Keywords

1 Introduction

The proximity compatibility principle (PCP) is an approach to display design proposed by Wickens [1–7]. Under the high task proximity between two tasks, we can divide attention to simultaneously conduct two tasks without degrading task performances. In such a case, the display location to perform both tasks has a significant effect on the performance of both tasks. In other words, both performances are enhanced when both displays are located closely, or when both displays are similar. Under the low task proximity between two tasks, we must focus attention to each task in order to simultaneously conduct two tasks without degrading task performances. Even in such a case, the display location to perform both tasks has a significant effect (impact) on the performance of both tasks. Contrary to the high task proximity condition, both performances are enhanced when both displays are located distantly, or when both displays are dissimilar. The explanation above is summarized in Fig. 1. The solid line and the dotted line correspond to high and low task proximity conditions, respectively. The conception is that PCP may be applicable to the design of automotive display.

Fig. 1.
figure 1

Schematic explanation of basic concept of PCP (Proximity Compatibility Principle).

Full size image

The aim of this study was to propose an effective method for displaying driving environment and properly transmitting this to drivers using PCP principle. The display was designed by changing two types of proximity, that is, the proximity of the display and the proximity of the task itself. The proximity of the display was controlled as the type of the display or the distance between two displays. The proximity of the task was controlled as the difference of two tasks performed on the two displays. More concretely, the similar proximity of the task was either two analogue tasks or two digital tasks. The different proximity of the task was the mixture of digital and analogue tasks. These two proximities were all within-subject variables.

2 Method

2.1 Participants

Eight healthy male aged from 21 to 24 years took part in the experiment. All had held a driver’s license for 2–4 years. The visual acuity of the participants in both young and older groups was matched and more than 20/20. They had no orthopedic or neurological diseases. All signed the informed consent after receiving a brief explanation on the contents and the procedure of the experiment.

2.2 Apparatus and Task

A driving simulator system developed by Murata et al. [6, 7] was used. The participants were required to simultaneously carry out a simulated driving task (main task) using an accelerator and the brake. The participant was required to minimize the deviation from the predetermined line and keep the lane location using a steering wheel.

The following four types controlled by the two proximity factors (display proximity and task proximity) were selected as secondary tasks.

Secondary task 1: reaction task to the approach of a following vehicle. When the following vehicle approached within the predetermined distance, the participant was required to react using a cross switch of steering wheel.

Secondary task 2: lane changing task in which the participant was required to change the running lane.

Secondary task 3: speed maintaining task in which the participant was required to maintain the predetermined speed range from 45 km/h to 55 km/h.

Secondary task 4: reaction task in which the participant was required to react with a cross key of steering wheel when the distance to the destination was equal or less than the predetermined value.

Secondary task 1 and Secondary task 2 are processed mainly by a right hemisphere. Secondary task 3 and Secondary task 4 are processed mainly by a left hemisphere.

Combining the four types of secondary tasks above, the task proximity was determined as follows. Task proximity included four kinds of dual tasks.

High task complexity: Secondary task 1 (Left display) + Secondary task 2 (Right display), Secondary task 3 (Left display) + Secondary task 4 (Right display), and Secondary task 2 (Left display) + Secondary task 1(Right display).

Low task complexity: Secondary task 1 (Left display) + Secondary task 3(Right display), Secondary task 1 + Secondary task 4(Right display), Secondary display 2(Left display) + Secondary task3(Right display), and Secondary display 2 (Right display) + Secondary task4 (Left display).

The display proximity was controlled by the distance between two displays. The distance between displays for the close and the distant displays were 3 cm (visual angle of 2 degrees) and 17 cm (visual angle of 10 degrees), respectively.

2.3 Design and Procedure

The experimental factors were the display proximity (two: levels (close and distant)) and the task proximity (basically, low and high). All were within-subject factors. A total of 16 conditions specified by the task proximity and the display proximity were conducted by each participant. The participant conducted each condition for five minutes. The order of performance of display proximity condition was counterbalanced across the participants. Four participants conducted the low display proximity condition firstly. Other four participants conducted high display proximity condition firstly. For each display proximity condition, the order of performance of eight task proximity conditions was randomized across the participants.

Before starting an experimental session, the participants were allowed practice the experimental task until they were accustomed to the operation. The participants were required to carry out simultaneously a simulated driving task and 16 kinds of dual secondary tasks controlled by two proximity factors above. The performance measures were tracking error in the simulated driving task, the percentage correct and reaction time in the secondary task.

3 Results

3.1 Reaction Time in Secondary Task

As a result of a two-way (task proximity by display proximity) ANOVA was conducted on the reaction time of each of four secondary tasks. As for Secondary task 2, 3, and 4, no significant main effect or interaction were detected. Only for Secondary task 1 in which the participant was required to react using a cross switch of steering wheel when the following vehicle approached within the predetermined distance, a significant task proximity by display proximity interaction was detected (F(1,7) = 7.250, p < 0.05). In Fig. 2, the reaction time is plotted as a function of display proximity and task proximity in Secondary task 4 (reaction task when the distance to the destination was equal or less than the predetermined value). Figure 3 shows the reaction time as a function of display proximity and task proximity in Secondary task 1(reaction task to the approach of a following vehicle).

Fig. 2.
figure 2

Reaction time as a function of display proximity and task proximity (reaction task when the distance to the destination was equal or less than the predetermined value).

Full size image
Fig. 3.
figure 3

Reaction time as a function of display proximity and task proximity (reaction task to the approach of a following vehicle).

Full size image

3.2 Number of Errors in Secondary Task

A similar two-way (task proximity by display proximity) ANOVA conducted on the number of errors in secondary task revealed no significant main effect or interaction for all of four types of secondary tasks.

3.3 Tracking Error in Main Driving Simulator Task

A similar two-way (task proximity by display proximity) ANOVA conducted on the tracking errors in the main driving simulating task revealed no significant main effect or interaction for all of four types of secondary tasks.

4 Discussion

The proximity compatibility principle (PCP) (see Fig. 1) [2–5] is one of the cognitive scientific approaches to the display design. Under the high task proximity between two tasks, this principle predicts that we can divide attention to simultaneously conduct two tasks without degrading task performances. In such a case, the display placement to perform both tasks has a significant effect on the performance of both tasks. In other words, both performances are enhanced when both displays are located closely.

A method for displaying the secondary task and properly transmitting this to drivers using PCP principle was empirically explored. The display was designed by changing two types of proximity, that is, the proximity of the display and the proximity of the task itself. The proximity of the display was controlled as the type of the display or the distance between two displays. The proximity of the task was controlled as the difference of two tasks performed on the two displays. High proximity tasks were those which were cognitively processed with the same hemisphere. In this study, Secondary tasks 1 and 2 were mainly processed using a right hemisphere, while Secondary tasks 3 and 4 were mainly processed using a left hemisphere. Low proximity tasks were those which were cognitively processed using different hemispheres (One (Secondary task 3 or 4) was mainly processed by a left hemisphere, and another (Secondary task 1 or 2) was mainly processed by a right hemisphere).

As shown in Fig. 3, the reaction time plotted as a function of display proximity and task proximity also did not obey the prediction by PCP. If the prediction by PCP is applicable to this condition, the close display led to better performance under the high task proximity condition, while the distant display led to better performance under the low task proximity condition.

The reaction time shown as a function of display proximity and task proximity obeyed PCP as shown in Fig. 2 (reaction task when the distance to the destination was equal or less than the predetermined value). As PCP predicts, the close (distant) display led to better performance under the high (low) task proximity condition. It has been demonstrated that the performance under the dual task condition does not necessarily obey the prediction by proximity compatibility principle. In other words, in the range of this experiment, it seems that PCP is not necessarily a universally applicable display design guideline.

Future research should continue to explore whether PCP should be universally and effectively applicable to the automotive display design. As far as this study is concerned, the PCP prediction is not robust and does not universally true. Whether PCP is applicable or not seems to depend on the contents of the secondary task. It must be further discussed under what condition of PCP is applicable as Wickens [6, 7] predicted.