Highway safety is a crucial concern globally, particularly in developing countries that are undergoing fast urbanization and infrastructure development, such as India. Globally, India ranks fourth in road crashes and first in road fatalities (World Health Organization (WHO), 2023). According to the Ministry of Road Transport and Highways (MoRTH) (Annual Report, 2022) report, about 0.46 million crashes occurred in 2022, of which 33.80% were fatal. According to MoRTH data (2020-2022), the number of crashes has steadily increased by 24% since 2020. As per the Highway Safety Manual (HSM, 2009) about 65% of the crashes are attributed to human factors. A similar scenario can be seen from the Indian dataset, showing 69% of the drivers as over-speeding or human error (Ministry of Road Transport and Highways (MoRTH), 2023). The age groups <18-25 and 25 to 35 account for 25.5% and 25.3% respectively of all fatal road accidents for the year 2022 (Annual Report, 2022).

There are numerous elements that contribute to human factors, including physiological and psychological characteristics such as vision, hearing, and reaction times (Bayat et al., 2024). Driving is a complex task that involves a wide range of cognitive processes, such as perception, attention, learning, memory, decision-making, and action control. The cognitive processes are associated with human errors (Saurin et al., 2004) that consist of mental abilities that deal with learning, problem-solving, remembering, decision-making and time and motion anticipation (G. G. Fisher et al., 2019). Among these, anticipations (time and movement) are key in hazard perception and safety while driving, as they deal with the cognitive ability to apprehend the speed and distance of moving objects (Pelser et al., 2005; Stahl et al., 2016). Considering the time and movement anticipation, a driver can comprehend the speed of the vehicles with the change in the image in the eyes with respect to time, called speed perception (HSM, 2009). This is attributed to the perception in depth, which is essential for making decisions in critical situations, such as driving through curves and turns, at intersections, merging lanes, around roadside objects, in changing road conditions, and at pedestrian crossings. Perception in depth relies heavily on accurately calculating distances and spatial relationships between elements in the driver's environment. It allows drivers to measure their distance from other vehicles, pedestrians, obstacles, and side features. The ability to accurately perceive and judge the speed of one's own vehicle as well as the proximity to other cars on the road is necessary for maintaining a safe driving speed and responding to changing traffic conditions. Individual characteristics, environmental factors, and visual signals can all affect speed perception. There was a 42% rise in traffic crashes at various junction types between 2020 and 2022, particularly T-intersections, Y-junctions, and 4-arm junctions classified as merging areas as per MORTH (MoRTH 2023, 2022). On the other hand, the contribution of 2-wheelers and car drivers has consistently risen from 2020 to 2022, i.e., 38.8% to 43% for 2W, and for cars was approximately 19 percent. Hence, given the growing traffic volumes and diverse road users, understanding the linkages between human factors and highway safety is a crucial part.

While driving, drivers can make various types of errors, including decision errors, recognition errors, and performance and non-performance errors (Monyo et al., 2021; Shaon et al., 2025). Decision errors can involve too fast a reaction for a condition (e.g., breaking due to overestimation of speed), or the failure to acknowledge the unexpected action of the other user (e.g., other drivers' lane changing operation or pedestrian crossing roads). These decision errors are mostly due to time and movement anticipation, as cognitive functions are used for prediction and decision-making (Seyfzadehdarabad et al., 2023). The time and movement anticipation is a cognitive ability rather than a personality trait (Seyfzadehdarabad et al., 2023). These decision errors caused about 54% of driving errors and hence are necessary to evaluate. Also, these errors are affected by the driver’s gender, age, driving experience, cognitive capabilities, and other distractions such as the use of mobile phones (Kim et al., 2023; Tinella et al., 2024). Using a phone while driving divides attention, significantly affecting driving safety (Rahman et al., 2021; Rajesh et al., 2017). This cognitive distraction caused by phone use affects driving performance and the sending and receiving of text messages, resulting in a significant decrease in speed and difficulty maintaining a stable speed and headway distance (Haigney et al., 2000; Oikonomou et al., 2025).

Thus, it can be concluded that human error is a major factor in most highway crashes. Hence, the study of human cognitive abilities can help reduce potential human errors. The Schuhfried company developed the Vienna Test System (VTS), a computerized psychological assessment tool, in the 1980s (Schuhfried, 2013). VTS is a systematic, objective testing procedure with extensive criteria that is utilized for psychological diagnostics globally. The VTS is a comprehensive psychological assessment (also called a neuropsychological assessment) tool designed to evaluate cognitive abilities, personality traits, and specific competencies. For accurately measuring sensory, motor, and cognitive responses, tests designed in VTS use specialized input/output devices (earphones, response buttons on the panel and pedals). Pictures, lights, sounds, or animations controlled with precise timings can all be used as test stimuli. This flexibility marks VTS well-suited to simulate driving-related tasks (like responding to traffic signals, perceiving brief scenes, or tracking moving objects) under controlled conditions (Grundler & Strasburger, 2020). Some applications of this test battery include the recruitment of marine pilots, high-speed rail drivers and drone operators, and the assessment of athletes' skills, etc (Guo et al., 2019; Johnsen et al., 2023; Seyfzadehdarabad et al., 2023; Wilczyńska et al., 2021). In Austria and Germany, a traffic psychology assessment is required as part of the medical-psychological test for regaining a driving license after being removed due to accidents or serious traffic violations (Sommer et al., 2008). Psychometric tests have long been used to predict safe driving performance in many countries worldwide. Medical-psychological tests are used to evaluate cognitive abilities (e.g., in Austria, Spain, and Hungary) and to select the most qualified applicants in transportation firms. Psychometric assessments have stayed mostly similar for decades, despite significant changes in truck and bus technology, as well as overall road safety infrastructure (Vetter et al., 2018; Vujanić et al., 2016).

One of the key features of VTS is the ability to assess time and movement anticipation. These abilities are essential for driving activities, including passing, estimating distance, and responding to changing traffic. The Time and Movement Anticipation test (code-named ZBA) is the main VTS module for this. The exam assesses both spatial and temporal anticipation (i.e., "where will the object be at that time?" and "when will the object arrive?"). This closely resembles the demands of driving in the real world. For example, a driver must anticipate the speed and distance of approaching vehicles while overtaking. Likewise, maintaining a safe following distance involves estimating how long it will take the vehicle ahead of you to reach a specific location. It's stated that the ZBA test evaluates the ability to estimate the speed and direction of object movements, and that accurately assessing speeds and distances requires spatial and temporal orientation. Poor performance (significant time or position errors) indicates a lack of ability to anticipate motion, a characteristic that has been connected in studies to odd driving habits and collision involvement.

To the best of the author's knowledge, only a handful of studies have shown a relation between novice driver training and reductions in crashes and near-crashes (Epstein et al., 2022; Fisher et al., 2017, 2024). These training programs have focused either on hazard anticipation (Crundall et al., 2012) or attention maintenance (Pradhan et al., 2011). Further, Zhang et al. (2018) found that the training failed with careless drivers but worked with careful drivers. In a developing nation like India, there are very few studies on the time and movement anticipation of young drivers and the factors that affect them. This study will illustrate the underrepresented driving context in previous studies of greater traffic heterogeneity, along with non-lane-based traffic conditions. Developing nations like India have highly heterogeneous traffic with weak lane discipline, reflecting driver behaviour that can accept narrow gaps and still manoeuvre smoothly. For this type of driver behaviour, the time and movement assessment has not been well studied. Also, crashes involving motorised two-wheelers are increasing worldwide, and about 60-70% of these crashes are due to errors in anticipation (Jaffard & Elslande, 2012). Further, if the patterns and factors affecting time and movement anticipation are well known, the driver’s training module can be prepared accordingly; hence, the present study examines decision errors related to time and movement anticipation among young drivers in India. The effects of gender, age, vision, driving experience, driving license possession, and mobile phone use on decision errors using VTS are studied, which can help develop driver training programs and interventions that are effective and tailored to the needs of this demographic.

The novelty of the present study is as follows

Travelling at high speed can increase the risk of a crash if the driver fails to account for the speed of the upcoming vehicle or its distance from the vehicle. If the driver fails to perceive these factors, stopping within the safe stopping distance will not be possible, leading to a collision with the object/vehicle. As per the study by Pascal, driving at fast speeds can have catastrophic repercussions, since drivers may fail to foresee an incoming hazard and lose control of their car, resulting in a crash (Cœugnet et al., 2013).

The study by Hunt et al. (2011) highlighted the significance of gap-acceptance judgments for road user safety, emphasising the need to evaluate both the distance and speed of oncoming cars. It emphasizes that people frequently depend entirely on distance when making gap estimations, ignoring the speed of the oncoming vehicle, which can lead to dangerous actions, particularly when the oncoming vehicle is driving unusually fast. Jurecki and Stańczyk (2011) subjected 30 drivers to three different tests in order to evaluate their behavior prior to a collision. The first test, a psychological exam, assessed drivers' simple and complex reaction times. In simple reaction-time tests, drivers press a key when a stimulus (a light) appears on the screen. Complex reaction time tests require drivers to press a key when a specific colour of light appears on the screen. The second test was conducted on a test track where drivers are forced to react to prevent a collision with another car at a crossroads. The third test used a simulation in which the same scenario was replicated as conducted in the second test. The analytical report indicated that reaction time values on simulators are lower than those on the track, for both simple and complex reaction times. Mańkowska et al. (2015) studied the effects of visual perception on reaction time, anticipatory time, and movement in basketball players. Only females were considered for the purpose of the study. They concluded that time and movement anticipation are necessary for any sports person to handle complex situations effectively, and that this ability is mostly influenced by an individual's perceptual abilities. Horváth et al. evaluated the effect of the six-week training programme on car racing drivers using VTS. They found that the drivers were more accurate on word-reading tests during exercise compared with rest conditions. Further, they found that car drivers' performance improved, with reduced reaction times. No significant differences were found in time or movement anticipation with training.

Stahl et al. (2016) studied the effect of the in-vehicle interface on the anticipation of novice and experienced drivers. They found that novice drivers show a significant effect on time anticipation. Wu et al. (2017) found that gender, driving experience, driving speed, and road type affect the speed perception of drivers. They found that inexperienced drivers make more errors on wider roads than on narrow roads. Further, they concluded that females make the most estimation errors at both slow and fast speeds. Saadat et al. (2018) evaluated the effect of waterpipe smoking on the psychomotor performance of drivers using VTS. They found that waterpipe smoking enhanced the two-hand coordination, concentration and attention. No significant differences were observed for time and movement anticipation. Another study by Saadat et al. (2019), involves hospital nurses who napped later during their night shift (from 3:00 to 6:00 AM) and demonstrated poorer time-movement anticipation abilities compared to those who napped earlier (from 12:00 to 3:00 AM). Since time-movement anticipation is essential for tasks such as safe driving, hazard management, and preventing road traffic collisions, these findings suggest that nurses who nap later in their shift may be at greater risk of road traffic accidents while commuting after work. Pelser et al. (2005) evaluated the strategy for selecting haul truck operators using the psychomotor test of VTS. They utilized Cognitrone, Determination Test, ZBA test and two-hand coordination tests. They found the positive skewness for ZBA and the two-hand coordination test, and hence, respondents performed well on these tests.

Mohanty and Samal (2019) used a questionnaire-based approach to study the role of young drivers in road crashes in the city of Bhubaneswar, India. The 15-25 age group was split into four groups. The study found that individuals between the ages of 18 and 21 are more likely to have catastrophic vehicle crashes. Over-speeding of vehicles accounting for 80 kmph and above is considered a disaster for riders. Moreover, not wearing a helmet increases the risk of a severe accident by approximately 400 times. Monyo et al. (2021) studied the factors influencing older driver errors. They found that gender, age, lightning conditions, location, alcohol involvement, distraction, etc., were factors affecting decision errors and recognition errors. They concluded that drivers are more likely to make decision errors (approximately 54%) than recognition errors. Babic et al. (2022) found that gender, road type, visibility conditions and vehicle speed affect the drivers’ speed perception. Mamcarz et al. (2023) examined the personality traits of the young driver as a link between the reaction time and traffic perception using the VTS. Various VTS tests, including Reaction Time (RT), ZBA, and others, were administered to 60 participants. They found that gender affects the personality factors, self-control, motor time, risk avoidance and time anticipation. The results showed that females had higher self-control scores but lower time-anticipation scores. Murphy and Voyer (2023) studied the effects of laterality and time estimation on accident proneness, using a sample of psychology undergraduate students. They found that the participants who made the positive error (underestimating time) were more susceptible to crashes. Wu (2024) found that speeds were perceived as lower when the vehicle moved in the same direction as the driver. Further, the author concluded that errors were more frequent in a simple road environment than in a more complex one. Marković et al. (2025) concluded that there was overestimation at lower speeds and underestimation at higher speeds in the study conducted at night. On the other hand, Papić et al. (2020) found that pedestrians underestimated higher speeds while crossing roads.

From the above-mentioned literature, it is observed that age, driving experience, and gender have significant effects on reaction time. However, the effects of age and gender, particularly on reaction time for time and movement anticipation, have not been well studied. Hence, the present study examines time and movement anticipation among Indian drivers. First, the distribution of time and movement anticipation deviation is studied, and then the effects of age and gender are explored for the dataset.

For the present study, the computerized psychological evaluation tool VTS, is used to measure several aspects of cognitive and psychomotor abilities. It comprises a variety of standardized tests intended to evaluate multiple aspects of a person's personality, emotional intelligence, and cognitive performance. VTS is used by researchers for cognitive studies of time and movement anticipation (Seyfzadehdarabad et al., 2023; Veliks et al., 2023). The VTS have high reliability for time and movement anticipation tests, ranging from 0.94 to 0.99 (Schuhfried, 2013).

Test units ZBA S5 and ZBA S2 module in VTS are used to evaluate time and movement anticipation for the present study. The details of the test are given below.

After obtaining the data, the deviation in reaction time (hereafter, called deviation time) is studied for both time and movement anticipation. The conflict and simulation analysis need the distribution of deviation times as an input (Astarita & Giofré, 2019), and hence an attempt is made to find the best-fitted distribution for the deviation time.

To fit the distribution of the speed perception test data, first, the family of distributions is prepared by studying the literature, and it includes Log-Normal, Normal, Weibull, Generalized Extreme Value (GEV) and Log-Logistic distributions as given in Equations (1) to (5).

Normal:

(1) f(x)=1σ2πe−12(x−μσ)2

where µ – mean, 𝜎 – standard deviation

Log-normal:

(2) (x)=exp⁡(−12(ln⁡x−μσ)2)xσ2π

where 𝜎 – shape parameter (σ>0), µ – location parameter.

Weibull:

(3) f(x)={kλ(xλ)k−1e−(xλ)kx≥00x<0

where k – shape parameter, λ – scale parameter.

Generalized Extreme Value (GEV):

(4) f(x)={1σexp⁡(−(1+k(x−μσ))−1k)(1+k(x−μσ))−1−1kk≠01σexp⁡(−(x−μσ)−exp(−(x−μσ)))k=0

where 𝜎 – shape parameter (σ>0), µ – location parameter, k – shape parameter.

Log-logistic:

(5) f(x)=(β∝)(x∝)(β−1)(1+(xα))β2

where α – scale parameter, β – shape parameter.

The Kolmogorov-Smirnov (KS) non-parametric test is applied to evaluate the goodness of fit (Bari et al., 2021; Dong et al., 2015). The KS value checks the difference between the cumulative frequency distribution (CDF) curve (C. S. Bari et al., 2022; Navandar et al., 2021). In the present study, a 90% confidence interval is considered. 90% confidence level is considered, as the p-values were marginally higher than 0.05 for all distributions. The null and alternate hypotheses for the present study are as follows.

H₀: the deviation time follows the specified distribution.

H₁: the deviation time does not follow the specified distribution.

The best-fitting distribution is the one with the minimum KS value. Then, the effects of age and gender on the deviation are critically examined.

The present study examines time anticipation (speed perception) and movement anticipation (deviation errors) using VTS. Younger participants are considered for the study, and the effects of age, gender, driving experience, license possession, vision, and mobile phone use are studied. In the study, the positive deviation, i.e., the participant presses the button before the ball crosses the line, is termed as overestimation of speed. On the other hand, the negative deviation is termed as underestimation of speed. The data on deviation time is found to follow the GEV distribution (Table 1). The shape factor was negative in all cases, indicating larger values farther from the line (Table 2).

The analysis of deviation time by gender revealed that females underestimate speed more than males do, considering all combined data and all three motions. This means that females have a slightly lower time-anticipation ability than males when observed as a combined group (without age variation). Similar findings are reported by Wu et al. (2017). The same results were obtained in a study by Wedagama & Dissanayake (2009), where female motorists and motorcyclists were more prone to fatal accidents than males. Considering age-wise variations, females in the 18 to 20 group overestimate more than males for slow and fast motion (Table 3). The statistical analysis showed that for slow and medium motion, there is a significant difference in gender-wise deviation times, but not for fast motion (expressways, NHs) (Table 4). For slow cases, such as urban or village roads, there is a significant difference in deviation times, and hence the variation in responses can be seen between males and females. Similar findings are reported in the study by Wu (2024), where higher errors are observed in low-complexity areas, such as suburban sections, while lower errors are observed in high-complexity areas, such as urban roads. On the other hand, Barat and Das (2025) found that as the speed of the two-wheeler increases, the speed estimation errors increase. Babic et al. (2022) found that the speed estimation is higher on rural roads. Also, they concluded that female respondents are more accurate in speed perception, which contradicts the present study and studies by Wu et al. (2017) and Hussain et al. (2019). Also, Garach et al. (2022) found that females perceived higher speeds better than males. Marković et al. (2025) found that female participants made more errors than male participants at lower speeds. Also, they found that participants had lower accuracy in speed perception on rural roads.

Considering the standard deviation, females show greater variability in deviation times across all three motions than males. It can be said that, though males underestimate speeds more in slow and medium motion (urban road/village road) conditions, they are underestimating speeds less on highways (fast case). On the other hand, in the underestimation case, the female shows higher median values for the low- and fast-motion cases. Similar results are observed by Larue et al. (2018). They concluded that for slower-moving trains, the speed perception errors were lower. Also, they found that the speeds were underestimated by upto 44% and thus concluded that people can’t appreciate that the trains can move faster. On the other hand, the present results are in contradiction with the results obtained by Marković et al. (2025). They found that the males are underestimating the speeds. These gender differences in speed perception are significant for lower-speed motion. Similar to this, Marković et al. (2025) and Papić et al. (2020) showed that males are underestimating speeds more than females as pedestrians.

In the 18-20 age group, a significant difference between genders is observed only for the overestimation of slow and medium motion. It can be seen that for all other cases of underestimation and fast overestimation, there is no significant difference in speed perception (Table 4). It means that the younger and novice drivers are exhibiting more or less the same behavior. Overall, female drivers tend to overestimate the speed of moving objects in space, both for slow and fast motion (Table 3). On the other hand, for the 21-30-year age group, significant differences are observed for slow and medium motion in both over- and underestimation cases.

It can be seen that with an increase in the experience level, the deviation times in the overestimation case decrease from slow motion to fast motion (Figure 4 (a)). Similar findings are obtained by Marković et al. (2025) and Papić et al. (2020). Further, at the same experience level, the overestimation increases from slow motion to fast motion. The present findings contradict those with Wu et al. (2017). They found that, with increased speed, the percentage of underestimation increases among those without driving experience. For those with driving experience, underestimation is observed only at higher speeds. The Kruskal-Wallis results showed a significant difference for the overestimation case, with deviation times in the driving age groups 1 to 4 and >11, and 5 to 10 and >11, for both slow and medium motion (red part of Table 6). For fast motion, no significant difference was observed.

Further, those who don’t have a driving license show a significant difference in speed estimation for the slow motion (Table 7). The effect of the phone is seen for the underestimation case (Table 7). The observations are based on the VTS and questionnaire survey; hence, the authors acknowledge that there may be a difference in the actual variation in deviation time due to the phone case. Helmut Schrom-Feiertag et al. (2019) found that pedestrians under-estimate speeds in the virtual environment as compared to the field. Hussain et al. (2019) found that simulator and field speeds vary, but the tendency and distribution remain the same.

The present study's findings suggest that the policy may consider including the time and movement anticipation test in the driving license test. Those with good speed perception can judge the speed of other vehicles and pedestrians also control their own speed. Specific speed perception programmes can also be justified for the target population (like young drivers). These implications are exploratory in nature and should be interpreted with caution until replicated in larger samples.

The present study was carried out to evaluate the factors that affect the time and movement anticipation of young drivers. The data from the ZBA S5 test in which participants who overestimated the speed follow a GEV distribution. Data analysis revealed that the positive error deviation for male participants was greater than for female participants, indicating speed overestimation. Among the 18-20 age group, female participants overestimated speed to a greater extent than male participants. With increased driving experience, the tendency to underestimate speed increases. Further, novice drivers make more errors at slower speeds. Similarly, no significant difference in error deviation was found between participants with normal vision and those with spectacles. Phone users made more errors in deviation time than non-users when estimating speed in slow motion. The results remained the same for ZBA S2 across simple and complex test progressions and cumulative tests, indicating that using a phone decreases the cognitive attention required for driving, thereby increasing crash probability, as per the literature.

The authors acknowledge that results may vary with more samples. Further, in the present study, univariate analysis was carried out across different factors as suggested in previous studies (Kerautret et al., 2023; Liu et al., 2021; C. Wu et al., 2017; C. C. Wu, 2024). In future studies with larger data samples, multivariate analysis, including interaction effects, can be conducted. In the future, the same methodology can be applied to different driver types (e.g., bus drivers, truck drivers) and to different travel frequencies. Also, the effect of anger and aggression can be checked (Nergiz, 2025). Furthermore, these tests can be useful for reaction-time modelling. Real-time accident analysis can be performed using the parameters from the anticipation results.

Hrishikesh Deshpande: Data curation, Software, Visualization, Writing – original draft, Writing – review & editing. Chintaman Bari: Conceptualization, Data curation, Funding acquisition, Software, Supervision, Visualization, Writing – original draft, Writing – review & editing. Ashish Dhamaniya: Resources, Supervision, Visualization, Writing – original draft, Writing – review & editing.

The authors report no competing interests.

During the preparation of this work the authors used Grammarly in order to have grammar checking. The output was reviewed and revised by the authors who take full responsibility for the content of the publication.

This manuscript presents original work that has not been previously published or disseminated in any form.

The current study, conducted with human participants, was reviewed and approved by the Internal Research Committee (IRG) of the Mahindra University, Hyderabad, in accordance with the funding and data collection (IRP/2024/3). It was advised to take informed consent from participants during data collection, and it was followed strictly. Participants provided their written informed consent to participate in this study.

No external funding was used in this research.

Data sharing is not applicable, as no datasets were generated or analysed for this study.

Code sharing is not applicable, as no custom code or scripts were developed or used for this study.

Handling editor: Amir Pooyan Afghari, Delft University of Technology, the Netherlands.

Reviewers: Anshuman Sharma, Indian Institute of Technology Varanasi (BHU), India; Muhammad Adnan, Hasselt University, Belgium.

The handling editor takes full responsibility for the final acceptance decision. The editor's decision may differ from the recommendation made by one or more reviewers.

Submitted: 27 January 2026; Accepted: 2 October 2026; Published: 23 October 2026.