Edited by: Monica Capra, Claremont Graduate University, United States
Reviewed by: David Herrero, Universidad Europea del Atlántico, Spain; Carmelo P. Cubillas, Autonomous University of Madrid, Spain
This article was submitted to Cognition, a section of the journal Frontiers in Psychology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Risky decision-making is highly influenced by emotions and can lead to fatal consequences. Attempts to reduce risk-taking include the use of mindfulness-based interventions (MBI), which have shown promising results for both emotion regulation (ER) and risk-taking. However, it is still unclear whether improved emotion regulation is the mechanism responsible for reduced risk-taking. In the present study, we explore the effect of a 5-week MBI on risky driving in a group of repeat traffic offenders by comparing them with non-repeat offenders and repeat offenders without training. We evaluated the driving behavior of the participants through a driving simulation, and self-reported emotion regulation, both before and after the intervention. At baseline, poor emotion regulation was related to a more unstable driving behavior, and speeding. The group that received mindfulness training showed improved performance during risky driving situations and had fewer accidents, although their overall driving behavior remained largely unchanged. The observed trend toward improved emotion regulation was not significant. We discuss whether other effects of MBI – such as self-regulation of attention – could underlie the observed reduction in risky driving in the initial stages. Nonetheless, our findings still confirm the close relationship between emotion regulation skills and risky driving.
Daily life involves constant decision-making with regard to what actions to take and some situations can lead us to take certain risks, e.g., when we are in a rush or in a bad or even euphoric mood. In fact, the factors that modulate the process of risky decision making include emotion (
In driving environments in particular, risky driving and low perception of risk have been found to be influenced by emotions (
Due to the influence of ER on risky driving, one promising strategy for reducing road fatalities could be the use of interventions that, through the improvement of ER skills, lead to safer driving behavior (
Mindfulness, understood as a trait, is the natural mindful tendency of each individual (
Thus, in the light of these findings, we aimed to test the effectiveness of a MBI in reducing risky driving behavior in a group of repeat traffic offenders, measuring behavioral change through a driving simulation. This type of measurement has been used to study real driving behavior, with a correspondence between real and simulated driving (
Our sample was composed of 89 participants (29 women; age range between 18 and 63 years,
To group drivers into repeat and non-repeat offenders, we used the following self-reported traffic violations as criteria: attendance of a rehabilitation course for drivers at least once, a loss of points according to the Spanish penalty system for traffic rule violations, being fined at least twice for risky driving behavior (alcohol or drug use, not using a seat belt, or exceeding speed limits), or reporting as having usually exceeded speed limits by more than 20% of the permitted speed. Sixty repeat offenders, meeting at least one of these criteria, and 29 non-repeat offenders, meeting none of these criteria, completed the baseline and post-intervention evaluations.
Half of the risky drivers were selected for a 5-week MBI program dependent on their weekly availability, which was established prior to testing, to gather the greatest number of participants for the intervention groups. At four different time points along the 2-year period of data collection, we grouped the participants who coincided at the same weekday availability, resulting in a quasi-randomized controlled trial. The drop-out rate of the mindfulness training following the second session was 6 out of the 30 participants.
Thus, in the current study, we compared the following three groups: non-repeat offenders (NR,
Demographic variables and driving experience (mean and standard deviation).
NR | R | R-M | |
---|---|---|---|
Age | 32.38 (14.6) | 35.03 (14.66) | 35.7 (14.75) |
Sex (women/men) | 11/18 | 6/24 | 12/18 |
Education level | 3.55 |
3.67 |
3.4 |
Estimated km driven in life | 228867.24 (424848.94) | 324144.83 (401172.55) | 373083.33 (606785.13) |
Interval of estimated km driven per year by car | 5.1 |
6.59 |
5.83 |
Education ranged from 2 (Primary studies) to 4 (Superior level studies).
Km driven/years is measured in estimated intervals, with 5 = 6,000-9,000; 6 = 9,000-12,000; and 7 = 12,000–15,000 km.
We used two complementary questionnaires, focusing on the awareness of emotion and its regulation and different types of ER strategies, respectively.
To group participants into repeat and non-repeat offenders they reported on demographic variables (sex and age), km driven per year and in life, months of holding a driver license, number of rehabilitation courses, number of lost points, number of traffic fines, and frequency of exceeding speed limits.
The Spanish version of the Difficulties Emotion Regulation Scale (DERS;
The Spanish version of the Cognitive Emotion Regulation Questionnaire (CERQ;
For the driving simulation, we used the HRT motorcycle simulator, which consists of a seat, handlebar, pedals, accelerator, brakes, turn indicators, and claxon (see
Indices calculated from data recorded by the HRT included: average and variance of speed (km/h), of speed in a risk situation (km/h), and of exceeded speed limits (km/h), length of time spent exceeding speed limits (sec), average throttle rotation (%), variance of steering wheel (rad), front and rear brake force (kg), number of accidents, and the average rating of performance in each risk situation, calculated by the HRT, ranging from A (good performance) to D (accident), taking into account variables, such as speed when entering the risk situation and distance to crash with an object. This last value is coded from 1 to 4 with lower values indicating greater risk-taking and worse performance in a risk situation. Exceeded speed limits, speed in risk situations, and performance ratings are calculated for the two urban courses only, since the HRT software does not register measures of speed limits or risk situations in the mountain road scenario.
The mindfulness-based training was adapted from the Mindfulness-Based Stress Reduction (MBSR) program widely used in research (
The participants were selected according to their self-reported traffic violations, which were requested by means of an online survey (see participants section). The baseline and post-intervention evaluation was the same. Participants came to the research center and, as a part of a broader project, filled in the questionnaires and completed the driving simulation, with the order based on the availability of the facilities (HRT and computer room) and participants’ temporal availability. The average interval between both evaluations was approximately 4 months (Mean = 142.26 days,
As in the literature recommended, we used an intention-to-treat approach (
JASP statistical software (Version 0.11.1, JASP Team, 2020, freely available at
We found no differences between the three groups in terms of gender (Pearson’s
Mixed-factor ANCOVAS were conducted to analyze the effect of the intervention on the driving behavior indices and the ER strategies. The experimental design includes time (baseline and post-intervention evaluation) and subscales as within-subject factors, and group (NR, R, and R-M) as between-subject factors, using age and interval between evaluations as covariates. To support our hypothesis with Bayesian methods, the Bayes Factor (BF10) was calculated for all possible models compared to the null model, as well as the BFInclusion for each predictor (
For variables with significant time × group interactions, mediation analysis was conducted. To explore the magnitude of the intervention effect, an ANCOVA was carried out on the Behavior Shift Index (BSI;
As accident rate is not a continuous variable, these rates are analyzed by categorizing the difference between baseline and post-intervention evaluation into improved (difference > 0), unchanged (difference = 0), and worse performance (difference < 0). A multinomial regression was conducted, and risk ratios were estimated with group (NR, R, and R-M) as the between-subjects factor and age and interval between evaluations as covariates.
Finally, we explored whether ER and driving behavior are related, conducting a partial correlation analysis between the BSI values of all questionnaire subscales and the driving behavior indices, using age as a covariate. We also confirmed a general relationship between ER and the driving behavior indices at baseline in the whole sample (
Analysis of the driving performance ratings revealed a time × group interaction [
Effect of the intervention on the behavioral indices of the driving simulation.
No significant time × group interaction effects (
Relationship between evaluation of performance in risk situations, as calculated by the Honda Riding Trainer (HRT), and other HRT indices.
Average speed (km/h) | Variance speed (km/h) | Average speed risk situations (km/h) | Variance speed risk situations (km/h) | Average exceeded speed limits (km/h) | Variance exceeded speed limits (km/h) | Length of time spent exceeding speed (s) | Mean throttle (%) | Front brake (kg) | Variance steering wheel (rad) | Accidents (sum) | |
---|---|---|---|---|---|---|---|---|---|---|---|
Evaluation of performance in risk situation (1–4) | −0.752 |
−0.549 |
−0.847 |
−0.611 |
−0.642 |
−0.528 |
−0.762 |
−0.563 |
−0.338 |
−0.691 |
−0.473 |
Correlations of BSI values | −0.464 |
−0.26 |
−0.624 |
−0.313 |
−0.184 | −0.219 |
−0.253 |
−0.273 |
−0.082 | −0.434 |
— |
Correlations of BSI in the R-M group | −0.685 |
−0.453 |
−0.835 |
−0.542 |
−0.231 | −0.317 | −0.438 |
−0.407 |
−0.322 | −0.528 |
— |
Rear Brake Index did not show any significant relationship. Behavior Shift Index (BSI) for accident rate could not be calculated as variable is not continues. Partial correlation coefficient reported represent Pearson’s
The multinomial regression analysis of the differences in accident numbers between baseline and post-intervention evaluations revealed differences between the R-M and both control groups in the comparison between worse and better performance (
No time × group interaction effect was found for ER [DERS:
Differences in the emotion regulation (ER) scales between the baseline and post-intervention for each group. Total scores are presented for the Difficulties Emotion Regulation Scale (DERS;
In summary, participants who were trained in mindfulness do not show differences in ER but showed improved performance in risk situations and had fewer accidents in comparison with both control groups. It is also worth noting that while age is an important factor in the prediction of driving behavior, this factor has almost no influence on the magnitude of improvement observed as a consequence of the intervention.
In the current study, we explored the effect of MBI on driving behavior and ER. We evaluated the performance on a driving simulation and self-reported ER scores of a group of repeat offenders trained in mindfulness and compared these measures with those of two control groups, one of repeat offenders and another of non-repeat offenders. We found that the intervention had an effect on accidents and evaluation of performance in risk situations, but no effect on ER and most of the behavioral indices. However, driving indices were closely related to the performance ratings at baseline, while the magnitude of change was related to the one of performance ratings, being greatest in the mindfulness trained group.
The R-M group of our study had fewer accidents and performed better in a risk situation, although no differences were found in terms of the other driving indices, such as speed, acceleration, and driving direction. However, these indices are closely related to performance ratings in risk situations, and the magnitude of change observed in these measures is strongly associated with the change in performance ratings in the R-M group, pointing to the possibility that most of the indices are enhanced in a similar way. As mentioned earlier, the length of the intervention could have played a role in the non-significance of some of these effects, which might be greater in a follow-up study.
Previous research on the effect of mindfulness training on driving behavior is still scarce. In fact, there is only one study exploring changes in driving simulation, which found a (non-significant) reduction in traffic violations in students enrolled in a Buddhism course (
Furthermore, our findings showed that age is one of the most important predictors of risky driving behavior on the simulator, characterized by speeding, an instable direction, and low evaluations in the performance in risk situations, which is in line with previous research (
In the present study, we found no differences in ER following the 5-week MBI, although the pattern of results points to less ER difficulties and a reduction of negative ER strategies, such as ruminations, catastrophizing, and blaming oneself or others. Studies with a longer intervention – usually 8 weeks – have found enhanced ER (for a review, see
In the driving context, research has focused on driving anger and aggressive driving, applying a wide variety of approaches, including behavioral, cognitive, and relaxation techniques (for a review, see
Moreover, training in mindfulness may not only reduce driving anger, but might also produce other changes in ER strategies that affect aberrant driving behavior. Thus, it is important to measure changes in emotion regulation or expression in general. This issue was addressed in a study with Chinese bus drivers, where cognitive therapy, using the same type of ER instruction as that used in the present intervention, resulted in the greater use of positive ER strategies (
Taken together, our results provide first evidence of a behavioral change following MBI in repeat offenders, a high-risk group for road accidents and fatalities. Since in the current study, behavior is measured in a simulated traffic environment, and not only with questionnaires or decision-making tasks, the results are promising and suggestive of real-life behavior. Additionally, it should be noted that, even though a motorcycle simulator was used, these results may indicate safer driving behavior in general, as well in other vehicles such as cars and bikes.
Although research has pointed to ER as the mechanism underlying safer driving behavior (
We hypothesize that the first behavioral changes may be faster and easier to measure than differences in ER, which may be more stable over time. The improvements in attentional control, which are enhanced by MBI (
Although our findings indicate that MBI lead to a safer performance in risk situation, more research is needed to confirm our results and to study long-term effect. Since our sampling was based on the temporal availability of the participants, complete randomized trials are needed with a greater number of participants, as well as studies using longer MBI programs, to explore whether longer training improves ER and other indices of driving behavior.
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation, to any qualified researcher.
The studies involving human participants were reviewed and approved by the Human Research Ethics Committee of the University of Granada (n° 204/CEIH/2016). The patients/participants provided their written informed consent to participate in this study.
ACt, ACn, AM, and SB designed the experiment. LM-C and SB carried out the testing of participants. CV-L and EC-V designed and performed the intervention. SB and ACt analyzed the data. SB and LM-C drafted the manuscript. All authors contributed to the critical revision of the manuscript and approved the final version.
CV-L and EC-V were employed by the company Presentia.
The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
We would like to thank the support of the Andalusian Regional Government, and the European Regional Development Fund (ERDF), to the Brain, Behavior, and Health, scientific excellence unit (SC2), ref: SOMM17/6103/UGR.
The Supplementary Material for this article can be found online at: