Left-Turn Driving Behavior Based On Situational Anger: Day Or Night


 Anger is a key factor affecting drivers' subjective judgment and driving skills. The influence of anger on driving behavior has been widely studied, but there is a lack of comparative research under different lighting conditions. Through driving simulation experiment, this paper studies the influence of anger on left-turn driving behavior under two light conditions of day and night. In the experiment, 32 licensed participants were divided into two groups, one in emotional neutrality and the other in anger. Among them, the emotional state of anger is induced by a traffic related video. The results showed that compared with daytime participants, participants at night had higher anger intensity, shorter gap acceptance and post encroachment time (PET) when left-turn driving. In addition, compared with emotion neutral participants, angry participants tended to accept shorter gap acceptance and post encroachment time (PET) when turning left. This indicates that participants failed to respond correctly to left-turn driving behavior in a state of anger. However, the response of gender differences to situational driving anger was not affected by light conditions. The anger intensity of male participants during the day and night was higher than that of female participants, and the gap acceptance and post encroachment time (PET) during left-turn were shorter than that of female participants. This shows that male participants are more likely to produce high-intensity anger and are more likely to make dangerous driving decisions in a state of anger. This paper puts forward some suggestions on the identification of anger and the prevention of angry driving.

common 40 . Relevant studies show that our country is about 60.72% of motor vehicle drivers have experienced "road rage" 7 . Therefore, it is necessary to study the effect of anger on left-turn driving behavior.
Spielberger's State-The Character Anger Model 46 is a well-known theoretical framework about the impact of anger on tra c safety. Spielberger de nes the tendency to frequent and intense anger as characteristic anger, and from the temporary state of physical feelings and anger to state anger. According to this theory, Deffenbacher, Getting, and Lynch 10 distinguish between two modes of driving anger, which are trait driving anger and situational (state) driving anger. Trait driving anger refers to the tendency of the driver to become angry while driving, while state driving anger refers to the physical arousal and anger of the driver in the event of irritable situations. Trait-based driving anger is a relatively persistent and stable personality, while situational driving anger is an emotional state. Although many studies have shown that trait anger has a signi cant impact on driver behavior, drivers are also directly affected by situational driving anger 1,58 .Ho Dong Chao 27 , Līva Ābele 2 based on situational (state) driving anger patterns on the driver's driving performance, road decision-making, driving style, visual attention distribution, and other studies also show that situational driving anger will hurt the driver's driving behavior. However, there has been limited research on the effects of situational anger on left-turn driving behavior.
In China's tra c environment, tra c scenes such as vehicle lane change without turning lights, vehicles in intersection lane white solid line change, non-motorized tra c lanes, or even retrograde are very common in daily life and can arouse the anger of the majority of drivers to varying degrees 19 . Sullman 49 is based on the Driving Anger Scale 10 de nes four types of driving-related anger-inducing situations: impaired progress, dangerous driving, hostile gestures, and impolite driving. Wickem 52 also points out that behaviors such as walking back and forth (33%) and driving slowly (20%) tend to cause anger among drivers. Anger is an important cause of drivers' angry driving 50 . Most tra c accidents caused by angry driving behavior occur at night 32 . However, there has never been an investigation into whether lighting conditions can cause changes in drivers' anger.
In addition, the night driving environment is also a key factor affecting driving behavior. Night driving is dangerous for all road users 54 . Although only 21%-23% of vehicle mileage 56 is required to travel at night,51% of fatal accidents and 29.5% of accidental injuries occur at night. Road users have three times the mortality rate at night 6 . Joanne M. Wood 53 points out that dim lighting conditions can lead to a signi cant reduction in the driver's reaction time and pedestrian identi cation distance, thereby increasing the risk of driving at night. Second, driver drowsiness at night is also a major cause. In uenced by circadian rhythms, people will become more sleepy at night, and drowsiness leads to reduced drivers' awakening and functional sensory-motor skills, leading to poor driving decisions 35 However, there is little research on the effect of lighting conditions on left-turn driving behavior.
In the study of situational driving anger, the previous methods of inducing anger mainly included experience recall, story situation, and lm induction 24,22 .In recent years, the lm-induced method has been widely used in the study of situational driving anger, Zhang Tingru 58 ,Ho Dongchao 27 , LīvaĀbele 2 also demonstrated that the effectiveness of lm-induced methods in the induction of anger.
In this paper, the driving simulator experiments under different light conditions were carried out by using the lm induction method to induce the participants' anger, so as to solve the following problems: rst, will the anger be affected by the light conditions? Second, is the left turn driving behavior of participants affected by light conditions under the state of anger? The results of this study contribute to a deeper understanding of the mechanism and consequences of situational driving anger, and put forward some suggestions to control the negative effects of situational driving anger.

Method Participants
In this research, a total of 32 participants with normal vision or correction, no color blindness or other eye diseases, and who had a driver's license and at least one year of driving experience were interviewed anonymously and voluntarily. They were randomly divided into 2 groups of 16 people, equally male to female. All the participants were persuaded that their information will remain anonymous and the total data will be applied in research, increasing the tra c safety knowledge. Each participant is paid CNY 50. The participants were informed that they could quit the experiment at any time and were asked to sign an informed consent form before the experiment began.The study was approved by the Ethics Committee of Sichuan Normal University and all experimental procedures were performed in accordance with the Declaration of Helsinki.

Experimental equipment
The current study is carried out using a xed base all-instrument open cab simulator system. The driving simulator cab uses three 42-inch LED screens to provide a 180-degree horizontal view, one screen for the front view image, and two screens for the side mirror display. The simulator includes all automotive controls, such as a 1060-degree rotating feedback-powered steering wheel, clutch, brake and accelerator pedals, shifter (for manual transmissions), mirrors and side mirrors, steering signal lights, similar to actual cars. The simulator is equipped with a screen panel that displays the speedometer, tachometer, steering signal indicator, and the audio system in the simulator is connected to the analog scene, providing tra c sound and engine noise to reproduce the actual vehicle's surroundings, thereby enhancing the authenticity of the driving experience on the simulator.

Experimental design
This study was designed using a 2(control group/anger group) *2(day/night) hybrid experiment design. The internal factor of the participant is the emotional state. In the experiment, we used a ve-minute tra c-related movie clip taken from the driver's point of view to induce the anger of the participants. Before the experiment, participants were shown a ve-minute-long daytime and night video clip taken from the driver's point of view to induce anger. Participants were asked to imagine themselves driving in the video clip vehicle. Half of the participants performed left-turn tasks in anger-driven situations, while the other half completed the same tasks without anger-induced neutral emotions. The factor between the participants is the lighting conditions of the driving. Each group of participants was required to complete the left-turn driving task under different lighting conditions.

Experimental materials
The rst video is in the daytime: The participant's vehicle is driving normally in the inner lane of a fourlane city road. After about 30 seconds, a car overtakes the participant's vehicle in the outside lane and cuts in without using a turn signal. Participants honked their horns to warn the driver that such behavior was dangerous. But the vehicle in front began to deliberately block the participants' progress at a very slow pace. When the participant tries to get rid of the vehicle in front by changing the lane, the vehicle in front suddenly accelerates. This continued for about 3 minutes, with the driver of the vehicle in front throwing debris out of the window and accelerating away. In the experiment, the difference between the two videos was that the night participants were not only affected by the rude behavior of the vehicle in front but also by the vehicle coming in the opposite direction, while the participants were mainly affected by the rude behavior of the vehicle in front. Chen Yu 7 , Zhang Tingru 58 , Feng Zhong, Xiang 19 , Deng Yuan chang 13 , Ren Wei wen 40 . The above driving behavior will arouse the driver's anger.

Experimental scenarios
In the virtual environment we built, we designed a three-lane city road about 8000 meters long and 3.5 meters wide. According to the speed limit for intersections under tra c regulations, we set the same speed (30 km/h) for the through train and the left turn. At the same time, to create tra c con icts, we have designed different starting points for the through train and the left-turn train, as shown in Figure 1.
The test area of the experiment were points A 30m and B 64 m which far from the con ict area.In the experiment, the left-hander drove into the test area at a speed of 30 km/h. To avoid the perceived effect of speed changes on participants, the through train also needs to pass through the intersection at 30 km/h. In addition, participants can see that the pass-through vehicle is approaching. In this study, the con ict area is de ned as the overlapping area between the left vehicle track and the straight vehicle track.
The experimental scenario is shown in Figure 1: (Green indicates the direction of travel of the left-turn; yellow indicates the direction of travel of the straight car, and the red ag indicates the area of con ict).

Procedure
Participants arrive at the lab to sign an informed consent form for the experiment and ll in the relevant personal information (age, gender, driving experience) anonymously. The experimenters then divided participants into two groups on average, and everyone has 15 minutes to practice driving to familiarize themselves with the simulator's operation and simulation environment. The variables involved in our experiment are summarized in Table 1. Before the start of the experiment, a group of participants needed to complete the emotional assessment of the intensity of the three emotions (calm, anger, and fear) that they felt at the time, with a score of 7 (1 was not at all and 7 was very strong). In the experiment, we recorded and managed the emotional scores of the participants. Among them, fear was not scored and managed in this experiment, because fear was not involved in this experiment.
At the beginning of the experiment, the experimenters showed one group of participants a video clip of the daytime situation, re-evaluated the participants' emotional state after watching, and then conducted a simulated driving experiment. When participants completed daytime driving tasks, they needed a veminute break to complete the emotional score after the rst experiment and calm their emotions.
The second experiment began. The experimenters were asked to focus again on video clips from the night-time situation and repeat the emotional score, followed by a night-time simulated driving experiment. The third emotional score was performed after the end of the experiment. The other group was an experimental control group that conducted simulated driving experiments without watching videos and mood scores. In the experiment, each group of participants was required to complete two consecutive driving sessions, one in the daytime and one at night. The ow of the experiment is shown in

Situational Driving Anger Effectiveness Analysis
The effectiveness of emotional induction affects the effectiveness of experimental results. In this study, we used the Jeon 20 experiment method to ask participants to complete an emotional assessment (7 out of 1) of the intensity of the three emotions they felt before, after, and after the experiment (The score is 7, with 1 not at all and 7 very strong) and collect the relevant data. The emotion evaluation data from 16 participants were analyzed by using a 3(experiment phases)  Table 3.
It can be seen in this table that compared to pre-experiment and post-experiment, the emotional intensity of participants during the experiment had changed signi cantly. Specially, we can nd out that though participants had completed the experiment, the anger intensity of participants is still higher than they do not watch the video, this suggests that the anger that is aroused will last for some time, just as shown in Figure 2.
Our experiment focused on the intensity of participants' anger. We rated emotions 2(emotion state: anger/calm) *2(experiment phase: Pre-induction/post-induction) mixed ANOVA between the participants. In terms of the scores of calm emotion state, there were signi cant differences in emotional intensity between participants during different experiment stages (F(2,93)=51.11,p=1.06E-15,Fcrit=3.09), participants had lower levels of calm emotional intensity (5.34 vs 2.53) after watching the video compared to the emotional intensity before the experiment; there were also signi cant differences in anger intensity (F(2,93)=71.58,p=1.51E-19,Fcrit=3.09), participants had a higher intensity of anger (3.25 vs. 6.06) after watching the video compared to the pre-experiment emotional intensity. These results showed that the intensity of anger changed signi cantly after the participants began the experiment, and it proved the effectiveness of the lm-induced method used in this paper in the study of situational driving anger. As shown in Figure 3.

Analysis of participants' emotional intensity
First, we analyzed the mixed variance of the emotional scores between the participants on gender and emotions 2(emotion state: anger/calm) *2(gender: male/female). In terms of calm intensity, the gender difference had a signi cant effect on the participants' emotional intensity (F(2,93)=12.19,p=0 .0007,Fcrit=3.94).Female participants had a higher level of calm than male participants (4.25 points vs 3.13 points); in terms of anger intensity, the gender difference also had a signi cant effect on participants' emotional intensity (F (2,93)=4.48,p=0.04, Fcrit=3.94). Male participants had a higher intensity of anger than female participants (4.79 points vs 4.15 points). These results showed that male participants were more likely than female participants to develop anger and have a higher intensity of anger. As shown in Figure 4: At the same time, we also performed an ANOVA of light and emotion 2(emotion state: anger/ calm) *2(lighting conditions: day/night). In terms of calm intensity, the lighting conditions had a signi cant effect on the participants' emotional intensity (F(2,93)=10.26,p=0.002, Fcrit=3.94), participants in the daytime had a higher level of calm intensity (4.21 points vs 3.17 points) compared to participants in the night, and lighting conditions also had a signi cant effect on participants' emotional intensity in terms of anger intensity ( F(2,93)=12.06,p=0.0008, Fcrit=3.94), participants at night had a higher intensity of anger (4.98 points vs 3.96 points) than participants in the daytime. These results showed that participants were more likely to develop anger and have a higher intensity of anger at night. As shown in Figure 4.
Finally, a mixed ANOVA between light and gender 2 (lighting conditions: day/night) * 2 (gender: male/female) showed that there was no interaction between light and gender on the intensity of calm (F (2,93) =0.59, p=0.44, Fcrit=3.94); but a signi cant interaction between light and gender on anger intensity (F (2,93) =6.01, p=0.02, Fcrit=3.94). As shown in Table 4. P<a=0.05 Regardless of lighting conditions, male participants had a higher intensity of anger than female participants (4.33 points vs 3.58 points;5.25 points vs 4.71 points). These results showed that male participants were more likely than female participants to have a higher intensity of anger. As shown in Figure 5.

Analysis of the effect of situational driving anger on left-turn behavior
The driving data from the 32 participants were analyzed by using a 2(emotional state)  Table 5.  Fcrit=3.99). These results showed that participants at night had shorter gap acceptance (4.75s vs 6.28s) and shorter post encroachment time (PET) (3.38s vs 4.47s) when turning left than participants in the daytime. As shown in Figure 7: We also tested 2 (emotion state: anger/control) * 2 (lighting conditions: day/night) * 2 (behavior: gap acceptance/ post encroachment time) to investigate whether different emotions and light conditions interact on left turn behavior. The results showed that there was no interaction between different emotions and light conditions on gap acceptance of the participants' left turn behavior (p=0.72> a=0.05); there was also no interaction between the different emotions and light conditions on the post encroachment time (PET) of the participants' left turn behavior (p=0.42> a=0.05). These results indicated that participants showed no signi cant differences in left-turn driving behavior during daytime and nighttime anger. As shown in Figure 8: Finally, to test the differences in gender in the experiment, we conducted four variables 2 (emotions: anger/control) *2(lighting conditions: day/night) *2(behavior: gap acceptance/ post encroachment time) *2(gender: male/female) mixed variance analysis. The experimental results show that the gender difference is not signi cant.

Discussion
Analysis of the effectiveness of the lm-induced method The focus of this study was the intensity of anger among participants before and after the start of the experiment. In the experiment, we observed a decrease in calm emotional intensity (5.34 points vs 2.53 points) and an increase in anger intensity (6.06 points vs 3.25) after watching the video clip. The results indicate that the participants' anger was successfully aroused. This is consistent with the study of Zhang Tingru 58 , that is, the lm-induced method is effective in situational driving anger research.
Compared to other interaction methods, the experimental method based on the driving simulator may be considered to lack ecological effectiveness. Our experiment is very similar to the experience of anger while driving. In actual tra c, although an event that provokes anger usually lasts only a few minutes, the anger it had aroused may continue into a subsequent driving situation and has an impact on driving behavior 47 . Some people may think that the anger may quickly disappear when performing demanding driving tasks. However, the results showed that despite the decrease in anger intensity during the driving simulation experiment (4.09 vs 6.06 points), however, the anger is still signi cantly higher than before the anger induction (4.09 points vs 3.25 points).

Effect of lighting conditions on situational driving anger
In conclusion, the results of this study show that the anger intensity of participants at night is signi cantly higher than that during the day. Because the risk of driving at night is higher than that during the day, and the reduced visibility at night increases participants' cognitive burden on the environment 30 , participants are more likely to have a high degree of anger. Relevant studies show that drivers will not reduce their speed due to reduced visibility at night 28 . Therefore, participants need to focus on the surrounding environment to ensure the safety of driving behavior. When this attention is interrupted by sudden rude behavior, participants will instinctively produce stress psychological response, resulting in the increase of angry emotional intensity 40 . At the same time, due to the in uence of circadian rhythm, humans usually become more sleepy at night 25 . Driving for a long time makes the participants bear both physical and mental pressure. Participants need to maintain a high degree of tension to deal with various possible situations. When the front party suddenly rude driving behavior, participants will be more sensitive to rude provocation, resulting in an increase in anger. This is consistent with Li Ming's 32 research, that is, before the outbreak of "road rage", the actor's individual psychology mostly had adverse tendencies and pathological signs such as anxiety, tension, irritability, depression, impetuosity, anger, evasion and jealousy. Chen Yu 7 also pointed out that the more frequent you drive, the more mileage you drive, the easier it is to drive angrily.
The results showed that male participants were more angry than female participants regardless of light conditions.The main reasons for this difference between male and female subjects are as follows: First, women are more likely than men to predict the potential risks and future consequences of their behavior.
They will avoid any violent behavior that may cause each other's anger to form a self-protection mechanism 16 . Secondly, female drivers respect the rules more than male drivers. Meanwhile, compared with female drivers, male drivers have relatively low public self-awareness, which leads male participants to pay more attention to their emotions and have a higher degree of anger when encountering abnormal driving behavior 26 .Finally, compared with male participants, female participants were more likely to adopt a more adaptive attitude to express anger, while male participants were more aggressive.The research of Li Ming 32 , Yan Lixin 55 and others shows that male participants have a higher probability of anger during driving than female participants. But Bjorklund 5 , Cong Haozhe 8 , LīvaĀBele 2 and others pointed out that female drivers are more likely to produce road anger. Pelin Deniz 38 also showed that masculinity and femininity regulate the relationship between driving anger and driving anger expression of young drivers.
It can be seen that the impact of gender differences on anger is controversial. In this study,our experimental results show that gender differences have a signi cant impact on the degree of anger of participants.

The effect of situational driving anger on left-turn behavior
The results of this paper showed that participants who were in a situational driving rage had shorter gap acceptance and post encroachment time (PET) when turning left than participants with neutral emotions.
This is consistent with studies by Zhang Tingru 58 , in which participants in anger had shorter wait times when turning left, resulting in shorter acceptable gaps. Beatriz González-Iglesias 20 points out that anger can lead to tra c violations by drivers. Li, Zhang, Sawyer, Zhang, and Hancock 34 . It is also con rmed that angry drivers tend to show a higher risk-taking tendency. From a behavioral perspective, driving anger can adversely affect driving behavior, increasing the risk of tra c accidents. Secondly, visual perception is the main source of driving information. Easterbrook 17 pointed out that high awakening negative emotions (such as anger) can lead to tunnel vision, a peripheral vision loss, but the central vision remains the same. Therefore, the horizons of drivers in situational driving anger will be narrow, and participants will focus mainly on tra c events ahead and miss safety-critical information in the surrounding view. All of these factories lead to wrong driving behavior decisions. Zhang Tingru 58 also pointed out that situational driving anger reduces the range of the driver's visual attention. Taken together, anger leads to incorrect driving decisions made by participants that increase the risk of tra c accidents.

Effect of lighting conditions on left-turn behavior
The experimental results showed that the gap acceptance and post encroachment time (PET) was shorter when the participants performed left-turn driving at night (6.06s vs 5.5s;4.06s vs 4.53s

Limitation
This study has some limitations but they could be addressed in the future. First, the driver's reaction in the simulator differs from the real world, and the adopted simulator adopted reduces the authenticity and effectiveness of the experimental results. In the future, the comprehensive use of modern technologies such as VR and EEG can be used to reduce the authenticity of the experimental results and the real-world difference.
Second, the participants' emotional scores were mainly self-reported data, and it was not possible to determine whether the participants had the mentality to meet social expectations. The study could consider objective data collection methods to minimize data errors. Third, other situations besides state driving anger that could cause participants' anger was also not considered, and the impact of verbal aggression on driver driving anger is an interesting topic of future research. Finally, most of the study samples are young drivers with a small sample size, and the experimental results are not generally representative. In the future, the study sample size should be considered to expand, and more participants with different driving experiences should be selected for experiments. In the future, we will work to provide a dynamic tra c model based on the situational driving rage to predict changes in driving behavior as participants drive continuously in the lane.

Conclusions
The purpose of this study was to study the effect of situational driving anger on left turn driving behavior under different lighting conditions. The unique contribution of this study is that by improving the experimental design and strictly controlling the potential confounding factors, it provides some causal relationships among illumination, situational driving anger and left turn driving behavior. The study successfully proved that light conditions signi cantly affected the anger intensity of participants.
However, light conditions do not necessarily affect gender's response to anger situations. In addition, the study found that the gap acceptance and post encroachment time (PET) of angry participants at night were shorter than that during the day. This shows that drivers in angry state are more careless at night than drivers in emotional neutral state.
The results showed that angry participants were more likely to drive dangerously at night. Based on the results of this study, we put forward the following suggestions to reduce the adverse effects of situational driving anger. First, develop real-time anger recognition technology and necessary mitigation strategies to help drivers manage emotions. Second, the driver's negative attitude in the state of anger is the key to affect the accident risk caused by anger. Therefore, we can increase the learning of correctly identifying and dealing with negative driving attitudes in driver education. Third, a hazard warning system can be installed in the vehicle to help angry drivers effectively obtain road information. Finally, the driving anger intervention strategy should be applied to different light environments.

Declaration Data availability
The data and material of the current study will be available from the corresponding author upon reasonable request.  The changes of emotion intensity in different experiment phase  The effect of light and gender on emotional intensity The in uence of different emotional states on left-turn behavior Figure 7 The effect of light and sex on left-turn behavior Figure 8 The effect of different lighting conditions on left-turn behavior