Impaired Cognitive Performance and Driving: A Comparative Study of Older Adults and Younger Adult Drug Users

Reyhaneh Bakhtiari ^1,*, Michelle V. Tomczak ^1,*, Stephen D. Langor ¹, Aaron Granley ², Farah Campbell ², Anthony Singhal ^1,3,*

1. Department of Psychology, University of Alberta, Edmonton, Canada

2. Impirica, Edmonton, Canada

3. Neuroscience & Mental Health Institute, University of Alberta, Edmonton, Canada

* Correspondences: Reyhaneh Bakhtiari, Michelle V. Tomczak and Anthony Singhal

Academic Editor: Ute Brüne-Cohrs

Special Issue: Driving Safety in Healthy Aging and Age-Related Diseases

Received: September 02, 2025 | Accepted: January 27, 2026 | Published: February 05, 2026

OBM Geriatrics 2026, Volume 10, Issue 1, doi:10.21926/obm.geriatr.2601334

Recommended citation: Bakhtiari R, Tomczak MV, Langor SD, Granley A, Campbell F, Singhal A. Impaired Cognitive Performance and Driving: A Comparative Study of Older Adults and Younger Adult Drug Users. OBM Geriatrics 2026; 10(1): 334; doi:10.21926/obm.geriatr.2601334.

© 2026 by the authors. This is an open access article distributed under the conditions of the Creative Commons by Attribution License, which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original work is correctly cited.

1. Introduction

Normal aging is associated with cognitive decline affecting driving competency [1]. The process of aging entails various psychological and physiological changes and is often shown through an overall decreased processing speed combined with increased forgetfulness in a short and long term memory association [2]. Transport Canada reported that 34.4% of licensed Canadian drivers are over the age of 55 and actively driving while driver fatalities increase after the age of 65 and more fatalities are caused by male older drivers compared to female older drivers [3].

The most affected sensory function in older adults is vision due to the deterioration of the visual field resulting in a decrease of visual sensory information processing [4,5]. Once visual acuity worsens, stationary and moving objects cannot be perceived accurately and efficiently which has been linked to an increase in crashes in older adults [1]. Older adults commonly report visual difficulties in daily tasks that are also related to problems they experience while driving, such as decreased visual processing speed, sensitivity to light, dynamic vision, near vision and visual search, problems driving at night due to decreased night time vision, etc. [6].

Drug-impaired driving has become a pressing issue in North America, particularly with cannabis legalization. Approximately 8% of drivers have ingested recreational drugs, primarily cannabis, with prevalence highest among younger drivers (20-24 years old) [7]. In Canada, drug-impaired driving incidents can occur approximately every 45 minutes, with 43% increase from 2018 [8]. According to a study by the Canadian Centre on Substance Abuse (2000-2008), nearly 35% of fatally injured drivers in Canada tested positive for drugs, including both legal and illicit substances [9]. It is increasingly clear that psychotropic (capable of affecting the mind, emotions or behavior) drugs contribute to impairment in driving performance. It has been estimated that at least 10% of all people killed or injured in crashes were taking psychotropic medication, which might have been a contributory factor to the crash.

Cannabis and cocaine are the most frequently detected substances in drug-impaired drivers, after alcohol [10]. Research shows that psychoactive substances, including cannabis and cocaine, induce long-lasting neuroadaptations at the molecular, cellular, circuitry, and behavioral level [11]. Driving performance deteriorates with increased cannabis or cocaine levels, influenced by factors like consumption method, tolerance, age, dosage and physiological absorption [12,13]. Consequently, drug-induced cognitive impairment elevates the risk of automobile collisions, injury, and fatalities.

Cannabis use, whether acute or chronic, impacts higher cognitive functions, including memory, attention, processing speed, executive function, as well as impulsivity and inhibitory control [13,14]. Recreational drug and alcohol users are being studied while displaying “short term” cognitive impairment [15]. Besides cognitive decline being correlated with age, literature suggests that the specific form of cognitive impairment such as drug induced cognitive impairment can adversely affect driving performance [1,16]. Further, literature also displays differences between short-term and long-term drug users where neuropathological changes are taken into account and how these changes may adversely affect the overall driving performance, either while under the influence, or during wear off stages [12,15]. Recreational drug and alcohol use is a crucial topic when studying cognitive impairment. Users often desire the rewarding effects and feeling of alcohol and drugs leading to continuous use often resulting in addiction. The nucleus accumbens, a brain structure located in the basal forebrain, is linked to other brain structures involved in dopamine and serotonin release related to recreational drug use.

Research has shown that cannabis and cocaine use significantly impairs driving abilities, posing substantial road safety risks [8,17]. In Canada, for instance, 27% and 16% of driver fatalities tested positive for cannabis and cocaine, respectively, in 2019 [18]. Cannabis use specifically impairs driving stability, leading to increased lane position variability, steering wheel position variability, and inappropriate line crossing in a dose-dependent manner up to five hours after consumption [19].

Urine tests are commonly used in forensic settings to identify individuals who have recently used these substances. Notably, cannabis and cocaine testing have specific detection periods and characteristics that need to be considered when evaluating the results from such tests. For instance, THC’s inactive metabolite THC-COOH can be detected in urine for 3-4 days in occasional users and up to 90 days in chronic heavy users, while cocaine’s metabolite benzoylecgonine (BE) can be detected for 2-4 days after use. Despite the psychoactive effects wearing off in these time frames, frequent use can lead to lasting changes in cognitive abilities, impacting driving performance [20]. Despite ample evidence showing clear degradation in the ability to drive safely due to both acute and chronic cannabis and cocaine use there is a need to investigate the impact of frequent use of these substances on cognitive performance important for driving.

Theories of cognitive functioning and driving impairment due to age related disease or recreational drug use coherently agree that driving is a combination of complex psychological and motor processes that heavily rely on cognitive functioning [1,9]. Age related cognitive decline and deterioration in cross motor skills are associated with an increase in accident risk connected to automobile fatalities [16].

The present study aims to compare the cognitive performance of frequent cannabis and/or cocaine users with healthy younger and older adults, using a set of tablet-based cognitive tasks that have been shown to predict on-road driving behavior [21,22,23]. In the case of [21], the tablet-based cognitive screen predicted driving performance in elderly drivers, some of which were at-risk for dementia. We used a similar cognitive screen to investigate the effects of frequent drug use on reaction time, judgment, memory, and motor control, and compared the performance of frequent drug user groups with healthy controls to better understand the potential implications for driving safety.

The participants with positive drug results were primarily frequent users recruited from a drug recognition evaluator training center. Based on the positive urine tests, indicative of recent use, and their frequent user status, we inferred that these individuals’ fitness to drive is likely compromised. By incorporating older adults that are actively driving, this study enables a critical comparison between the impacts of aging and frequent substance use on driving performance, despite the natural cognitive decline associated with aging [24].

In this study, the following hypotheses were investigated:

Hypothesis 1: Differences in performance among frequent drug user groups.

Hypothesis 2: Older adults will outperform frequent drug user groups.

Hypothesis 3: The control younger group will outperform both, frequent drug user groups and the control older group.

2. Method

2.1 Participants

In the present study we compared performance on the tablet-based cognitive tasks in five participant groups:

1) adults with a positive cannabinoids urine test (CAN group).

2) adults with a positive cocaine urine test (COC group).

3) adults with a positive cannabinoid and cocaine urine test (CAN/COC group).

4) healthy younger adults age 18 to 25 (Ctrl-Younger group).

5) healthy older adults age 65 and above (Ctrl-Older group).

2.2 Recruitment and Data Collection Process

The drug using group data was collected in a drug recognition evaluator (DRE) training facility in Jacksonville, FL where police officers are trained to screen participants for possible drug use and then perform a urine toxicology to confirm recent drug consumption and possible impairment. Participants were recruited via advertising the study at this facility and word of mouth. Most of these individuals were frequent drug users and completed a self-report questionnaire screening for past and current drug use, and medical conditions such as epileptic and cardiac conditions. After completing intake paperwork and consent within 2-4 hours after arrival, participants completed a urine toxicology panel screening for substances present in the urine (MedTox EZ Screen Cup, Joldon Diagnostics BioCup Panel 12 and the Alcopro Gabapentin Urine Dip Drug Test). The intake paperwork also screened for driving history, drug usage, medication use, and consent for urine toxicology and study participation. The tablet based cognitive assessment tool was also explained ahead of time and included in consent forms. Once the cognitive tasks were performed, a trained facilitator collected the tablet based data with each participant individually and explained each step of the process accordingly, therefore participants were monitored during their performance and were able to ask questions throughout the entire data collection process. Handy sampling was used to facilitate this study that shows similarities to a field research study.

Based on the results of the urine test, the following three groups were identified and chosen for the current study: the cannabis group (CAN, n = 114, f = 19, m = 95, other = 0, 37.41 ± 11.69) included participants with only a cannabinoid positive urine result, the cocaine group (COC, n = 60, f = 13, m = 47, other = 0, 52.45 ± 7.86) included participants with only cocaine-positive urine tests, and the cannabis and cocaine group (CAN/COC, n = 162, f = 24, m = 137, other = 1, 47.99 ± 12.94) included participants with urine tests positive for both cannabinoid and cocaine . The young adult control (Ctrl-Younger, n = 278, f = 186, m = 90, other = 2, 19.89 ± 5.36) group of undergraduate students was recruited through the University of Alberta. At the time of participation (2018), the Ctrl-Younger and Ctrl-Older groups were not screened for drug use via a urine test. However, cannabis was not legal in Canada at this time and self-report suggested that no drugs were consumed.

The control older adult (Ctrl-Older, n = 47, f = 22, m = 25, other = 0, 66.06 ± 6.85) group, consisting of active drivers over the past six months, was recruited as a part of our previous study. The inclusion criteria for this group were an active driving record (each participant was actively driving within the past six months), a valid driver’s license, and aged 65 years and older [21]. Table 1 and Figure 1 show the gender and age information for each group. No information about their educational background was collected.

Table 1 Demographic Distribution.

Figure 1 Age Distribution by Group.

2.3 Tablet Based Cognitive Test Battery

We utilized a set of tablet-based cognitive tasks (TBCT) to assess driving-related cognitive performance (Table 2). Developed by Impirica, a Canadian company in the business of driving assessments, these tasks formed a continuous series designed to evaluate various cognitive skills. As shown in previous studies [22,23], the tasks used in this research have been effective in discriminating between safe and unsafe drivers, and included:

1. Reaction Time Task: Participants were asked to press a button as quickly as possible after a visual cue. The cue’s location was congruent (80% of the cases) or incongruent (20% of cases) with the correct button, assessing attention switching.
2. Judgment Task: Participants were required to press ‘Go’ after a cue to initiate moving a cart, while avoiding obstacles that passed in front of it perpendicularly. An additional ‘Stop’ control option was available to help avoid obstacles.
3. Memory Task, participants are asked to recreate a previously shown geometric shape with their finger, after a short mask screen is presented. The number, complexity of geometric shapes, and the duration of the mask screen presentation changed over time.
4. Motor-Control Task (Bi-Manual Sensorimotor Control): Participants were instructed to hold the tablet-like steering wheel, and follow a moving target ring while avoiding fixed or unpredicted obstacles across four stages of increasing speed.

Table 2 Task and dependent measures from TBCT.

These four tasks have been shown to effectively predict safe and unsafe drivers among older adults [21]. The dependent measures that were used to predict driving performance in previous studies [22,23] are listed in Table 2.

All tasks of the TBCT; Reaction Speed, Decision Making, and Bi-manual perceptual-motor tasks were adapted from the DriveABLE Cognitive Assessment Tool (DCAT). The DCAT is a reliable measure of cognitive processes needed for safe driving and predicts actual on-road performance in cognitively impaired drivers [25]. The DCAT was developed based on a number of standardized neuropsychological tests, including the Visual Field Test, Rabbitt Card Sort, Rod and Frame Test, Sitting- Rising Test, Cognitive Reflection Test, and span of Attentional Field, Speed of Attention Shifting, Corsi Block Tapping Test were collected. The subtasks of the DCAT were tasks designed to be used in conjunction with a touch-screen and a 3-button base and allow for easier administration [26].

2.4 Procedures

Both control groups (Ctrl-Younger and Ctrl-Older) completed all four tasks in a single 20 to 30-minute session, following the fixed order outlined in Table 2. Participants in the drug groups completed the assessment within 3 hours of providing a urine sample, with most finishing within 45 minutes. Due to time constraints, drug group participants completed two tasks in a block of time. Data points exceeding 1.5 times the interquartile range (IQR) were excluded as outliers.

2.5 Data Analysis

There is a significant age difference between all groups (F-statistics: 601.17, p-value < 0.0001, with post-hoc pair-wise comparison, holm adjusted p-value < 0.01 for all pair-wise comparison except for CAN/COC and COC). Therefore, we used univariate analysis of covariance (ANCOVA) to examine group differences in cognitive task performance. Moreover, Shapiro-Wilk and/or Levene tests revealed significant or marginal violations of normality and homogeneity of variance assumptions for all measures. Therefore, we employed a non-parametric version of ANCOVA using Rfit (Rank-based estimation for linear models) in R [27].

For each analysis, we compared a full model (including age and condition) to a reduced model (excluding age) using the Drop in Dispersion Test. If the test result was significant, we used the full model; otherwise, we used the reduced model. For significant results in either model, we conducted post-hoc analyses with Holm adjustment, a family-wise error rate (FWER) multiple comparison adjustment approach, to further investigate group differences.

To further investigate drug effects, we conducted post-hoc analyses using independent 2-group Mann-Whitney U Tests between control older adults and drug groups with significant differences. By removing age control, we examined whether drug use accelerates cognitive decline equivalent to 15-30 years of normal aging. Specifically, we asked: Are frequent drug users’ cognitive abilities comparable to those of healthy adults 15-30 years older?

3. Results

This section compares cognitive performance across five groups: Ctrl-Younger, CAN, CAN/COC, COC, and Ctrl-Older. Descriptive statistics for cognitive task performance across groups were presented in Table 3. The data indicated that the younger control group generally performed better than the other groups, including older adults in various tasks, including Reaction Time, % Success, % Correct Shape, doubling of: % Correct Shape, and % Time Inside Target. These findings suggested that cognitive performance declined with age, which was consistent with previous research. Moreover, this table indicated that the drug groups exhibited varying levels of performance across different tasks. Specifically, the COC group showed worse performance in tasks such as Reaction Time, and % Success than CAN groups. Overall, the results suggested that the different drug conditions had distinct effects on cognitive performance. Notably, the % Premature Go variable exhibited a skewed distribution, with most participants scoring 0, resulting in a median value at the boundary of the distribution, and a tail of higher scores extending to the right. Such a pattern was not observed for other variables.

Table 3 Descriptive Statistics (Mean ± Standard Deviation) for Measures of Cognitive Task Performance of each Control Group.

The results, presented in Table 4, revealed that age significantly influenced all measures, except % Premature Go and Reaction Time in the Judgment Task, and % Surprised Object Avoided in the Control Task. Therefore, for all measures, except these three the full model was used. Reduction in Dispersion test results, p-values, and robust multiple R² values, were displayed in Table 4.

Table 4 Task Performance by Group (age-related variability is accounted for).

Post-hoc analysis results for the significant group effects were reported in Table 5 and Table 6. A significant group effect was observed in all measures except the ones age was not significant. The within drug group comparisons showed that the CAN group performed faster than the CAN/COC group in Reaction Time. No other significant differences were observed between the drug groups.

Table 5 Post-Hoc Analysis of Task Performance (Holm-adjusted p-values).

Table 6 Post-Hoc Analysis of Task Performance (Holm-adjusted p-values).

The Control Older group outperformed the CAN, CAN/COC, and COC groups in % Correct in the memory task. Additionally, the Control Older group performed better than the CAN/COC group in Go Count and % Success of judgment task, and better than the COC group in % Success of judgement task. Moreover, compared to the CAN/COC group, the Control Younger group performed faster in Reaction Time and had better scores in Judgment % Success, Go Count, Reaction Time, and Memory Duration. Additionally, the Control Younger group outperformed the CAN group in % Correct in memory task, and the COC group in % Correct in the memory and % Success in the Judgment task.

The comparison between performance of the control older and drug group, identified to be significant in Table 6, with the effect of age being excluded was presented in Table 7. The Wilcoxon rank sum test results showed significant differences between the Control Older and frequent drug users after excluding the age effect. Specifically, the Control Older group outperformed the CAN group in % Correct of the memory task. Moreover, the Control Older group outperformed the CAN/COC and COC groups in % Success, Go Count of the judgement task, and % Correct of the memory task. Figure 2, Figure 3, Figure 4 showed comparison between different groups across multiple task measures, after controlling for the effect of age.

Table 7 The results of the Wilcoxon rank sum test comparing older adults and drug users with age effect excluded.

Figure 2 Performance in Reaction Time Task. Ctrl-Younger (p-value < 0.001) and CAN (p-value = 0.042).

Figure 3 Performance on the Judgement Task.

Figure 4 Memory Task Performance.

4. Discussion

This study aimed to investigate the impact of healthy aging, as well as cocaine and/or cannabis use on cognitive tasks crucial for driving. We compared performance of older adults, younger adults and frequent drug users who tested positive for cocaine and/or cannabis, utilizing a validated cognitive test battery [21,22,23].

This approach allows for a more nuanced understanding of the adverse cognitive impact, complementing existing research on long-term consequences. The following sections provide a more in-depth examination of the differences between the older control group and drug groups, as well as within-group variations in the drug groups.

4.1 Hypothesis 1: Within Drug Group Users Differences: The CAN Group Were Faster Than the CAN/COC Group

The only significant difference within the drug groups emerged in the Reaction Time Task, where the cannabis group demonstrated faster reaction times compared to the group combining cannabis and cocaine. No significant performance differences were observed between the cannabis and cocaine groups. Age was identified as a significant confounding factor in this task, so the declining effect of age has been removed in this comparison.

This finding suggests that the combination of cannabis and cocaine, may have an additive and detrimental effect on processing speed. Notably, the absence of differences between the cannabis and cocaine groups in this task may indicate an exacerbated negative impact of poly-substance use. However, the interpretation of this null finding should be interpreted with caution due to the smaller sample size of the cocaine group (n = 20) compared to the combined cannabis and cocaine group (n = 62), which may have reduced statistical power and limited detection of group differences. Despite this limitation, the current study contributes to our understanding of the cognitive and behavioral consequences of poly substance use, particularly involving cocaine, where evidence is limited [28].

Processing speed, a fundamental aspect of cognitive function, facilitates various cognitive processes and reflects the efficacy of neural circuitry [29]. Our results indicate that cannabis use is associated with less cognitive impairment in processing speed relative to combined cannabis and cocaine use. Acute cannabis intoxication has been shown to impair psychomotor speed and visuomotor processing, potentially linked to increased glucocorticoid exposure in the prefrontal cortex [14]. However, the effects of chronic cannabis use on processing speed are mixed. Some studies, including those conducted by [30] among older adults and [31] in a 14-year prospective study among adolescents, found no significant impact, whereas others have reported contradictory results [13]. In contrast to the mixed findings on cannabis, acute cocaine administration has been found to enhance psychomotor speed, albeit with a sharp reduction in concentration [17,32]. Nonetheless, long-term cocaine use is characterized by broad cognitive impairment [13]. This disparity highlights the complex relationship between substance use and processing speed.

Differences between cannabis and both cannabis and cocaine may also be attributed to differences in the timing and “coming down” experiences between the two groups. Cocaine’s effects are short-lived (30-40 minutes), followed by an abrupt decline [33]. In contrast, cannabis-induced euphoria gradually fades after 3-4 hours [12]. Together, these findings underscore the complex interactions between substance use and cognitive function, emphasizing the need for continued research to elucidate the specific effects of cannabis and cocaine on processing speed.

4.2 Hypothesis 2: Older Control Adults Outperformed Drug Groups

Aging is characterized by significant brain changes, including enlargement of cerebral ventricles, decreased brain volume, neuronal shrinkage, and myelin loss. These cortical changes contribute to cognitive decline, manifesting as slower reaction times, impaired spatial attention and memory, reduced visual acuity, impacting fitness for driving [34]. This section explores the cognitive differences between frequent drug users and older adults, comparing substance-related cognitive decline with ageing-related decline. This comparison aims to provide insight into the intersection of cognitive decline due to substance use and aging. Our results show that in both Judgment, and Memory Tasks the older adults outperformed the drug groups.

4.2.1 Judgement Task

In the Judgment Task, older adults demonstrated superior performance compared to both CAN/COC and COC groups. Specifically, they showed higher accuracy in avoiding collisions with passing lines and enhanced vigilance and response ability, reflected in a greater number of ‘Go’ button presses. This task requires participants to accurately assess the speed of both moving lines and the box, making timely decisions on when to press the ‘Go’ button to pass the box while avoiding hitting the moving lines. Additionally, participants must quickly press the ‘Stop’ button to prevent collisions. This dual-control aspect—particularly timely ‘Stop’ button presses—is crucial for successful task performance. Successful performance relied on a combination of cognitive and motor skills, including decision-making, reaction time, spatial awareness, and motor control.

The task’s demands mirror real-world scenarios, such as driving, where rapid decision-making is critical, like stopping or navigating intersections. By evaluating these skills, the Judgment Task provides valuable insights into an individual’s ability to navigate complex, dynamic situations.

The prefrontal cortex (PFC), specifically the dorsolateral PFC and anterior cingulate cortex, plays a critical role in decision-making and executive function [35]. Animal studies have shown that long-term cocaine exposure in rats leads to reduced neural activation in these areas [36,37]. Consistent with animal findings, studies in humans have revealed abnormal brain activation, functional connectivity, and anatomical structure in chronic cocaine users and individuals prenatally exposed to cocaine [35]. Moreover, research by [38] indicates that THC consumption leads to dose-dependent impairments in fine motor control and motor timing. Additionally, long-term cannabis use has been linked to cognitive declines in various domains, including decision-making and processing speed [39,40,41]. These cognitive impairments may contribute to the observed results in the CAN/COC group, suggesting a potential underlying mechanism.

4.2.2 Memory Task

In the Memory task, older adults exhibited superior performance, surpassing all drug groups -including CAN, CAN/COC, and COC- in accurately redrawing shapes. Notably, no significant differences were found in completion time, ruling out speed-accuracy tradeoff (SAT) as a factor. The Memory task involves redrawing a geometric shape after a brief mask screen, assessing working memory capabilities. Working memory temporarily holds information, facilitating reasoning, decision-making, and guiding behavior. As a crucial cognitive skill, working memory plays a vital role in driving, where information must be processed and retained to navigate safely.

Research on acute cocaine’s impact on working memory has yielded mixed findings. [42] review found no significant effects, while [43] review suggested small to moderate deficits. However, [44] review resolves this discrepancy by proposing that chronic cocaine exposure triggers compensatory neuroplastic changes. These changes enable normalized cognitive performance during acute intoxication but lead to declined performance during withdrawal, which is in line with findings that recreational cocaine users exhibited significant impairments similar to individuals with cocaine use disorder [45]. This interplay between cocaine exposure, neuroplasticity, and cognitive function emphasizes the importance of considering population characteristics (cocaine dependence vs. non-drug-using controls) and timing post-intoxication. Despite differing conclusions regarding acute cocaine administration, both [42] and [43] noted moderate deficits in working memory in long-term cocaine users compared to control groups. These reported deficits were also consistent with previous meta-analyses exploring cognitive functioning in cocaine users [46].

Research on cannabis use’s effects on working and short-term memory yields mixed results. Acute exposure has varying impacts on working memory, lasting several hours post-consumption, with prolonged effects in chronic user’s memory [13,41]. Acute use increases disinhibition and impairs specific memory components [41], particularly working memory [13]. Chronic cannabis use also shows mixed effects on working memory, with impairments reported in some studies, but not others [13,41]. Notably, neuroimaging studies reveal brain circuitry abnormalities despite behavioral similarities [47]. Heavy cannabis use in older adults has been linked to altered brain structure, including thinner cortices in the hippocampus, an area critical for memory [48]. This cortical loss may exacerbate age-related cognitive decline. Overall, impaired memory-related cognitive performance shows small-to-medium effect sizes across domains [39,40,41].

Summarizing results in association to the second hypothesis it seems as older adults’ superior performance in Decision Making and Memory Tasks remained consistent, even without accounting for age-related decline. In this study, we compared older adults, with a mean age of 66.06 ± 6.85 years, to frequent drug users across three groups: CAN (37.41 ± 11.69 years), CAN/COC (47.99 ± 12.94 years), and COC (52.45 ± 7.86 years). Notably, frequent drug users’ decision making abilities deteriorated to levels worse than those of older adults 18+ years their senior (CAN/COC) and 13.6+ years (COC). Furthermore, working memory abilities in frequent drug users were significantly impaired, performing worse than older adults 29 years (CAN), 18 years (CAN/COC), and 13.61 years (COC) their senior. This effect was particularly pronounced in the CAN group, highlighting concerns about cannabis’s impact on working memory, despite its perceived lower risk.

4.3 Hypothesis 3: Younger Control Group Outperformed Drug Groups but Not Older Adults

Cocaine and cannabis impair distinct cognitive abilities crucial for driving. Aging-related decline similarly affects essential cognitive domains, including processing speed, memory, and decision-making. Consequently, we expected younger controls to outperform both older controls and frequent drug users.

Improved performance of younger controls compared to the CAN/COC group was observed in all measures of all tasks, except the Control Task. Moreover, younger adults outperformed both CAN and COC groups in Memory Task accuracy and surpassed the COC group in Judgement Task success rates. These findings align with previous research documenting reduced cognitive performance among frequent drug users, as discussed in preceding sections. Notably, the pronounced performance gap between younger controls and CAN/COC groups, relative to other frequent drug users, suggests more severe impairment in the poly substance user group. This underscores the importance of considering poly substance use on cognitive function.

Interestingly, the Control Task lacked sensitivity to differentiate between groups, despite its effectiveness in identifying risky drivers when combined with other tasks. This highlights the importance of using a comprehensive assessment battery to accurately detect cognitive impairments.

After adjusting for age using a rank-based approach, we found no significant differences in cognitive performance between older and younger adults. The rank-based approach robustly accounted for age-dependent changes, ensuring a fair comparison. Notably, older controls in this study were active drivers during the last six months and were deemed fit to drive. This explains why, after controlling for age, no performance differences emerged between older and younger adults in cognitive tasks related to driving fitness. However, significant performance differences were observed between older and younger adults before adjusting for age. This suggests that aging is the primary contributor to these differences. Consequently, accounting for age eliminated group differences.

5. Conclusions and Limitations

Driving is a vital aspect of quality of life throughout the life span of an individual, requiring coordinated cognitive functions. Although both aging and drugs impact cognitive abilities, it is not clear whether their patterns of cognitive decline are similar or not. This study investigated cannabis and/or cocaine users’ performance in standardized tablet-based cognitive tasks assessing driving fitness. The results revealed that frequent drug users performed worse overall than both older and control groups, particularly in memory and decision-making domains. More specifically, when comparing the performance of frequent drug users to older control adults, the results showed that frequent drug users performed worse in memory and judgment domains, despite the older adults being much older.

Further, we observed significant age differences among the frequent drug user samples. Notably, the CAN group was approximately 10 and 15 years younger than the CAN/COC and COC groups, respectively. This observational study, although limited to frequent drug users volunteering for drug recognition evaluator (DRE) training, indicates that cannabis legalization may disproportionately impact younger adults. Further investigation is necessary to explore this relationship.

It is important to emphasize that our study lacked information on duration, dosage, and timing of use post-intoxication. Additionally, while our urine tests detected the presence of drugs, we did not have information on the levels detected, as the data was collected in a Law Enforcement Facility in the US. The pattern of drug usage, history, frequency and intensity was also missing. Furthermore, we did not collect breath samples to assess alcohol use, which may interact with other substances and affect performance. The thresholds for detecting drug presence in urine samples were determined by the company’s specifications (e.g., 50 ng/ml for cannabinoid and 300 ng/ml for cocaine). Moreover, older and younger adults in both control groups were not tested for drug use, which may affect the interpretation of our results if some control participants were using substances. These limitations underscore the inherent challenges in conducting experimental research on drugs such as cocaine due to experimental control and ethical considerations. Therefore, most existing literature focuses on the long-term effects of repeated cocaine use, typically assessed after the drug has been cleared from the body, as indicated by negative urine samples [42]. Although we can only assume the frequent drug users in our study were unfit to drive, this is an interesting result showing the extent of cognitive decline in frequent drug users.

Eventually, older adults will most likely have to retire from driving at some point in their lives due to “life expectancy exceeding driving fitness expectancy” in the United States. Research shows that at around 70 years of age and older fatal crash rates increase significantly and are highest for individuals 85 years of age and older. However, literature also suggests that age alone is not a very reliable variable that predicts on-road driving and as seen in this study, older adults performed well in most tasks. As older drivers increasingly rely on driving and their own personal vehicle transportation, it is becoming more important to use and develop time and cost effective screening tools that are valid and reliable, easily accessible and preferably standardized to properly screen and assess driving skills and crash work of older adults. Based on existing research, it is crucial to perform statistical analysis controlling for age when it comes to cognitive impairment and cognitive assessments in relation to driving to identify correct differences in impairments patterns associated with age related decline, or drug-related cognitive decline.

These different impairment patterns need to be further investigated perhaps with better drug screening technologies to more accurately differentiate blood/urine metabolites and associated cognitive performance differences, recording the history of drug use, frequency and dose. A study of this nature is complex and has many challenges. Controlling for confounding factors with drug using participants is difficult particularly where self-report is part of the assessment. For example, the cocaine (COC) group showed urine metabolites associated with cocaine consumption, but it is possible they were coming off a short-lived high at the time of assessment. Also, we were not able to assess alcohol use in our participants.

A further limitation displayed in this study is that the participants were pre-determined based on availability and willingness to participate. We could not randomly sample, which may have affected the results. This kind of study is wrought with sampling problems and follow-up research will be aimed to avoid these sampling issues.

Further, our study also lacks information on comorbid physical conditions such as vision and hearing loss that comes with aging, however, the participants were actively driving, therefore we assumed that annual hearing and vision screenings were completed and passed the elderly assessment process in regards to on-road driving, however, we did not particularly screen for hearing or vision loss. Furthermore, we did also not screen for medication use that could interfere with cognitive skills in regards to on-road driving which portrays an additional limitation.

Another limitation with our study is that younger adults are generally more familiar with technology, including experience with gaming technology on touchscreen devices. Therefore, the younger control group as well as younger participants in the drug groups may have had an advantage over other groups in this study that could exacerbate and mitigate the observable effects of the drugs in different scenarios.

When interpreting lower cognitive scores in the frequent drug user group, caution is necessary to avoid over-pathologizing results [49]. As [44,50] noted, impaired performance may still fall within demographic norms, particularly given our study’s unmatched age groups and lack of information on pre-existent neurodevelopmental factors and behavioral traits in frequent drug users. Notably, comorbid conditions like major depressive disorder are linked to cognitive deficits in attention, memory, learning, executive function, and processing speed [51]. Such conditions that were not controlled in this study, may impact performance beyond drug use. Therefore, when interpreting our findings, it’s essential to consider the complex interplay between substance use, demographic factors, and potential comorbid conditions to avoid misattributing cognitive deficits solely to drug use.

Despite these limitations, our study offers a distinct contribution to the field. Its unique design features two control groups spanning the lifespan, allowing for comprehensive comparisons. Additionally, the study investigates the cognitive effects of cannabis, cocaine, and combined cannabis and cocaine while participants are still intoxicated, as confirmed by positive urine tests. By utilizing cognitive tasks relevant to driving, this study provides valuable insights into substance use’s cognitive impacts in an important daily life activity while navigating ethical concerns surrounding the experimental administration of drugs.

Acknowledgments

All participants completed consent forms approved by the University of Alberta Research Ethics Board and received a $10 USD honorarium. No feedback on their performance was provided, but participants were allowed to view the results of their urine test. The brand name for the Tablet Based Cognitive Tasks (TBCT) is Vitals Mobile and was developed by Impirica.

Author Contributions

R.B. and M.T. are co-first authors of this manuscript. R.B., M.T., A.G., & A.S. conceived of, and designed the study. A.G. supervised the drug users’ data collection, and R.B., M.T., and S.L. collected the healthy older and younger adult data. R.B., M.T., A.G., and F.C. conducted the data analyses. R.B., M.T., and A.S. wrote the first draft of the manuscript, and all authors edited and revised the manuscript.

Funding

This work was supported by a Natural Sciences and Engineering Research Council of Canada (NSERC) Grant awarded to A.S.

Competing Interests

At the time of this study A.G. and F.C. were employees of Impirica, and A.G. had ownership interest in Impirica.

AI-Assisted Technologies Statement

During the preparation of this manuscript, the authors utilized Meta AI’s Llama 3.2 to enhance readability and language clarity. Although this tool was employed, the authors thoroughly reviewed and edited the content to ensure accuracy and integrity. The authors assume full responsibility for the published article’s content.