can be seen that whether the meter scale is marked red or not, there is no significant effect on the reading of the speedometer, the fuel gauge and the water temperature gauge. The special color matching of the meter has a significant influence on the reading of the speedometer and the water temperature meter, but to the tachometer and the fuel gauge it has no significant effect. For the tachometer in A (2, 29) and D (14, 32) instrument layouts, the scale red treatment has a significant impact on it.

3.2 Eye Movement Data Analysis

3.2.1 Influence of Dashboard Layout on the Number of Fixation Points

During the test, the subjects stare at the stimulate material to form the fixation, as shown in Fig. 3.

Fig. 3.
figure 3

Fixation on the picture

Full size image

In this experiment, the more the number of gaze points, the more the subject was unable to determine the target. The attention of subject was attracted by the content of the dial which indicated that the search and reading efficiency was low. Some areas of the dashboard are used as the Area of Interest (AOI) and the number of fixation points are counted in each AOI under different layouts and color matching pictures. As shown in Fig. 4.

Fig. 4.
figure 4

Means of the number of fixations of different pictures

Full size image

Through the statistical analysis of the number of gaze points of each instrument under different layouts and color matching, it can be clearly found:

  1. (1)

    Regardless of the layout and color matching, the number of gaze points of the F-class picture (that is, the dial of the special color design) is generally more. Compared with the A-class picture, the speedometer, the tachometer, the fuel gauge and the water temperature table got more gaze points. The increase was 13.7%, 24.8, 16.4%, and 68.1%, which indicated that the subjects was easy to be affected by the design and the reading efficiency was also reduced. Among them, No.33 and No.34 pictures has more than three color and blue color is used. No.35 picture has a special color orange, so it is not advisable to use more than 3 color schemes and special colors such as blue and orange.

  2. (2)

    Overall, the design of Class A pictures is conducive to the overall reading of the instrument. The number of gaze points in the reading process of the speedometer, tachometer, fuel gauge and water temperature meter are relatively small, and it is not difficult to find. In the reading of the speedometer, the number of C-class gaze points is less than the number of A-class gaze points. From the point of view of the number of gaze points, the use of C-class is more conducive to the reading of the speedometer. But there is a problem, that is, it is necessary to consider whether the difference in the number of fixation points between the A and C-class pictures is significant. If the difference is significant, it is necessary to consider the influence of the dashboard layout on the driver’s reading. When driving a car, if the speedometer is considered as an important meter and the driver observes the highest frequency and has the greatest influence on the driver, it is recommended to select C-class. If the reading of the other three meters is considered comprehensively, it is recommended to select class A because the overall reading of the meter the Class A layout is the best.

Therefore, the difference analysis is made on the number of gaze points of the A-class and C-class layouts and the color matching speed, and the significant level is α = 0.05. According to the analysis, in terms of the speedometer, the difference between the number of gaze points in the layout and color matching of class A and class C is not significant (P = 0.135 > 0.05), indicating the choice of class A or class C layout and color matching speedometer. The reading has no effect. For the tachometer, fuel gauge and water temperature meter, the number of gaze points in the C-class layout is 23.5%, 28.7%, and 46.8% higher than that of the A class. Therefore, the layout and color matching of the A-class is obviously better than that of the C-class. So, A-class layout and color matching are recommended.

3.2.2 The Impact of Different Areas of the Dashboard on Visual Search

When the panel was read, the instrument panel layout in different areas decided to have different visual characteristics, so the instrument panel was divided into three areas (left area, right area and middle area), as shown in Fig. 5.

Fig. 5.
figure 5

Visual search area division

Full size image

The average number of fixation for the three regions during the visual search for the four instruments is shown in Table 7. Average number of gaze in a region equals to the total number of gaze in the area/total people.

Table 7. The number of average fixation count during the search
Full size table

According to the average number of gaze times in the visual search process, it was found that the first search in the search process was usually the right region, accounting for 36.29%, indicating that the region was most concerned by the subjects, followed by the middle. The area, accounting for 34.14%, received the least attention in the left area, only 29.57%. In the actual driving process, the driver may search in order according to his own habits on the dashboard. Vehicles of different brands will have different degrees of difference when setting the dashboard, and the driver’s driving experience is more sufficient. The habit effect of the vehicle dashboard on the subjects will be greater. On the contrary, if there is no driving experience or less driving experience, the reading of the instrument panel is more inclined to daily reading habits than driving habits or driving experience. Therefore, the speedometer and tachometer can be set in the middle and right areas of the instrument panel, and the dials (such as fuel gauges and water temperature gauges) that are not commonly used can be set in the left area of the instrument panel, which is beneficial to improve the search efficiency of the driver.

4 Conclusions

According to the experimental results, the following conclusions are obtained:

  1. (1)

    During the driving, the speedometer is most concerned by the driver, and other instruments are secondary concerns. In the design of the car dashboard, A type of layout (left and right structure, tachometer and speedometer are distributed on the left and right sides of the instrument panel, fuel gauge, water temperature gauge in the speedometer, tachometer) is conducive to the overall reading; If the speedometer is read separately, the C type of layout is better for reading.

  2. (2)

    Subjects have more fixation under special color schemes, indicating that it is difficult to get information, which makes the reading efficiency of the instrument panel lower. In a word, the appropriate special color matching is good for reading, but complicated color matching will reduce the driver’s reading efficiency.

  3. (3)

    When designing the instrument, it is not recommended to set the scale vertically, which will lead to the extension of the reading time, reduce the driver’s reading efficiency, and be unsafe.

  4. (4)

    According to the visual search habits of the subjects and the experimental research results of the number of fixation in the visual search, the main instruments (speedometer, tachometer) are set on the right and middle areas of the instrument panel (visual search efficiency is 70.43%); The dial (fuel gauge, water temperature gauge) that is not commonly used is set in the left area of the dashboard, which can improve the search efficiency of the driver.