Exploring the sustained and divided attention of novice versus experienced drivers

Driving is a complex task as it requires continuous information processing in a constantly changing environment. The dynamic nature of driving necessitates is at a high level of attention, accuracy, and vigilance for safe performance. Therefore, appropriate engagement, maintenance, and regulation of sustained attention are critical to driving safety. In addition, driving at times requires the simultaneous information processing of multiple tasks and, therefore, appropriate engagement in, and regulation of, divided attention is also necessary for safe driving.

Sustained attention is the ability to maintain concentration for a prolonged period. for achieving this, there must be intrinsic maintenance of an individual’s alert state, where there is no external input. This is ability to sustain attention and redeploy this attention flexibly in response to relevant environmental factors is necessary to avoid adverse driving events. Unfortunately, the ability of individuals to maintain a high sustained attention performance in repetitive and monotonous tasks such as driving is limited. The significance of this is the paramount because the failure to appropriately sustain and regulate attention can result in attention lapses (absent-minded behaviors resulting from transient reductions in the sustained attention of the driver;  driving errors (inadvertent and inadequate actions resulting from observational failures due to lapses in sustained attention and consequently accidents.

While driving is considered to be a highly complex task (the following learning to drive), it can be perceived as second nature and an automatic process. Due to this it is perception and the repetitive and monotonous nature of driving, many drivers feel it is acceptable to perform a secondary task while driving, which may affect driving competence and safety. This is known as distracted driving, and it requires the utilization of divided attention.

Divided attention involves processing and responding to specific information while at the same time engaging in multiple tasks. Driving under distracting conditions greatly relies on an individual’s executive attentional resources. The Yerkes-Dodson law states that performance and arousal have an inverted-U-shaped relationship. As arousal increases, performance will improve up to a certain threshold and then deteriorate. Secondary tasks or distractors may increase arousal and performance initially, but the performance may not be sustained. Additional distractors can further compete for the neural resources required for safety in driving. Although initially heightened arousal can improve performance, the added cognitive load may result in performance deterioration, which increases the risk of motor vehicle accidents.

Statistics have shown that VMAs can be committed by individuals of a wide variety of ages, which indicates that both novice young adult drivers and experienced adult drivers commit VMAs. Research has found that novice young adult drivers have a higher motor vehicle crash rate in motorized countries than in any other age category, which is an alarming statistic. For example, in Australia, the road mortality rate in 2019 for 18- to 20-year-olds was 9 in every 100,000 persons compared to an average of 4.7 for the entire population. Furthermore, the literature has reported that a substantial number of major VMAs are the result of momentary distractions or lapses in sustained attention (Dockree et al., 2005). Distracted driving – for example, using a mobile phone while driving – contributes to a significant percentage of drivers’ behaviors causing road safety issues 

Distracted driving occurs in motorists of all ages. Increased task complexity is shown to delay appropriate responses and affect driving performance risks, and such influence would decrease with more experience. As such, a greater number of years of driving experience has been shown to correlate with better driving performance, where there are fewer errors and attention lapses. Selective attention is found to improve with age and is significantly associated with fewer driving risks. Increased cognitive control could potentially reduce risk-driving behavior in novice drivers. The higher rate of VMAs in young novice drivers could be related to less skilled use of sustained and divided attention that is necessary to avoid accidents arising from attention lapses and errors due to distracted driving or other reasons. A review of the literature revealed that research had been done on the factors affecting driving performance in novice young and experienced drivers, but less research reports a comparison of the sustained- and divided-attention function of novice young adult drivers with that of experienced adult drivers. Thus, it is apparent that there is a critical need for research on this topic to address this knowledge gap and to provide insight into driver education.

This research aimed to explore the sustained and divided attention of novice young adult drivers compared to that of experienced adult drivers. The Sustained Attention-to-Response Task (SART) was used as the main tool for assessing attention, with the addition of a secondary task with low and high levels of cognitive workload being introduced into SART for assessing divided attention. Participant attention function was also assessed by the Color Trails Test (CTT) and Digit Span test (DST). It was hypothesized that the sustained- and divided-attention performance of novice young adult drivers would be significantly lower than that of experienced adult drivers. It was further hypothesized that the low-load condition would result in a higher SART performance compared to the negligible-load condition. The high-load condition would produce performance deterioration due to a too-high arousal level in both groups.

Design and its  procedure:

The participants were requested to have seven hours of sleep the night before the assessments to ensure they were fatigue-free. The data collection process was conducted in a quiet private room that provided a minimal distraction. The participants’ demographic and driving history information was collected, followed by the administration of the CTT and the DST.

The RSME was then administered, which was followed by the first block of the SART and the second RSME administration. A second SART block was then conducted, followed by the final RSME administration. The total time required was 60 to 90 min per participant. Each participant received an AUD$40 gift voucher as compensation for their time and travel expenses.

Sustained attention to the sponse to the task:

• The SART is a task used frequently to examine sustained attention (Robertson et al., 1997). For this study and research, the conventional fixed SART design was adopted to examine sustained attention in terms of being a “negligible-load” (NL) condition. The fixed SART design was modified to include two secondary tasks – the “low-load” (LL) and “high-load” (HL) conditions.
• This modification allowed for the divided attention of participants to be examined.
• A single block of trials consisted of 648 trials, allowing for 71 trials per digit.
• Each block consisted of the conventional SART with NL demand and a modified SART with the LL and HL conditions. 
• The participants were used for required to complete two blocks each, which resulted in 1296 single digits. Blocks were randomly selected from a pool of six blocks by rolling a six-sided dice. Each block lasted 20 min and there was a short rest break as required mid-block. A break of five to 10 min was allocated between blocks.
• The participants were instructed to respond as fast as possible in both blocks.
• They were given time to familiarise themselves with the assessment before it commenced by completing a two-minute practice test. The conventional SART for assessing sustained attention: Having adopted the conventional fixed SART design, the digits one to nine appeared centrally on the laptop screen, one at a time, in ascending order.
•  Each digit was also presented for a period of 150 ms. The participants were instructed to click the left mouse button using their right forefinger for digits one to nine for go trials, excluding digit three. The participants were requested to inhibit a response when digit three appeared as these were no-go trials. This formed the NL condition. 
• 
  To prevent the participants from developing a habitual rhythmic response pattern and succumbing to a speed-accuracy trade-off, each digit was followed by an inter-stimulus interval that randomly varied between 1000 ms and 1500 ms. Five random digit sizes were presented: 100, 120, 140, 160, and 180 in Arial font. This was done to increase the processing demands and decrease the likelihood of the participants developing a personal search template for target trials.

The modified SART to access divided attention:

The modified SART with the addition of secondary tasks was designed to study the participants’ divided attention. The block components of the modified SART were separated into sub-blocks, the length of which was randomly selected from 27, 36, and 45 trials. Two cases of sub-blocks – the LL and the HL conditions – were included to study the participants’ divided attention. The LL and HL conditions represented the two levels of cognitive workload. The LL condition posed a low cognitive workload, which required memorizing a five-digit number composed of the same five digits – for example, 77777. The HL condition posed a high cognitive workload, which required memorizing a five-digit number consisting of random and unrepeated digits – for example, 31489. For both LL and HL conditions, the five-digit number was presented in the screen center at the start of each sub-block. The participants were required to memorize and rehearse the number presented during the sub-block. At the end of the sub-block, another five-digit number was presented, and the participants were required to indicate if this number was the same as the number previously presented. The participants were required to respond with a left mouse click if the number was the same and a right mouse click if the number was different.