The present study tested a theoretically-grounded model examining individual difference and contextual predictors of texting while driving in a sample of young adult drivers. The main findings supported the hypothesis that individuals lower in mindfulness tended to report more frequent texting while driving. In addition, findings suggested that this relationship was explained by individual differences in emotion-regulation motives (i.e., texting to ameliorate negative emotional states) and attention-regulation motives (i.e., limiting texting to allow greater attention to day-to-day experience). Path analysis more strongly supported emotion-regulation motives as a mediator of how frequently young adults text message while they drive. These resulting linking individual differences in dispositional qualities with texting behavior is important because texting-while-driving is associated with an increased the risk of motor vehicle accidents. Nonetheless, many young adults engage in this behavior. Nearly one-third of the present sample of young adult drivers reported sending or reading texts while driving with some regularity. Understanding predictors and correlates of texting-while-driving may help to indentify relevant targets for interventions to reduce this dangerous behavior.
The current results are consistent with past research that showed how individuals low in mindfulness are more likely to have difficulty regulating emotions and may be more likely to employ maladaptive strategies to manage emotions (Arch & Craske, 2010
; Baer et al., 2006
; Creswell et al., 2007
; Feldman et al., 2007
). The present study extended prior work by being the first to demonstrate a link between mindf ulness and the use of texting as an emotion regulation strategy. It is also the first to find that emotion-regulation motives for texting are linked to texting-while-driving and that emotion regulation motives may influence consumer behaviors that would facilitate more frequent texting.
Attention-regulation motives were associated with mindfulness and with texting-while-driving in a univariate analysis but not the multivariate path analysis. This effect may have been diminished by the nature of the measure created for this study to capture this construct. In particular, the score for this scale exhibited relatively low internal consistency, which may be attributable in part to the scales’ brevity as well as efforts to cover a variety of facets of these constructs and avoid redundant language in the items (see John & Benet-Martinez, 2000
). On the one hand, measures with low alpha values may be acceptable in the early stages of research into a new psychological phenomenon (Nunnally & Bernstein, 1994
), as is the case with this study. Nonetheless, it is important to note that associations between attention-regulation motives and other variables may be attenuated in the present study.
Results suggest that the six predictor variables in the present model account for a little over one-fifth of the variance in texting-while-driving (a medium effect size); however, the effects of the individual predictors on each mediator tended to exhibit effect sizes in the small range. This suggests that there are likely a host of other relevant predictor variables that future research should consider. For instance, variables related to the theory of planned behavior (e.g., attitudes about texting-while-driving, perceived social norms) prospectively predicted nearly 30% of the variance in texting-while-driving (Nemme & White, 2010
). Thus, future research integrating both mindfulness and the theory of planned behavior holds prom ise for understanding psychological factors contributing to texting-while-driving.
The present study has several limitations that should be noted. First, the use of cross-sectional data limits the ability to make inferences about temporal relationships between variables. Second, the use of self-report data to assess texting-while-driving may have compromised accuracy due to recall errors and social desirability. This latter issue may have been attenuated somewhat in the present study given that texting-while-driving had not been outlawed at the time of data collection. Future studies would benefit from more objective measures of texting-while-driving, for instance the use of mobile phone records or automated event recorders. Third, the sample used in the present study (female college students) offers both strengths and limitations. On the one hand, this sample could be conceptualized as being at-risk for texting-while-driving given that younger drivers are more likely than more experienced drivers to text and drive (e.g., Marist, 2010
). The use of an all-female sample may limit generalizability in light of potential gender differences in the experience and regulation of emotions; however, gender differences in texting-while-driving have not been observed in other samples of young drivers (e.g., Lenhart et al., 2010
; Marist Poll, 2010
; Nemme & White, 2010
). In sum, results may not generalize to drivers who are male, older, less educated, or more frequent drivers than this sample. Nonetheless, results may help illuminate relevant factors for drivers in an at-risk age group.
Although the results of the present study should be regarded as preliminary in light of these limitations, they offer potentially helpful clues to inform targets of intervention to decrease the public health risks posed by texting-while-driving. First, it is informative to note that all three contextual variables examined in the present model (frequency of driving, owning a mobile phone with a full keypad, and a mobile phone plan with unlimited texting) were independent, significant predictors of texting-while-driving. Future experimental research could examine whether manipulating one (or all) of these variables would result in a reduction of texting-while-driving. Indeed, such technology-based environmental modifications have been proposed to curb texting-while-driving (Johnson, 2009
). However, in the present study, emotion regulation motives remained a significant independent predictor of texting-while-driving above and beyond these three contextual variables, raising the prospect that modifying these contextual variables would still leave a relevant risk factor unaddressed.
Mindfulness-based interventions may offer some promise in terms of addressing emotion-regulation as a potential risk factor. For instance, mindfulness-based interventions have been shown to successfully reduce other risky and self-destructive behaviors that may serve an emotion-regulating function, including substance abuse (Bowen & Marlatt, 2009
), binge-eating (Kristeller, Baer, & Quillian-Wolever, 2006
), and non-suicidal self-injury (Neacsiu, Rizvi, & Linehan, 2010
). Furthermore, laboratory studies of individuals without prior meditation experience suggest that even brief practice of mindfulness exercises may help to promote reduced emotional reactivity (Arch & Craske, 2006
), more rapid recovery from a negative mood (Broderick, 2005
), and less distress in response to repetitive thoughts (Feldman, Greeson, & Senville, 2010
). As such, it is possible that brief mindfulness exercises practiced before or while driving may help to alleviate potential affective and cognitive factors that may serve as triggers for texting-while-driving. Further research is needed to examine the feasibility, desirability, and—ultimately—efficacy of such interventions.
In conclusion, the present study provides initial evidence that individual differences in mindfulness are associated with a reduced likelihood of texting-while-driving. The study also suggests that individuals low in mindfulness may engage in texting-while-driving as a means to regulate negative emotions. These novel findings suggest that the constructs of mindfulness and emotion regulation deserve further attention from researchers, policy makers, and clinicians interested in understanding and preventing texting-while-driving.