Perception-reaction Time Research Articles

Visual Perception and Visuomotor Reaction Speed Are Independent of the Individual Alpha Frequency.

While the resting-state individual alpha frequency (IAF) is related to the cognitive performance and temporal resolution of visual perception, it remains unclear how it affects the neural correlates of visual perception and reaction processes. This study aimed to unravel the relation between IAF, visual perception, and visuomotor reaction time. One hundred forty-eight (148) participants (28 non-athletes, 39 table tennis players, and 81 badminton players) investigated in three previous studies were considered. During a visuomotor reaction task, the visuomotor reaction time (VMRT) and EMG onset were determined. In addition, a 64-channel EEG system identified the N2, N2-r, and BA6 negativity potentials representing the visual and motor processes related to visuomotor reactions. Resting-state individual alpha frequency (IAF) in visual and motor regions was compared based on sport experience (athletes vs. non-athletes), discipline (badminton vs. table tennis), and reaction performance (fast vs. medium vs. slow reaction time). Further, the differences in the IAF were determined in relation to the speed of neural visual (high vs. medium vs. low N2/N2-r latency) and motor (high vs. medium vs. low BA6 negativity latency). Group comparisons did not reveal any difference in the IAF between athletes and non-athletes (p = 0.352, ηp2 = 0.02) or badminton and table tennis players (p = 0.221, ηp2 = 0.02). Similarly, classification based on the behavioral or neural performance indicators did not reveal any effects on the IAF (p ≥ 0.158, ηp2 ≤ 0.027). IAF was not correlated to any of the behavioral or neural parameters (r ≤ 0.10, p ≥ 0.221). In contrast to behavioral results on cognitive performance and visual temporal resolution, the resting state IAF seemed unrelated to the visual perception and visuomotor reaction speed in simple reaction tasks. Considering the previous results on the correlations between the IAF, cognitive abilities, and temporal sampling of visual information, the results suggest that a higher IAF may facilitate the amount and frequency but not the speed of information transfer.

Factors affecting behavior of mobile phone use while driving and effect of mobile phone use on driving performance

This study analyzed factors affecting behavior of mobile phone use while driving and its effects on driving performance, in terms of speed, lateral position, steer deviation, steer speed, following distance, perception–reaction time, and occurrence of a near miss situation. To investigate the factors affecting behavior of mobile phone use while driving, 1106 respondents from four different regions in Thailand participated in the questionnaire survey study. Theory of Planned Behavior (TPB) was used to explain these factors including two additional extended factors which are risk perception and law enforcement knowledge. The outcome of this part shows that attitude, norm, and law enforcement knowledge significantly affect the intention and behavior of the drivers. Even though approximately 90 percent of drivers realized that using a mobile phone while driving was dangerous and against the law, they have reported that they still use mobile phone while driving. To determine the effect of mobile phone use on driving performance, a 2-lane, straight rural highway, with a leading vehicle and an unexpected “STOP” sign, were simulated in order to examine the driving performance of drivers “without a phone”, “talking on a phone call”, and “texting a message” conditions. The results found that using mobile phone while driving can reduce speed and following distance, but increase lateral deviation, steer deviation, steer speed, perception-reaction time, and number of near misses leading to higher risks for road crashes.

Using vehicular trajectory data to explore risky factors and unobserved heterogeneity during lane-changing

This study aims to investigate contributing factors to potential collision risks during lane-changing processes from the perspective of vehicle groups and explore the unobserved heterogeneity of individual lane-changing maneuvers. Vehicular trajectory data, extracted from the Federal Highway Administration’s Next Generation Simulation dataset, are utilized and 579 lane-changing vehicle groups are examined. Stopping distance indexes are developed to evaluate the potential collision risks of lane-changing vehicle groups. Three mixed binary logit models and three mixed logit models with heterogeneity in means and variances are established based on different perception reaction time. Model estimation results show that several variables significantly affect the risk status of lane-changing vehicle groups, including the mean values of clearance distance and speed differences between the leading vehicle in the current lane and the subject vehicle, standard deviations of clearance distance, and speed differences between these two vehicles, as well as standard deviations of the speed difference between the subject vehicle and the following vehicle in the target lane. Interestingly, the influences of the last three variables differ considerably across the observations and the mean of the random parameter for standard deviations of clearance distance between CLV and SV is associated with the mean speed difference between CLV and SV. Since one of the explanations is individual heterogeneity, personalized designs for advanced driver assistance system would be an effective measure to reduce the risk.

Predicting driver reaction time and deceleration: Comparison of perception-reaction thresholds and evidence accumulation framework

Prediction of driver reaction to the lead vehicle motion based on the perception-reaction time (PRT) is critical for prediction of rear-end crash risk. This study determines PRT at various spacings in approaching and braking conditions, and examines the association of PRT and deceleration rate with crash risk. For these tasks, a total of 50 drivers’ behavior was observed in a driving simulator experiment with 4 different scenarios – reaction to a decelerating lead vehicle, reaction to a stopped lead vehicle, perception of a lead vehicle’s speed change, and perception of a slow-moving lead vehicle. The study tested three hypotheses of PRT including perception and reaction thresholds and the evidence accumulation framework using a visual variable (tau-inverse). It was found that the drivers neither reacted after a specific PRT from the start of perception nor reacted at a specific value of tau-inverse. Rather, the drivers generally reacted when the accumulation of evidence (tau-inverse) over time reached a threshold. It was also found that the magnitude of deceleration rate depends on the tau-inverse at the start of braking and hence, higher crash risk was associated with higher level of urgency and insufficient brake force rather than longer PRT. This study demonstrates that the evidence accumulation framework is a promising method of predicting driver reaction in approaching and braking conditions for different types of lead vehicle, and the level of urgency is important for predicting the probability of crash.

Evaluating Relationships between Perception-Reaction Times, Emergency Deceleration Rates, and Crash Outcomes using Naturalistic Driving Data

Perception-reaction time (PRT) and deceleration rate are two key components in geometric design of highways and streets. Combined with a design speed, they determine the minimum required stopping sight distance (SSD). Current American Association of Highway Transportation Officials (AASHTO) SSD guidance is based on 90th percentile PRT and 10th percentile deceleration rate values from experiments completed in the mid-1990s. These experiments lacked real-world distractions, and so forth. Thus, the values from these experiments may not be applicable in real-world scenarios. This research evaluated (1) differences in PRTs and deceleration rates between crash and near-crash events and (2) developed predictive models for PRT and deceleration rate that could be used for roadway design. This was accomplished using (1) genetic matching (with Rosenbaum’s sensitivity analysis) and (2) quantile regression. These methods were applied to the Strategic Highway Research Program 2 (SHRP2) Naturalistic Driving Study (NDS) data. The analysis results indicated that there were differences in PRT and deceleration rates for crash and near-crash events. The specific estimates were that, on average, drivers involved in crash events took 0.487 s longer to react and decelerated at 0.018 g’s (0.58 ft/s2) slower than drivers in equivalent near-crashes. Prediction models were developed for use in roadway design. These models were used to develop tables comparing existing SSD design criteria with SSD criteria based on the results of the predictive models. These predicted values indicated that minimum design SSD values would increase by 10.5–129.2 ft, dependent on the design speed and SSD model used.

Effects of Non-Driving Related Tasks During Self-Driving Mode

Perception reaction time and mental workload have proven to be crucial in manual driving. Moreover, in highly automated cars, where most of the research is focusing on Level 4 Autonomous driving, take-over performance is also a key factor when taking road safety into account. This study aims to investigate how the immersion in non-driving related tasks affects the take-over performance of drivers in given scenarios. The paper also highlights the use of virtual simulators to gather efficient data that can be crucial in easing the transition between manual and autonomous driving scenarios. The use of Computer Aided Simulations is of absolute importance in this day and age since the automotive industry is rapidly moving towards Autonomous technology. An experiment comprising of 40 subjects was performed to examine the reaction times of driver and the influence of other variables in the success of take-over performance in highly automated driving under different circumstances within a highway virtual environment. The results reflect the relationship between reaction times under different scenarios that the drivers might face under the circumstances stated above as well as the importance of variables such as velocity in the success on regaining car control after automated driving. The implications of the results acquired are important for understanding the criteria needed for designing Human Machine Interfaces specifically aimed towards automated driving conditions. Understanding the need to keep drivers in the loop during automation, whilst allowing drivers to safely engage in other non-driving related tasks is an important research area which can be aided by the proposed study.

Calibrating Design Guidelines using Mental Workload and Reliability Analysis

The generic nature of road design is indiscriminate to age, race, or gender, as it is implicitly assumed that there are few behavioral differences between drivers while traversing various alignment elements (e.g., horizontal curves, tangential segments, etc.). For instance, the perception reaction time required, which is based on an 85th percentile value, on a tangent section is the same as that on a horizontal curve. This suggests that current guidelines do not consider the complexity that some geometric features might induce on drivers, and consequently, there is a need to address the many considerations of diversity. In this respect, human factors should be explicitly included in design guidelines. One aspect of human factors that has received little attention in the literature is related to the mental workload. In this study, a procedure is presented to estimate the mental workload for stopping sight distance. Then, reliability analysis is conducted to compare the change in the probability of non-compliance owing to the available sight distance and based on the mental workload. By analyzing data from 12 horizontal curves in Alberta, Canada, the probability of non-compliance dropped from 9.1% to 0.7%, and a moderate correlation with collisions was found. The results of the analysis showed that incorporating mental workload into the geometric design process can improve safety performance.

The neural basis of hazard perception differences between novice and experienced drivers - An fMRI study.

The neural mechanisms underlying hazard perception are poorly understood as to how experience leads to better driving skills. In this study we used functional magnetic resonance imaging (fMRI) to examine experience-related changes in brain activation during hazard perception task between novice and aged drivers. Additionally, region of interest (ROI) and seed-to-voxel analyses were conducted to examine experience-related functional connectivity changes during visual attention and saliency networks between novice (n=15, age 22.13 ± 3.38 years years) and experienced (n=16, age 41.44 ± 5.83 years) drivers. Experienced drivers had significantly lower hazard perception reaction time (1.32 ± 1.09 s) and miss rates (11.42 ± 8.36 %) compared to the novice (3.58 ± 1.45 s and 39.67 ± 15.72 %, respectively). Blood oxygen level dependent (BOLD) activation increased in occipital, parietal and frontal areas when executing hazard perception task in both groups. In general, during the task execution, experienced drivers showed greater activation in the occipital lobe, supramarginal gyrus (SMG), right anterior insular cortex (AIC), anterior cingulate cortex (ACC) and cerebellar regions compared to the novice drivers indicating more efficient visual attention and decision-making process during hazard perception task. Seed based functional analyses during the hazard perception task revealed greater connectivity between the ACC and the entire salience network (visual attention network) in the experienced group. Additionally, ACC had higher functional connectivity with the right frontal eye field (FEF), bilateral intraparietal sulcus (IPS) and lateral occipital areas in the experienced group. Our results suggest that better hazard perception in the experienced drivers is due to increase in the activation of executive attention regions and higher functional connectivity between bilateral occipital cortices and salience network. In conclusion, better hazard perception is highly dependent on emotional awareness, perception of motion velocity, spatial representation of the environment and executing control.

Investigation into passing behavior at passing zones to validate and extend the use of driving simulators in two-lane roads safety analysis

A passing maneuver allows drivers to maintain their desired speed on two-lane highways. However, it entails a high risk of collision with vehicles travelling in the opposite direction. Investigating drivers’ behavior while performing passing maneuvers could provide helpful information on the factors that influence this process. Driving simulators have become important tools for driving behavior research studies as they are safe, facilitate the controlled use of experimental variables, and generate detailed output data. It remains to be seen whether simulator results can be considered representative of real-life driving conditions. With respect to passing maneuvers, no study has made a comprehensive and direct comparison between drivers’ passing behavior in the field and driver behavior observed in a simulated environment.In this validation study, a fixed-base interactive simulator was used to collect data from fifty-four participants (eighteen Iranians and thirty-six Italians) involved in several traffic scenarios on a two-lane rural highway segment (obtained by varying the speed of opposing vehicles, lead vehicles and headways in the opposite direction). A 3D model and its environmental characteristics were realized from the real segment which had previously been surveyed with drones to collect videos and derive data on real passing maneuvers.The results for the two-sided K–S test revealed no statistically significant difference in the accepted gap, effective accepted gap, perception reaction time, and time to collision variables between the field and the simulator at the 95 % confidence level. However, when conducting a one-sided K–S test, some statistical directional differences were found in the cases of the accepted gap and perception reaction time variables, which exhibited lower values in the field compared to the simulator again at the 95 % confidence level. Although the passing duration was statistically higher in the simulator than in the field, the shape of the two distributions was not statistically different. Analysis showed that differences in the passing duration are due to the lower passing vehicle speed and lower speed difference with the simulator than in the field, which are caused by truncating headways in the subject direction in the simulator. The cultural background of participants did not result in any discernible difference in passing behavior. The results would support a more extensive use of driving simulators in future passing behavior studies.

Monte Carlo Simulation Approach to the Duration of Yellow Lights at Signalized Intersections Considering the Stochastic Characteristics of Drivers

In China, around 90% of traffic crashes at signalized intersections take place within the signal change intervals, especially during signal change from green to red. Hence, yellow time, which is a part of inter-green time, is of great significance to the safety of signalized intersections. The conventional calculation method for duration of yellow light (DYL) ignores the stochastic characteristics of drivers, which we believe is an important factor in this calculation. Therefore, the purpose of this research is to investigate a new approach to calculate DYL based on safety reliability theory in which the randomness of human factors is taken into consideration. Firstly, a comprehensive literature review concerning the conventional calculation methods of DYL is conducted. Secondly, a theoretical calculation method of DYL based on safety reliability theory is put forward which, different from the conventional methods, accounts for the stochastic characteristics of drivers. Additionally, a driving simulation experiment is designed to obtain two driving behavior parameters of Chinese drivers: perception–reaction time (PRT) and safe acceptable acceleration (SAA). Thirdly, a Monte Carlo simulation is employed to simulate the interactive process of PRT and SAA, and solve the proposed model. Finally, according to the Monte Carlo simulation results, a look-up table describing the relationship between DYL, safety reliability (50–90%) and approaching speed (15–40 km/h) is made. Results show that this method successfully incorporates the probabilistic nature of driving behavior. Taking the safety reliability into consideration can provide a more reasonable method to calculate the DYL of signalized intersections.

Effect of Physical Exercise and Gender on Information Processing and Choice Reaction Time of University Students

The study aimed to examine the impact of a team sport match on the cognitive coding-decoding system and the psychomotor reaction time on university students. Twenty-one healthy university learners (average age = 21.5 ± .3 years) were administered the WAIS-IV scales and Leed Test psychomotors before and after a single bout of exercise. The exercise treatments consisted of a 5-min warm-up, a 5-min cooldown, and handball game for 20 min. The means were compared by ANOVA at the .05 significance level. The results revealed that exercise treatment resulted in significantly enhanced performance across Speed of information processing (p<.000), visual perception (p<.005) and choice reaction time (p <.040). Furthermore, under the effect of gender, the performances of males in total reaction time were significantly (p<.05) higher than females at rest and after exercise. But no significant difference was noted in the speed of treatment information and visual perception time. It is suggested that physical activity can be used as a means of stimulating cognitive skills and student development in university training programs in order to optimize certain cognitive functions and improve learning.

Component-Based Interactive Framework for Intelligent Transportation Cyber-Physical Systems.

While emerging technology for self-driving automation in vehicles progresses rapidly, the transition to an era of roads full of fully connected and automated vehicles (CAVs) may take longer than expected. Until then, it is inevitable that CAVs should coexist and interact with drivers of non-autonomous vehicles (NAVs) in urban roads. During this period of transition, it is critical to provide road safety with the mixed vehicular traffic and uncertainty caused by human drivers. To investigate the issues caused by the coexistence and interaction with humans, we propose to build a component-based and interactive intelligent transportation cyber-physical systems (ITCPS) framework. Our design of the interactive ITCPS framework aims to provide a standardized structure for users by defining core components. The framework is specified by behavior models and interfaces for the desired ITCPS components and is implemented as a form of human and hardware-in-the-loop system. We developed an intersection crossing assistance service and an automatic emergency braking service as an example of practical applications using the framework. To evaluate the framework, we tested its performance to show how effectively it operates while supporting real-time processing. The results indicate that it satisfies the timing requirements of vehicle-to-vehicle (V2V) communication and the limited processing time required for performing the functions of behavior models, even though the traffic volume reaches the road capacity. A case study using statistical analysis is conducted to assess the practical value of the developed experimental environment. The results of the case study validate the reliability among the specified variables for the experiments involving human drivers. It has shown that V2V communication support has positive effects on road safety, including intersection safety, braking events, and perception-reaction time (PRT) of the drivers. Furthermore, V2V communication support and PRT are identified as the important indicators affecting road safety at an un-signalized intersection. The proposed interactive framework is expected to contribute in constructing a comprehensive environment for the urban ITCPS and providing experimental support for the analysis of human behavior in the coexistence environment.

Visual Performance and Motorcycle Safety-Related Impacts of Various High Beam Headlight Intensities

&#x0D; &#x0D; &#x0D; In Malaysia, the three main types of motorcycle accidents include collision with passenger cars, collision with other motorcycles and single-motorcycle accidents. An accident may occur due to the lack of road design consistency and visibility where most drivers make more errors in the absence of geometric features. One of the most important barriers to the more frequent use of high beam headlight is the length of visibility of the headlight beam in relation to the stopping distance of a vehicle. Furthermore, it is crucial to evaluate the current limits on the luminous intensity of high beam headlights that is considered visible for a driver to come to a stop. Therefore, this first stage study summarizes and investigates the stopping sight distances, braking distances that allow for the evaluation of the vehicle's performance to increase the capability of both the driver and vehicle with regard to safety. A total of ten experiments were carried out with a passenger car, the Perodua Myvi, on a road within a local university. The results found that Stopping Sight Distance (SSD) could be used to determine the SSD graph by considering the velocity (v0), driver's perception reaction time (tR), coefficient of breaking friction (fT), gravitational constant (g), deceleration rate (a) and roadway grade (G). The study also concludes that SSD could be used to determine the SSD graph by considering the velocity (v0), driver’s perception reaction time (tR), coefficient of breaking friction (fT), gravitational constant (g), deceleration rate (a) and roadway grade (G). Ultimately, the study also indicates that the time to switch to hi-beam (Automatic High Beam – AHB) is at SSD distance (meter or second).&#x0D; &#x0D; &#x0D;

Probabilistic Geometric Design of Highways: A Review

The design input parameters such as vehicular operating speed, deceleration rate and driver’s perception reaction time considered in the geometric design of highways exhibit considerable uncertainty in practical scenario. As such, in this context, the probabilistic approach or reliability based approach is essential in highway geometric design which is capable of incorporating the factor of uncertainty of design input parameters. The objective of this paper is, therefore, to present a state of the art review on the works dealing with reliability based methodology in highway geometric design and discuss future scope of development. Prominently, in the present study, the process of probabilistic or reliability based design in the reference of sight distance considerations and horizontal alignment have been reviewed, and thereby, some recently developed parameters such as Lfront-eye (the distance from the front of the car to the driver’s eye) and considerations as reported in various literature are recommended to be introduced in the framework of reliability based highway geometric design practice. Further, it is recommended in line with the reliability based design practice to incorporate the methodology to establish the link between Probability of Noncompliance (Pnc) and collision frequency obtained from accident data. Also, it is recommended that a reliability based optimization technique can act as a viable tool for ensuring uniform level of safety on a reliability based designed road.

Empirical evaluation of drivers’ start-up behavior at signalized intersection using driving simulator

Abstract Start-up behavior at signalized intersection mainly depends on perception reaction time of drivers to the green phase. This study investigated the start-up behavior at signalized intersections by considering reaction time, acceleration and jerk (the rate of change of acceleration) of drivers in the state of Qatar. Distributions for reaction time, acceleration and jerk were plotted and the mean and 50th percentile values were presented. Three demographic factors (i.e., gender, ethnicity and age) were analyzed using two-tailed/unpaired t-tests. The relationships between acceleration and reaction time, and jerk and reaction time were investigated by linear regression analyses. Descriptive analysis showed that drivers had a mean reaction time of 2.91 s. Furthermore, Arab drivers had significantly lower reaction time than non-Arab drivers. Regarding the jerk maneuvers, young drivers (below 30 years) displayed significantly higher jerk than drivers of 30 years or above. Results from linear regressions showed significant negative correlations in both models (i.e., reaction time on acceleration, reaction time on jerk). As this study targeted multi-cultural drivers’ population, the results of reaction time and jerk distributions could be used as inputs in simulation models which are developed for evaluating driver behavior and safety at signalized intersections in regions with multi-cultural driving population.

Cognitive Load Variability from Road Characteristics Should Influence a Safety Requirement for Vehicle Stopping Sight Distance

Current U.S. geometric road design standards are based on a prescribed value for a driver’s perception-reaction time (a constant value of 2.5 seconds), which represents the time necessary for a driver to safely stop the vehicle to avoid a crash (referred to as a stopping sight distance); however, these standards fail to consider how road complexity, driver risk perception, and visual stimuli can influence perception-reaction time. With over a million vehicle fatalities a year (WHO, n.d.), it is necessary to investigate methods of improving driver safety. The influence of road characteristics is considered with some road design policies, but not currently applied to stopping sight distance. This paper introduces theoretical considerations for increasing perception-reaction time (and thereby adjusting speed limits or road geometry) based on roadway complexity (volume of vehicles, road geometry, pedestrian crossings, frequency of adverse weather conditions, or other conditions).

A machine-learning based time constrained resource allocation scheme for vehicular fog computing

Through integrating advanced communication and data processing technologies into smart vehicles and roadside infrastructures, the Intelligent Transportation System (ITS) has evolved as a promising paradigm for improving safety, efficiency of the transportation system. However, the strict delay requirement of the safety-related applications is still a great challenge for the ITS, especially in dense traffic environment. In this paper, we introduce the metric called Perception-Reaction Time (PRT), which reflects the time consumption of safety-related applications and is closely related to road efficiency and security. With the integration of the incorporating information-centric networking technology and the fog virtualization approach, we propose a novel fog resource scheduling mechanism to minimize the PRT. Furthermore, we adopt a deep reinforcement learning approach to design an on-line optimal resource allocation scheme. Numerical results demonstrate that our proposed schemes is able to reduce about 70% of the RPT compared with the traditional approach.

Cognitive and Physical Factors Affecting High-Speed Driving Ability in Older Adults

Background and Purpose: The ability to drive is important for the elderly to maintain their mobility and independence. Therefore, to investigate age-related changes in driving-related functions and their relation to the braking ability of drivers driving at high speeds, subjects were organized by age into study groups (younger, middle-aged, and older). Methods: For a correlational comparative study, 36 male participants were recruited and instructed to complete the Assessment of Driving-Related Skills (ADReS), lower limbs muscle test, and driving tasks on a driving simulator under 3 speed conditions. For the driving tasks, 3 event phases were analyzed: perception-reaction time (PRT), brake-movement time, and fast pressing time. Results: Older drivers demonstrated a significant decline in certain driving-related cognitive and motor abilities, as well as hip flexion and knee extension muscle strength, and exhibited a significantly longer PRT under moderate- and high-speed conditions compared with younger drivers. A longer PRT under moderate- and high-speed conditions was significantly associated with poor driving-related cognitive and motor abilities. Conclusion: The increased PRT for older drivers under moderate- and high-speed conditions might increase their risk of vehicular accidents on highways.

Review and design policy of sight distance types for the interurban highway system

The paper introduces design policies, model developments and the implementation of sight distances (SDs) in the interurban highway network in Israel. The SD types are stopping sight distance (SSD), decision sight distance (DSD), passing sight distance (PSD) and constrained passing sight distance (CPSD). In Israel, the implementation of SDs was resolved on the basis of highway categories included in the interurban highway system. The paper reviews and documents the developments of highway design parameters and constructed models of the SD types, specifically the perception–reaction time and deceleration coefficient required for SSD, DSD model formation and innovative concepts related to CPSD based on PSD model components. Compared with US design policy, the Israeli design policy resulted in slightly lower design values for SSD, DSD and PSD, but higher CPSD design values. The paper also refers to SD considerations that might impact the construction stages of a highway project. The developments presented are an outcome of the new Israeli geometric design policy and controls for rural (interurban) highways.

Quantifying visual road environment to establish a speeding prediction model: An examination using naturalistic driving data

Speeding is one of the major contributors to traffic crashes. To solve this problem, speeding prediction is recognized as a critical step in a pre-warning system. While previous studies have shown that speeding is affected by road environmental design, research in predicting speeding behavior through road environment features has not yet been conducted. Furthermore, there is a large discrepancy between actual and perceived road environmental information given that a driver’s visual perception plays a crucial role as the dominant source of information in determining driver’s behavior. Thus, this paper aims to establish a speeding prediction model based on quantifying the visual road environment to improve the design of pre-waring systems, which can predict whether drivers are going to speed and provide them with visual or/and audio warnings about their current driving speed and the speed limit prior to the occurrence of speeding behavior. Twenty input variables derived from three categories including visual road environment parameters, vehicle kinematic features, and driver characteristics were considered in the proposed speeding prediction model. Especially, the road environmental design factors consisting of the visual road geometry and visual roadside environment as perceived by the driver’s eyes were quantified using a visual road environment model. Field experiments were conducted to collect naturalistic driving data concerning speeding behavior on the typical two-lane mountainous rural highways in five provinces of China. Random Forests, an ensemble learning method for regression and classification, were applied to build the speeding prediction model and variable importance was calculated. Additionally, logistic regression was used as a supplement to further investigate factors impacting on speeding behavior. A speeding criterion was defined with two levels in this study: a lower level (exceeding the posted speed limit) and a higher level (10% above the posted speed limit). Under both levels of the speeding criterion, the speeding prediction model performed well with high accuracy (over 85%). This model could use the value of the variables obtained from the current position to predict drivers’ speeding behavior at the future position located a sighting distance away. This interval was sufficient for a pre-warning system to give a speeding warning that a driver with normal perception-reaction time (around 2.5 s) could respond to. Findings in this study can be used to effectively predict speeding in advance and help to reduce speeding-related traffic accidents.