پیش بینی شخصیت از سرعت در رانندگان جوان: خشم در مقابل هیجان خواهی

Among personality factors, sensation seeking and anger are the main predictors of voluntary risky behaviors. The studies that compare the impact of these factors show that anger is a greater predictor of voluntary risky driving behaviors than sensation seeking. However, these studies usually average data from several risky behaviors, and it is possible that analyzing data from individual risky behaviors would yield different results. Speeding in particular corresponds more closely to the definition of sensation seeking than anger, and should be influenced more strongly by sensation seeking than anger. To test this hypothesis we conducted two studies among French young drivers (n1 = 143; n2 = 2038), where we asked participants the speed they would drive at in a given situation or the likelihood they would commit speeding, and used scales specific to driving situations to measure anger and sensation seeking with. Both studies reveal that driving sensation seeking is a better predictor of speeding than driving anger or either of its factors. The implications of these results are discussed.

Crash risk incurred while driving is an important concern in modern societies, and represents a high cost in terms of public health and material damages. As a consequence, a lot of research investigate personal determinants of risky driving behavior (Rothengatter, 1997), such as attitudes, subjective norms, perceived behavioral control, past behavior (as in studies based on the theory of reasoned action (Ajzen & Fishbein, 1980), or the theory of planned behavior (Ajzen, 1991)), age, gender, driving experience, or personality traits, (Aberg, 1993, Blasco, 1994, Elliott et al., 2005 and Letirand and Delhomme, 2005). Concerning personality traits, current studies rely on constructs that are well documented in the literature, both general and driving-specific (Iversen and Rundmo, 2002 and West and Hall, 1997). These studies either focus on one personality trait, or compare the relative impact of several personality traits on risky driving behaviors. The conclusions of such studies can be used to identify populations more likely to adopt risky road behaviors. Indeed, in given situations, drivers can adopt driving behaviors that are not adapted to the situation, and that as such may increase the crash risk in the situation. This tendency toward risky ill-adapted behaviors varies depending on dimensions of their personality, such as sensation seeking orientation (Dahlen et al., 2005 and Dahlen and White, 2006), driving anger (Deffenbacher, 2008, Deffenbacher et al., 1994, Delhomme and Villieux, 2005 and Villieux and Delhomme, 2007), locus of control (Montag and Comrey, 1987 and Warner et al., 2010), or impulsiveness (Burns and Wilde, 1995 and Elander et al., 1993). In particular, sensation seeking and anger have received a lot of attention, with several studies showing that high levels of either sensation seeking or anger increase the likelihood of several risky driving behaviors such as speeding, as well as the likelihood of crashes. The studies that compared the effects of sensation seeking and anger on risky driving behaviors (Dahlen et al., 2005 and Dahlen and White, 2006) reveal that the link between anger and risky driving behaviors is stronger than the link between sensation seeking and risky driving behaviors. 1.1. Sensation seeking Sensation seeking is defined as the personality trait that refers to individual differences in optimal levels of arousal and stimulation (Zuckerman, 1994 and Zuckerman, 2007), or more precisely “the need for varied, novel, and complex sensations and experiences and the willingness to take physical and social risks for the sake of such experiences” (Zuckerman, Kolin, Price, & Zoob, 1964, p. 10). It is most often measured using the 40-item Sensation-Seeking Scale (SSS; Zuckerman, 1994 and Zuckerman et al., 1964), which comprises four subscales: Thrill and Adventure Seeking, Experience Seeking, Disinhibition, and Boredom Susceptibility. Another commonly used scale is the 20-items Arnett Inventory of Sensation Seeking (AISS; Arnett, 1994). Age and gender are among the strongest determinants of the motivation to seek sensations. More precisely, findings show a decline in sensation seeking with age (Giambra, Camp, & Grodsky, 1992): sensation seeking rises between ages 9 and 14, peaks around age 20 and declines steadily thereafter (Zuckerman, 1994). Concerning gender, men score higher than women (Zuckerman & Kuhlman, 2000). A particular attention has also been paid to the impact of sensation seeking on risky driving behaviors. Indeed, higher sensation seekers are more likely than lower sensation seekers to adopt risky driving behaviors such as drunk driving, speeding, racing other drivers, passing in no-passing zones, and several other behaviors (Arnett, 1990, Arnett et al., 1997, Burns and Wilde, 1995, Clément and Jonah, 1984 and Furnham and Saipe, 1993). Jonah (1997) reviewed 40 studies investigating risky driving behaviors and showed that sensation seeking accounted for 10–15% of the variance of risky road behaviors, and was also positively correlated to road crash involvement. The tight relationship between sensation seeking, risky driving behavior and age led to a number of studies focusing on young drivers. Indeed, the high road crash and fatality rate for young drivers (they are twice as likely as older people to die in a fatal crash (European Road Safety Observatory, 2008 and Jonah, 1986)) can be partially explained by their lack of driving experience (Borowsky et al., 2009 and Borowsky et al., 2010), but also by their higher likelihood to seek sensations while driving (Arnett, 1990 and Cestac et al., 2011). Most of the aforementioned studies on risky driving used Zuckerman’s SSS (1994). However, SSS is a non domain specific scale. Given the greater validity of domain specific scales to assess personality traits when focusing on particular domains, Taubman et al., 1996 and Yagil, 2001) developed a scale specific to the driving context, the Driving Related Sensation Seeking (DRSS). DRSS measures the sensation-seeking trait as defined by Zuckerman with a special focus on the “thrill and adventure seeking” subscale (Zuckerman, 1994 and Zuckerman et al., 1964), expressed in the specific area of driving (adapted in French by Delhomme, 2002). As such, DRSS comprises items directly related to speeding and to other thrill-related driving behaviors. Like SSS, DRSS has been positively related to risky driving in general and speeding in particular, but is distinct from a mere intention to adopt risky behaviors (Delhomme et al., 2009 and Yagil, 2001). 1.2. Anger Anger as a personality trait is commonly defined as the tendency to experience anger and related states such as frustration and bitterness (Berkowitz, 1993, Clore et al., 1993, Russell and Fehr, 1994 and Shaver et al., 1987), and is most commonly measured through anger scales such as the State-Trait Anger Expression Inventory-2 (Spielberger, 1999). Moreover research on anger has distinguished trait anger, the general tendency to be angry, no matter the situation (Deffenbacher et al., 1996), and situational anger, the proneness to be angry in a given situation. Thus, several domain-specific anger conceptions have been developed, among which driving anger, defined as a situation-specific form of trait anger, related to driving (Deffenbacher et al., 1994 and Villieux and Delhomme, 2010). This concept is mainly measured though the 33-items Driving Anger Scale (DAS, Deffenbacher et al., 1994), consisting of six factors: “Illegal Driving”, “Traffic Obstructions”, “Hostile gestures”, “Progress impeded”, “Police Presence” and “Discourtesy”. People high on driving anger – people who get angry more frequently and more strongly when driving – act more aggressive (Deffenbacher, 2008 and Deffenbacher et al., 2003) and adopt more aggressive and risky behaviors (Deffenbacher et al., 2000 and Deffenbacher et al., 2001). In particular, they adopt more often behaviors listed in the Driving Anger Expression Inventory (DAX), and the Driving Survey (Deffenbacher et al., 2003, Deffenbacher et al., 2002 and Deffenbacher et al., 2003), among which speeding. Moreover, people high on driving anger are also more often involved in minor road crashes (Deffenbacher, Huff, Lynch, Oetting, & Salvatore, 2000). The relationship between driving anger and risky behavior has been found in different countries such as France (Delhomme and Villieux, 2005 and Villieux and Delhomme, 2007), the UK (Lajunen and Parker, 2001 and Lajunen et al., 1998), or Norway (Iversen & Rundmo, 2002). This relationship can be interpreted in two ways. First, people with a “risky” driving style are more likely to be angered by situations that prevent them from adopting this style (such as the ones presented in items from the subfactors “Progress Impeded” or “Traffic obstructions”). Second, adopting risky driving behaviors can be a way to vent the anger felt toward driving situations. For example, concerning speeding, Björklund (2008) showed that the more frequently a driver committed speeding, the less this driver would be irritated by other drivers’ reckless speeding. 1.3. Study of sensation seeking and anger in the same research The impact of sensation seeking and anger on risky driving behavior has been greatly documented, and these two factors are considered the strongest predictors of risky driving behaviors. As such, studies investigating the impact of both factors on risky driving behaviors, allowing to determine, out of the two, which is the strongest predictor, have been conducted by Dahlen and collaborators (Dahlen and White, 2006, Dahlen et al., 2005 and Moore and Dahlen, 2008), Fernandes, Hatfield, and Job (2010) or Schwebel, Severson, Ball, and Rizzo (2006). For example, Dahlen, Martin, Ragan, and Kuhlman (2005) showed that driving anger, measured with the DAS, explained the greatest part of variance in the adoption of various unsafe driving behaviors listed in the DAX or the Driving Survey, such as lack of concentration, loss of control and risky driving, compared to other personality factors such as sensation seeking, impulsiveness and boredom proneness. They also found that adding sensation seeking in a model with driving anger improved significantly the prediction of risky and aggressive driving, suggesting that studies of driving anger may benefit from the addition of these factors. Besides, driving anger and sensation seeking were not significantly correlated. The same conclusions are drawn in all the aforementioned studies: when controlling for several variables (such as driving experience, gender, age, etc.), both driving anger and sensation seeking are significant predictors of the risky behaviors from the DAX and the Driving Survey. Moreover, in these studies effect size for driving anger is consistently higher than effect size for sensation seeking. However, in these studies the behaviors from the DAX are averaged, which may hide important differences between these behaviors. In that regard, speeding is an interesting example. Indeed, it is the most common voluntary risky driving behavior that can be adopted by any driver, and it is the main variable through which drivers regulate the risk they are taking in driving situations. As such, following errors in regulation, adopting a speed that is not adapted to the driving situation plays a great role in crashes. Moreover, its links with sensation seeking as well as anger have been well documented. However speeding is a behavior that evokes sensation seeking more than anger, and as such should be more correlated with sensation seeking than anger. Thus, studying the effects of sensation seeking and anger may reveal results patterns that do not correspond to previous studies considering various risky driving behaviors. More precisely, our hypothesis is that speeding is better predicted by sensation seeking than by anger. Besides, in all of the above-mentioned studies, anger is measured with a driving specific scale (DAS), but sensation seeking is measured with a non specific scale (SSS or AISS). Therefore, the greater relationship risky behavior shares with driving anger than with sensation seeking could be in part attributed to the situational focus DAS implies, and not to anger as a personality trait. To be confident in the conclusions drawn by these studies, it appears necessary to use scales with the same level of specificity, while making sure that they measure independent enough concepts. Moreover, these studies have been run only with American student samples of moderate size (between 73 and 316 students). It would thus be interesting to use a larger sample from a non student population, and from a different culture, in order to investigate whether such a result is indeed generalizable. The goal of our study was to answer these questions. 1.4. Our study We aimed at investigating the effect of driving sensation seeking and driving anger on a particular risky driving behavior: speeding. We measured the two traits with driving-specific scales, respectively the Driving Related Sensation Seeking scale (DRSS; Delhomme, 2002 and Taubman et al., 1996) and the Driving Anger Scale (DAS, Deffenbacher et al., 1994 and Delhomme and Villieux, 2005). Both these driving-specific scales contain items that directly address attitude toward speeding. We focused on young drivers, as they are the age category with the highest fatality rate per head in the population. Indeed, there are about 100 fatalities per million young drivers in Europe, and this rate steadily declines with driver’s age (Organisation for Economic Co-operation and Development, 2006 and SafetyNet, 2009). This peak rate can be explained in part by the fact that young drivers are more likely to adopt ill-adapted speeds. Indeed, as they lack driving experience and have not yet developed driving schemas, they do not always evaluate correctly the risk in a given situation, and adopt speeds that are too high for the situation. As such, they are more likely than older drivers to put themselves in danger while driving, and they are less able than older drivers to anticipate and react appropriately when facing such danger. We ran two studies among young drivers (students and non students) from the MARC survey (“Mobilité, Attitudes, Risque, & Comportements”: Mobility, Attitudes, Risk and Behavior). In the pilot one, we tested whether the effect of driving anger on speeding remained higher than the effect of sensation seeking when using a specific sensation seeking scale among French young drivers. In the second one, we validated the conclusions from the pilot study on a larger sample.

Speeding, like several other risky road behaviors, is explained by a large number of factors, among which personality traits such as sensation seeking and anger play a great role. However, unlike some other risky road behaviors studied in previous research, speeding is predicted by sensation seeking more than by anger. It appears thus essential to take into account the specificities of risky behaviors, as well as to use domain-specific scales when studying these behaviors.