Smartphone use while driving: What factors predict young drivers' intentions to initiate, read, and respond to social interactive technology?

Computers in Human Behavior 76 (2017) 174e183

Contents lists available at ScienceDirect

Computers in Human Behavior journal homepage: www.elsevier.com/locate/comphumbeh

Full length article

Smartphone use while driving: What factors predict young drivers' intentions to initiate, read, and respond to social interactive technology? Cassandra S. Gauld a, *, Ioni Lewis a, Katherine M. White b, Judy J. Fleiter c, Barry Watson a a

Queensland University of Technology, Centre for Accident Research and Road Safety e Queensland (CARRS-Q), Kelvin Grove Campus, Kelvin Grove, 4059, Australia Queensland University of Technology, School of Psychology and Counselling, Kelvin Grove Campus, Kelvin Grove, 4059, Australia c Global Road Safety Partnership, International F ed eration of Red Cross & Red Crescent Soci et es, Route de Pr e-Bois 1, CH-1214, Vernier, Switzerland b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 16 June 2016 Received in revised form 12 June 2017 Accepted 17 July 2017

This study was guided by an extended Theory of Planned Behaviour (TPB) and identified factors that predict young, predominantly university student drivers' intentions to engage in initiating, monitoring/ reading, and responding to social interactive technology (e.g., Facebook, email) on a smartphone. Participants (N ¼ 114) were aged 17e25 years. The standard TPB constructs of attitude, subjective norm, and perceived behavioural control were assessed in an online survey, as well as the additional predictors of anticipated regret, moral norm, mobile phone involvement, and cognitive capture. The results of hierarchical multiple regression analyses showed the standard constructs accounted for 67%, 56%, and 65% of variance in intentions to initiate, monitor/read, and respond, respectively, with the extended variables contributing additional variance. For initiating behaviour, for example, attitude, subjective norm, PBC, and cognitive capture all had significant, positive relationships with intention, while moral norm had a significant, negative relationship. For responding behaviour, attitude, subjective norm, PBC, and cognitive capture all had significant, positive relationships with intention, while anticipated action regret had a significant, negative relationship. These different combinations of significant predictors of intentions for each of the three behaviours (i.e., initiating, monitoring/reading, and responding) suggest that they may be distinct and require different public education message content to influence young drivers’ behaviours. © 2017 Elsevier Ltd. All rights reserved.

Keywords: Theory of planned behaviour Social interactive technology Social media Smartphone Young drivers Moral norm Mobile phone involvement Anticipated regret Cognitive capture

1. Introduction A recent Australian survey found that 75% of mobile phone users now have smartphones; this figure more than doubled between 2011 and 2014 and is expected to reach 91% by 2017 (Telstra, 2014). Social interactive technology accessible on smartphones allows the user to communicate with other people via, for example, social networking sites (e.g., Facebook, Twitter), emails, and also texting and calling. This greater functionality of smartphones (e.g., ability to access the internet and social media), compared to standard mobile phones, means they have a greater potential to distract a driver. Indeed, evidence suggests that 68% of drivers from the Australian state of New South Wales have reported reading emails

* Corresponding author. E-mail address: [email protected] (C.S. Gauld). http://dx.doi.org/10.1016/j.chb.2017.07.023 0747-5632/© 2017 Elsevier Ltd. All rights reserved.

and 25% have reported updating their Facebook status or tweeting while driving (NRMA, 2012). These statistics highlight the fact that drivers are utilising the capabilities of their smartphone beyond talking and texting, thereby increasing their crash risk. Mobile phone conversations and passenger conversations have different effects on driving performance. Specifically, simulator research has found a higher level of driver error associated with having a conversation on a mobile phone (e.g., driver less likely to reduce their speed when approaching hazards) than having a conversation with a passenger (Charlton, 2009). This may be due to drivers experiencing a higher level of cognitive load when engaging in mobile phone conversations than when conversing with a passenger. While passengers are aware of the driving situation and may modify their expectation of the conversation accordingly, the person a driver is conversing with on a mobile phone does not have access to these cues and may, therefore, expect the driver to engage

C.S. Gauld et al. / Computers in Human Behavior 76 (2017) 174e183

in an intense conversation while negotiating difficult traffic situations (e.g., Crundall, Bains, Chapman, & Underwood, 2005; Hunton & Rose, 2005). Mobile phone conversations also tend to have a higher level of cognitive load than other in-vehicle distractions and are, therefore, more distracting (McKnight & McKnight, 1993). Mobile phone conversations are continuous and externally paced (Charlton, 2009) whereas other in-vehicle distraction such as the operation of a satellite navigation system or eating while driving are less distracting as they are usually controlled by the driver. In all states and territories of Australia, including Queensland where the current study was conducted, using a mobile phone in hand-held mode is illegal for all drivers. Drivers can be fined if their mobile phone is in their hand for any reason, including calling, texting, and any of the additional functions accessible on smartphones (e.g., Facebook, internet). Despite the illegal nature of handheld mobile phone use for all Australian drivers, the increased functionality of mobile phones may be encouraging drivers to use them in the hand-held mode (Rudin-Brown, Young, & Lenne, 2013). Also, it is possible that drivers are increasingly concealing their use from outside view, making detection (and enforcement) difficult (Gauld, Lewis, & White, 2014; Rudin-Brown et al., 2013). This concealment, in addition to other factors such as tinted car windows and the difficulty detecting a mobile phone at night, heightens the need for measures other than enforcement on its own to raise awareness of the dangers of smartphone use while driving. Young drivers aged 17e25 years are represented in over 20% of road crash fatalities (Department of Infrastructure and Regional Development, 2014) yet constitute only 12.4% of the population (Australian Bureau of Statistics, [ABS], 2015). Learner drivers (i.e., drivers with an initial licence, aged at least 16 years, who need to be accompanied by a supervising driver at all times) and provisional licence holders (i.e., intermediate licences with specific driving restrictions where the driver can drive alone; the first of which is known as P1 and is followed by P2) under the age of 25 years are not permitted to use a hands-free mobile phone in the Australian state of Queensland, where the current study was conducted. Young drivers aged 18e25 years, however, are more likely than any other age group to use a mobile phone while driving (AAMI, 2012) and a smartphone in particular, suggesting this age group is particularly vulnerable to road trauma. The current study utilised an extended Theory of Planned Behaviour ([TPB]; Ajzen, 1991) to investigate young, predominantly university student drivers’ intentions to initiate, monitor/read and respond to social interactive technology on a smartphone while driving. In addition to the standard predictors of attitude, subjective norm, and perceived behavioural control, the additional predictors of anticipated regret, moral norm, mobile phone involvement, and cognitive capture were assessed. As enforcement of the law regarding smartphone use while driving is challenging, it is proposed that the results of this research could potentially form focal points for public education messages targeting these risky behaviours. The following literature review outlines the problem of young drivers accessing social interactive technology, the importance of investigating the different behaviours of initiating, monitoring/reading, and responding, and the theoretical background (i.e., the TPB). Each of the individual predictors is then discussed along with a justification for its inclusion in this research. 1.1. Literature review 1.1.1. Young drivers Evidence suggests that young drivers may have an increased risk of being involved in road trauma as a result of using their smartphones. Young drivers aged 18e24 years are more likely to call, text,

175

and read emails on their smartphones than older drivers with 12% admitted to updating their Facebook status while driving and 14% admitted to taking a selfie and uploading it onto social media while driving (AAMI, 2012, 2015). Smith (2015) found that young people aged 18e29 years were more likely than any other age group to report feeling distracted when they use their smartphone. Simulator studies have shown that such distraction can increase the risk of yellow-light running (Haque, Ohlhauser, Washington, & Boyle, 2013) and substantially prolong reaction times to detect events originating in the driver's peripheral vision, such as a pedestrian entering a crossing (Haque & Washington, 2013). 1.1.2. Initiating, monitoring/reading, and responding to social interactive technology Currently, only a few studies have investigated the specific behaviours of initiating, monitoring/reading, and responding, which could be applied to the range of social interactive technologies. Waddell and Wiener (2014) found that drivers had greater intentions to engage in, and had reported more actual engagement in, responding behaviours than initiating behaviours and suggested that social pressure to respond may play an important role. Other research supports this conclusion, particularly within the population of young drivers (Atchley, Atwood, & Boulton, 2011; Nemme & White, 2010). Shi, Xiao, and Atchley (2016) categorised texting behaviours into ‘sending, ‘reading’, and ‘replying’ and found that drivers perceived replying and sending as more risky than reading. Contrary to these perceptions, recent research has shown that simply hearing a notification can significantly disrupt performance on an attentiondemanding task (Stothart, Mitchum, & Yehnert, 2015). Of particular note from the Stothart et al. (2015) study is the magnitude of the observed distraction effects which the authors found were comparable to those found when users actively engaged with their mobile phone for calls or texts. The current study addresses this gap in the literature by investigating the specific behaviours of initiating, monitoring/reading, and responding to social interactive technology on smartphones among young drivers. 1.1.3. Theoretical background The Theory of Planned Behaviour ([TPB], Ajzen, 1985) posits that attitude, subjective norm, and perceived behavioural control (PBC) together predict intention. Attitude is defined as how positively (or negatively) the behaviour is evaluated, subjective norm is the extent to which important others approve or disapprove of the behaviour, and PBC is the perceived ease or difficulty of performing the behaviour and can reflect past experience as well as consideration of obstacles (Ajzen, 1991). Overall, the relative importance of each of these constructs varies across behaviours and situations (Ajzen, 1991). In accordance with the tenets of the TPB the current study hypothesised that attitude, subjective norm, and PBC would predict drivers’ intentions to initiate, monitor/read, and respond to social interactive technology on a smartphone while driving in the next week. In particular, the more positive their attitude towards this behaviour, the more they believed it would be approved of by important referents, and the more control they perceived having over the behaviour, the more likely young drivers would be to intend to engage in these behaviours. 1.1.4. Additional predictors Providing their addition is justified on theoretical grounds, extending the TPB to include other predictors may help to explain additional variance in intention and/or behaviour over and above the standard TPB constructs (Ajzen, 1991; Armitage & Conner, 2001; Conner & Armitage, 1998). Past research investigating

176

C.S. Gauld et al. / Computers in Human Behavior 76 (2017) 174e183

predictors of people's intention to use a mobile phone while driving (not specifically a smartphone) often utilise an extended TPB and included other predictors such as moral norm (Nemme & White, 2010), anticipated regret (Gauld et al., 2014), and mobile phone involvement (White, Walsh, Hyde, & Watson, 2012). While this study is unique in its focus on initiating, monitoring/reading, and responding to social interactive technology on smartphones while driving, it extends upon these previous studies and also these additional three predictors. An exploratory measure of cognitive capture was also included. This measure was developed by the authors to determine whether a greater tendency to experience instances of cognitive capture were associated with stronger intentions to engage with social interactive technology while driving. It was hypothesised that anticipated regret, moral norm, mobile phone involvement, and cognitive capture would together significantly account for additional variance in intentions, over and above the standard TPB constructs, as explained in more detail below. 1.1.4.1. Anticipated regret. Anticipated regret relates to the extent that an individual anticipates they will feel regretful if they engage in a particular behaviour (Conner & Armitage, 1998). Anticipated regret was included in the current study as it is reasonable to assume that engaging in illegal and hazardous behaviours, such as accessing social interactive technology on smartphones while driving, may be accompanied by feelings of anticipated regret. It has also been suggested that it may be worthwhile to investigate anticipated inaction regret (see Gauld et al., 2014), which refers to the extent an individual anticipates they will feel regretful if they do not engage in a particular behaviour (e.g., Ajzen & Sheikh, 2013). Anticipated inaction regret may be particularly relevant for responding behaviours where the driver may perceive a social pressure to reply to the incoming communication immediately (Nemme & White, 2010; Waddell & Wiener, 2014). It is also possible that feelings of anticipated inaction regret may occur as the young driver anticipates the regret that may result from not initiating a text to let a friend know they are on their way (Gauld, Lewis, White, & Watson, 2016). The current study, therefore, included the constructs of both anticipated action regret (i.e., the possible negative affect associated with using a smartphone while driving) and anticipated inaction regret (i.e., the possible negative affect associated with not using a smartphone while driving).

calling and texting behaviours (Walsh, White, & Young, 2010). Past research has shown that the more involved people are with their mobile phones, the more likely they are to engage in behaviours with potential negative outcomes, such as mobile phone use while driving (Walsh, White, & Young, 2009; Walsh, White, Cox, & Young, 2011). Mobile phone involvement has been found to be a significant predictor of young people's intentions to use a mobile phone for calls and texts while driving (Gauld et al., 2014; Walsh et al., 2010; White et al., 2012). The current study, therefore, investigated the possibility that drivers who have a higher level of involvement with their smartphone are more likely to intend to initiate, monitor/ read, and respond to social interactive technologies while driving. 1.1.4.4. Cognitive capture. Cognitive capture is a form of inattentional blindness in which an individual is so focused on a secondary activity that they are not cognitively present in their current environment (e.g., Ververs & Wickens, 2000). Mobile phone use, in general, has been shown to result in inattentional blindness (Hyman, Boss, Wise, McKenzie, & Caggiano, 2010). The hazards from mobile phone use may be greater when driving because, due to the nature of inattentional blindness, drivers are often unaware of their poor performance and can be surprised when others tell them they missed seeing something (Hyman et al., 2010). Wickens and Alexander (2009) investigated the concept of attentional tunnelling, a closely related concept to cognitive capture, within the aviation industry and defined it as ‘the allocation of attention to a particular channel of information, or task goal, for a duration that is longer than optimal, given the expected cost of neglecting events on other channels, or failing to perform other tasks’ (p. 182). They suggested that this concept could account for crashes in a wide variety of settings, one of which was mobile phone use while driving (Wickens & Alexander, 2009, p. 183). The current study examined the predictive ability of a newlydeveloped measure of cognitive capture that is specific to smartphone use while driving and is based on key features of the aforementioned definition of attentional tunnelling (Wickens & Alexander, 2009). Whilst exploratory, it was hypothesised that there would be a positive association between greater self-reported experiences of cognitive capture and intentions to initiate, monitor/ read, and respond to social interactive technology while driving. 1.2. The current study

1.1.4.2. Moral norm. Moral norm refers to a person's sense of ethical responsibility based on society's values which can be used as a guide for deciding what is right and wrong (Ajzen, 1991). As accessing social interactive technology on smartphones while driving is an illegal behaviour, and a driver carries the responsibility of getting their passengers safely to their destination, it follows that people's intention to engage with social interactive technology, may also involve moral considerations. Within the road safety context, Gauld et al. (2014) found moral norm to be a significant predictor of young drivers' intention to engage in concealed texting (i.e., young drivers who regarded concealed texting as an immoral behaviour were less likely to intend to engage in it). Similarly, Nemme and White (2010) found that moral norm significantly predicted both drivers' intention and actual behaviour to both send and receive text messages. Whilst it will investigate different distinct behaviours, the current study will build upon these previous findings. 1.1.4.3. Mobile phone involvement. The concept of mobile phone involvement includes additional ways that people interact with their phone when they are not using it to communicate with others (e.g., checking for missed calls, thinking about their phone) and is differentiated from mobile phone use which specifically refers to

Despite the prevalence of young drivers initiating, monitoring/ reading, and responding to social interactive technology while driving, there are few, if any, studies that have investigated predictors of these behaviours. The current study, therefore, aims to address the research question: What factors predict young drivers' intentions to initiate, read, and respond to social interactive technology? To the authors’ knowledge, this study is unique in both its investigation of the specific behaviours of initiating, monitoring/ reading, and responding and its focus beyond calling and texting to the broader array of social interactive technology (e.g., Facebook, twitter, email). This study, therefore, aims to extend theoretical knowledge by applying the TPB to these new behaviours, by building on previous studies that applied the TPB to mobile phone use while driving (specifically calling and texting [e.g., Gauld et al., 2014; Nemme & White, 2010; Walsh et al., 2009]), and by assessing the impact of cognitive capture, not yet examined in the context of social interactive technologies and driving, by including a new measure of cognitive capture developed by the authors. As young drivers are increasingly accessing these additional capabilities, particularly in hand-held mode and often concealed from outside view, other initiatives such as public education messages are increasingly necessary to work alongside enforcement efforts.

C.S. Gauld et al. / Computers in Human Behavior 76 (2017) 174e183

Results of this study could help inform the content of future messages. 2. Method 2.1. Participants Data were collected in 2015 over a 4-month period from February to May. The sampling frame for inclusion in the study was that the participants were aged 17e25 years, held a provisional or open licence, owned a smartphone, and resided in the Australian state of Queensland. Participants (N ¼ 114; 88 females, 26 males) were primarily first year psychology students (n ¼ 90) who had self-selected via an online system at a large Queensland university. Additional participants (n ¼ 24) were recruited from university email lists and from a snowballing of the researchers’ family and friends. Snowball sampling was used to include participants from a wider population beyond 1st year psychology students in efforts to increase the representativeness of the sample. Based on the inclusion criteria, participants owned a smartphone (M ¼ 4.5 years, SD ¼ 1.8 years), were aged between 17 and 25 years (M ¼ 20 years, SD ¼ 2.6 years), 77% had a provisional licence (i.e., a licence which allows a person to drive unsupervised, subject to various restrictions) or an open (i.e., unrestricted) licence (23%). On average, the participants reported driving 7.3 h per week (SD ¼ 4.7 h) in either an automatic (51%) or a manual car (49%). The majority of participants (82%) reported driving most frequently in suburban areas. As commonly occurs in Australian universities, all first year psychology students received partial course credit for their participation and other participants were entered into a draw to win one of three $AUD50 shopping vouchers. 2.2. Procedure Ethical approval was obtained from the University's Human Research Ethics Committee. Prior to commencement of the online survey, information describing the project was provided; including what participation involved, expected benefits and risks, and confidentiality. Potential participants then responded to four screening questions to determine their eligibility to take part in the study (i.e., they live in Queensland, have a provisional or open licence, are aged 17e25 years, and own a smartphone). Completion of the online survey was considered as participants having provided their consent to participate. 2.3. Materials/measures The survey was based on the standard TPB self-report format (Fishbein & Ajzen, 2009). As outlined in Fishbein and Ajzen (2009), the questions were framed in terms of reference to the target behaviour, action, context, and time (i.e., the TACT principle). Thus, the three target behaviours were “initiating social interactive technology use while driving in the next week”, “monitoring/ reading social interactive technology while driving in the next week”, and “responding to social interactive technology while driving in the next week”. Similar to previous studies (e.g., Atchley et al., 2011; Shi et al., 2016), “initiating” was defined for the participants as starting a communication with someone; “monitoring/ reading” was defined as checking your smartphone for communications and/or reading them; and “responding” was defined as replying to a communication that was started by someone else. Participants were informed that “social interactive technology” referred to functions accessed on smartphone through which the user communicates with other people. Examples of social interactive technology were provided noting that such technology

177

included but was not limited to social networking sites (e.g., Facebook, Twitter), text messages, emails, phone calls, and Instagram. It was also stipulated that “while driving” included being stopped at traffic lights or in traffic; anywhere other than being in a parked vehicle and that “a communication’ was a general term that referred to the variety of means by which people share information with each other (e.g., text message, email). The additional predictors of anticipated regret (action and inaction) and moral norm were measured for each of the three target behaviours while mobile phone involvement and cognitive capture were investigated in a general sense (i.e., not specifically for initiating, monitoring/reading, and responding behaviours) due to the nature of the scales. Unless otherwise stated, items were scored on a seven-point Likert scale of (1) strongly disagree to (7) strongly agree. Some of the items were negatively worded and reverse scored prior to analysis. Various demographic variables (e.g., gender, age, highest level of education attained) and frequency of smartphone use (i.e., for initiating, monitoring/reading, and responding) were also assessed. Participants also responded to the question ‘Which of the following social interactive technologies have you ever accessed on your Smartphone while driving?’ They were then presented with 10 previously identified social interactive technologies and asked to mark as many responses as were applicable to them. 2.3.1. Intention Three items measured intention (e.g., “I intend to < initiate/ monitor/read/respond to > social interactive technology on my smartphone while driving in the next week”) were adapted from previous studies (e.g., Norman & Conner, 2006). It formed a reliable scale for initiating (Cronbach's a ¼ .90), for monitoring/reading (Cronbach's a ¼ .91), and for responding (Cronbach's a ¼ .91) in the current study. 2.3.2. Attitude Attitude was assessed using four, seven-point semantic-differential scales (e.g., “For me, social interactive technology on my smartphone while driving in the next week would be” (1) Good to (7) Bad) and was adapted from previous studies (e.g., Elliott & Armitage, 2009; Norman & Conner, 2006). It formed a reliable scale for initiating (Cronbach's a ¼ .89), for monitoring/reading (Cronbach's a ¼ .91), and for responding (Cronbach's a ¼ .91) in the current study. 2.3.3. Subjective norm Three items measured subjective norm (e.g., “People important to me would want me to < initiate/monitor/read/respond to > social interactive technology on my smartphone while driving in the next week”) and were adapted from previous studies (Horvath, Lewis, & Watson, 2012; Norman & Conner, 2006). It formed a reliable scale for initiating (Cronbach's a ¼ .83), for monitoring/reading (Cronbach's a ¼ .82), and for responding (Cronbach's a ¼ .85). 2.3.4. PBC Two items measured PBC (e.g., “I am confident that I could < initiate/monitor/read/respond to > social interactive technology on my smartphone while driving in the next week”) and were adapted from previous studies (e.g., Horvath et al., 2012). They were strongly and positively correlated for initiating, r(114) ¼ .61, p < .001 for monitoring/reading, r(114) ¼ .61, p < .001 and for responding r(114) ¼ .61, p < .001. 2.3.5. Anticipated action regret Anticipated action regret was measured with two items adapted

178

C.S. Gauld et al. / Computers in Human Behavior 76 (2017) 174e183

from previous studies (e.g., Gauld et al., 2014) (e.g., “If I < initiate/ monitor/read/respond to > social interactive technology on my smartphone while driving in the next week I would feel regret”). The items were strongly and positively correlated for initiating r(114) ¼ .69, p < .001, for monitoring/reading r(114) ¼ .68, p < .001 and for responding r(114) ¼ .74, p < .001. 2.3.6. Anticipated inaction regret Anticipated inaction regret was measured with two items modified from a previous study (Ajzen & Sheikh, 2013) (e.g., “To < initiate/monitor/read/respond to > social interactive technology on my smartphone while driving in the next week would make me feel sorry for doing it”). The items had a large and positive correlation for monitoring/reading r(114) ¼ .64; and a medium sized and positive correlation for initiating r(114) ¼ .37, p < .001 and for responding r(114) ¼ .38, p < .001. 2.3.7. Moral norm Two items, based on the scale developed by Godin, Conner, and Sheeran (2005), measured moral norm (e.g., “It would be against my principles to < initiate/monitor/read/respond to > social interactive technology on my smartphone while driving in the next week”) and were strongly and positively correlated for initiating, r(114) ¼ .64, p < .001, for monitoring/reading, r(114) ¼ .67, p < .001 and for responding r(114) ¼ .85, p < .001.

Statistical Package for the Social Sciences (SPSS version 18.0), identified the significant standard and extended TPB predictors of young people's intentions to initiate, monitor/read, and respond in the next week. Attitude, subjective norm, and PBC were entered in step 1 and the extended TPB variables of anticipated action regret, anticipated inaction regret, moral norm, mobile phone involvement, and cognitive capture were entered in step 2. 3.2. Descriptive analysis Table 1 shows that participants in the current study were most likely to text (80.7%), talk (73.7%), and use Facebook (53.5%) while driving. Table 2 describes how often participants reported engaging in initiating, monitoring/reading, and responding to social interactive technology while driving. For example, 32.4% of participants reported initiating a communication on social interactive technology on their Smartphone while driving at least 1 to 2 times per week; 60.7% of participants reported monitoring/reading social interactive technology at least 1 to 2 times per week; and 45.6% of participants reported responding to social interactive technology at least 1 to 2 times per week. This result shows that monitoring/ reading was the most commonly engaged in behaviour and responding the second most common in this sample of young drivers. 3.3. Correlations

2.3.8. Mobile phone involvement Mobile phone involvement was measured with the MPIQ which comprised eight items, developed by Walsh et al. (2010). It assessed participants’ cognitive and behavioural association with their mobile phone (e.g., “I often think about my mobile phone when I am not using it”). Previous studies (e.g., Walsh et al., 2010, 2011) reported a strong reliability for the MPIQ (a ¼ .78 to .80), and it was reliable in this study (a ¼ .85). 2.3.9. Cognitive capture Four items were designed by the authors to measure cognitive capture based on the key features of Wickens and Alexander (2009) definition of attentional tunnelling. Specifically, the cognitive capture measure in the current study included the following features: (1) focus on the smartphone interactions occurs at the expense of focusing on the driving task; (2) the duration of the smartphone interaction is longer than optimal; and (3) driver neglects the driving task or fails to perform other tasks necessary to drive safely. Items include “How often have you found yourself suddenly focussed on your Smartphone rather than on the road when driving?” and “How often have you accidentally failed to carry out a routine driving task (e.g., missed a turn, forgotten to indicate, forgotten to change gear) because you were using your Smartphone while driving?” (1) never to (7) very often). It formed a reliable scale (Cronbach's a ¼ .91). 3. Results 3.1. Overview of data analyses Descriptive analyses were conducted on the behaviours of initiating, monitoring/reading, and responding to social interactive technology while driving in order to determine the frequency with which the participants in this study were engaging in these behaviours and the types of social interactive technology they reported engaging with. To examine the relationship between the standard and extended TPB constructs and intentions to initiate, monitor/read, and respond, bivariate correlations were then conducted. Hierarchical multiple regression analyses, using the

Table 3 presents the means, standard deviations, and bivariate correlations for the independent variables and dependent variables for initiating, monitoring/reading, and responding respectively. For each behaviour, all of the standard TPB predictor variables were significantly and positively correlated with intention for each behaviour as well as anticipated inaction regret, and cognitive capture. Anticipated action regret and moral norm were significantly and negatively correlated with intention for each behaviour. Mobile phone involvement was significantly and positively correlated with intention for monitoring/reading and responding; however, the MPIQ was not significantly correlated with intention to initiate. 3.4. Hierarchical multiple regression analyses predicting intentions to initiate, monitor/read, and respond Three separate hierarchical multiple regressions were conducted to identify which of the standard TPB constructs predicted young drivers’ intentions to initiate, monitor/read, and respond to social interactive technology on their smartphone as well as the extent to which the extended variables predicted intention, over and above the standard TPB constructs. For each of the three behaviours (initiate, monitor/read, and respond), the standard TPB variables of attitude, subjective norm, and PBC were entered into step 1, and the extended variables of anticipated action regret, anticipated inaction regret, moral norm, mobile phone involvement, and cognitive capture were entered into step 2.1 3.4.1. Initiate For initiating behaviour, results of the regression analysis showed that the standard TPB variables entered at step 1 accounted for a significant 67% of the variance in intention to initiate social interactive technology on a smartphone while driving, R2 ¼ .67,

1 The regressions were also run with the demographic variables of gender, age, and education in step 1 and the significant predictors in the final model were the same.

C.S. Gauld et al. / Computers in Human Behavior 76 (2017) 174e183 Table 1 Social interactive technologies participants have ever Accessed while driving (N ¼ 114). Social Interactive Technology

% of participants who had ever accessed this social interactive technology while driving

Texting Talking Facebook Snapchat Email Instagram Twitter Viber Skype Tinder Other

80.7 (n ¼ 92) 73.7 (n ¼ 84) 53.5 (n ¼ 61) 41.2 (n ¼ 47) 30.7 (n ¼ 35) 26.3 (n ¼ 30) 3.5 (n ¼ 4) 3.5 (n ¼ 4) 3.5 (n ¼ 4) 2.6 (n ¼ 3) 9.6 (n ¼ 11)

Note: As participants could select more than one form of social interactive technology, the % column will add up to more than 100.

F(3,110) ¼ 72.68, p < .001. At step 1, the significant predictors were attitude, subjective norm and PBC. The addition of the extended TPB variables at step 2 added a significant 6.1% to the prediction of intention, DR2 ¼ .061, DF(5, 105) ¼ 4.632, p < .01. Overall, model 2 (R2 ¼ .73), containing both the standard and extended TPB constructs, was significant, F(8, 105) ¼ 34.649, p < .001. An inspection of the beta weights for the final model showed that attitude, subjective norm, PBC, and cognitive capture all had significant, positive

179

relationships with intention, while moral norm had a significant, negative relationship with intention (see Table 4). 3.4.2. Monitor/read For monitoring/reading behaviour, results of the regression analysis showed that the standard TPB variables entered at step 1 accounted for a significant 56.2% of the variance in intention to initiate social interactive technology on a smartphone while driving, R2 ¼ .562, F(3,110) ¼ 47.140, p < .001. At step 1, the significant predictors were attitude, subjective norm and PBC. The addition of the extended TPB variables at step 2 added a significant 17.4% to the prediction of intention, DR2 ¼ .174, DF(5, 105) ¼ 13.884, p < .01. Overall, model 2 (R2 ¼ .74), containing both the standard and extended TPB constructs, was significant, F(8, 105) ¼ 36.707, p < .001. An inspection of the beta weights for the final model showed that subjective norm, PBC, and cognitive capture all had significant, positive relationships with intention, while moral norm had a significant, negative relationship with intention (see Table 4). 3.4.3. Respond For responding behaviour, results of the regression analysis showed that the standard TPB variables entered at step 1 accounted for a significant 64.8% of the variance in intention to initiate social interactive technology on a Smartphone while driving, R2 ¼ .648, F(3,110) ¼ 67.480, p < .001. At step 1, the significant predictors were attitude, subjective norm and PBC. The addition of the extended

Table 2 Reported frequencies (%) of initiating, monitoring/reading, and responding to social interactive technology on smartphones while driving. How often do you do the following on your Smartphone while More than once per Daily 1e2 times per driving: day week

1e2 times per month

1e2 times per 6 months

Once a year

Never

Initiate communication on social interactive technology? 3.6 Monitor/read social interactive technology? 8.0 Respond to a communication on social interactive technology? 5.4

13.5 13.4 23.2

7.2 8.0 9.8

6.3 3.6 4.5

40.5 14.3 17.0

9.9 18.9 21.4 31.3 17.0 23.2

Table 3 Means, standard deviations and bivariate correlations for standard and extended TPB constructs for initiating, monitoring/reading, and responding to social interactive technology on a smartphone while driving.

1. Intention

2. Attitude

3. Subjective norm

4. PBC

5. Anticipated action regret

6. Anticipated inaction regret

7. Moral norm

8. Mobile phone involvement

9. Cognitive capture

I M/R R I M/R R I M/R R I M/R R I M/R R I M/R R I M/R R I M/R R I M/R R

M (SD)

1.

2.

3.

4

5

6

7

8

9

2.89 3.73 3.19 2.55 2.99 2.58 1.66 2.06 2.01 4.78 5.00 4.85 4.45 3.87 4.07 1.54 1.85 1.89 5.03 4.46 4.80 3.99 3.99 3.99 2.14 2.14 2.14

e e

.76*** .62*** .73*** e e e

.52*** .47*** .58*** 48*** .45*** .58*** e e e

.61*** .66*** .60*** .55*** .54*** .48*** .35*** .32*** .30** e e e

-.56*** -.59*** -.59*** -.59*** -.57*** -.56*** -.28** -.20* -.18* -.37*** -.50*** -.37*** e e e

.43*** .27** .27** .34*** .21* .20* .64*** .44*** .49*** .31*** .14 .14 -.22** .-.01 -.11 e e e

-.71*** -.72*** -.67*** -.63*** -.60*** -.67*** -.45*** -.33*** -.35*** -.50*** -.58*** -.52*** .75*** .75*** .75*** -.33*** -.15 -.02 e e e

.12 .27** .19* .16* .12 .16* .07 .12 .16* .29** .22** .26** -.03 -.01 .04 .02 .14 .23** -.13 -.18* -.13 e e e

.45*** .52*** .39*** .37*** .23** .26** .13* .24** .16 .36*** .22* .21* -.29** -.18* -.20* .24** .31*** .24** -.36*** -.30** -.27** .34*** .34*** .34*** e e e

(1.60) (1.67) (1.62) (1.41) (1.45) (1.46) (0.85) (0.97) (1.00) (1.16) (1.19 (1.20) (1.54) (1.70) (1.82) (0.78) (0.92) (0.97) (1.49) (1.62) (1.70) (1.15) (1.15) (1.15) (1.13) (1.13) (1.13)

Note. I ¼ Initiating, M/R ¼ Monitoring/Reading, R ¼ Responding. M ¼ mean; SD ¼ standard deviation. *p < .05.

**

p < .01.

***

p < .001.

180

C.S. Gauld et al. / Computers in Human Behavior 76 (2017) 174e183

Table 4 Hierarchical multiple regression analysis with standard and extended TPB constructs for initiating, monitoring/reading, and responding to social interactive technology on a smartphone while driving.

Step 1

Variable

Behaviour

B [95% CI]

b

R2

DR 2

sr2

Attitude

I M/R R I M/R R I M/R R I M/R R I M/R R I M/R R I M/R R I M/R R I M/R R I M/R R I M/R R

0.61 [0.45, 0.77] 0.35 [0.17, 0.53] 0.49 [0.33, 0.60] 0.38 [0.14, 0.62] 0.33 [0.09, 0.58] 0.37 [0.15, 0.59] 0.33 [0.15, 0.51] 0.61 [0.40, 0.82 0.43 [0.26, 0.61] 0.41 [0.23, 0.58] 0.13 [-0.03, 0.29] 0.26 [.01, 0.44] 0.30 [0.03, 0.58] 0.25 [0.03, 0.46] 0.41 [0.18, 0.65] 0.26 [0.08, 0.44] 0.37 [0.18, 0.55] 0.35 [0.18, 0.52] -.00 [-.17, .16] -.11 [-.27, .05] -.21 [-.36, .07] .06 [-.22, .34] .01 [-.20, .22] .09 [-.12, .30] 0.25 [ 0.45, 0.06] 0.27 [ 0.45, 0.08] 0.07 [ 0.24, 0.11] - 0.13 [- 0.28, 0.03] 0.03 [- 0.13, 0.19] - 0.03 [- 0.19, 0.14] .24 [-.07, .41] .43 [0.27,0.60] .24 [0.07,0.40]

.54*** .30*** .44*** .20** .19** .23** .24*** .43*** .32*** .36*** .12 .23** .16* .14* 25** .19** .27** .25*** -.00 -.10 -.24** .03 .00 .05 -.24* -.28** -.07 -.09 .03 -.02 .17** .29*** .17**

.67*** .56*** .65***

.67*** .56*** .65***

.73*** .74*** .73***

.06*** .17*** .08***

.17 .06 .11 .03 .03 .03 .04 .13 .08 .05 .01 .02 .01 .01 .03 .02 .04 .04 .00 .00 .02 .00 .00 .00 .02 .03 .00 .01 .00 .00 .02 .07 .02

Subjective Norm

PBC

Step 2

Attitude

Subjective Norm

PBC

Anticipated Action Regret

Anticipated Inaction Regret

Moral Norm

Mobile Phone Involvement

Cognitive Capture

Note. I ¼ Initiating, M/R ¼ Monitoring/Reading, R ¼ Responding B ¼ unstandardised regression coefficient; b ¼ standardised regression coefficient; sr2 ¼ squared semi-partial correlations. * p < .05. **p < .01. ***p < .001.

TPB variables at step 2 added a significant 8.1% to the prediction of intention, DR2 ¼ .081, DF(5, 105) ¼ 6.275, p < .01. Overall, model 2 (R2 ¼ .73), containing both the standard and extended TPB constructs, was significant, F(8, 105) ¼ 35.295, p < .001. An inspection of the beta weights for the final model showed that attitude, subjective norm, PBC, and cognitive capture all had significant, positive relationships with intention, while anticipated action regret had a significant, negative relationship with intention (see Table 4). 4. Discussion The aim of the current study was to identify factors that predict young drivers’ intentions to initiate, monitor/read, and respond to social interactive technology among young, predominantly university student drivers aged 17e25 years. This study was unique in both its investigation of the specific behaviours of initiating, monitoring/reading, and responding and its focus beyond calling and texting to the broader array of social interactive technology (e.g., Facebook, twitter, email). This study extended theoretical knowledge by applying the TPB to new behaviours, by building on previous studies that applied the TPB to mobile phone use while driving (specifically calling and texting), and by assessing the impact of cognitive capture, not yet examined in the context of social interactive technologies and driving, by including a new measure of cognitive capture developed by the authors. The most common social interactive technologies accessed among this sample of young drivers were text messages, phone calls, and Facebook. Overall, there was support for the ability of the standard TPB constructs to predict intention for each of the three

behaviours investigated, that is, initiating, monitoring/reading, and responding. Specifically, when entered into step 1 of the model, the standard constructs accounted for 67% of variance in intention to initiate, 56% of variance in intention to monitor/read, and 65% of variance in intention to respond to social interactive technology while driving. The more positive attitude held towards these behaviours, the more participants believed these behaviours would be approved of by important referents (subjective norm), and the greater the perception of control over participants had over these behaviours (PBC) were associated with a higher intentions to engage in all three behaviour. As expected, the relative importance of each of these constructs varied across each of the three behaviours examined (Ajzen, 1991). The expectation that the additional variables (i.e., anticipated action regret, anticipated inaction regret, moral norm, mobile phone involvement, and cognitive capture) in the extended TPB would have a significant impact on young drivers' intentions to initiate, monitor/read, and respond to social interactive technology on smartphone, over and above the standard TPB constructs, was supported overall for all three behaviours. The constructs together explained an additional 6% of variance for intention to initiate, an additional 17% of variance for intention to monitor/read, and an additional 8% of variance for intention to respond over and above the standard TPB constructs. Overall, the full model, including the standard and extended variables, accounted for 73% of the variance explained in young people's intention to initiate, 74% for monitoring/reading, and 73% for responding. Differences, however, were found in the significant predictors in the respective final models for initiating, monitoring/reading, and

C.S. Gauld et al. / Computers in Human Behavior 76 (2017) 174e183

responding behaviours and are discussed below. Cognitive capture and mobile phone involvement were measured for smartphone use in general (not specifically in relation to initiating, monitoring/ reading, and responding) and, as such, a general discussion of each of these three predictor variables is presented towards the end of this section. These findings, overall, support the suggestion that the three behaviours may be distinct with different underlying motivations. It is important to note, however, that caution must be exercised when interpreting these results in relation to previous studies which have tended to focus on texting and calling only rather than the additional social interactive technologies accessible on smartphones. 4.1. Initiating As expected and in accordance with the TPB, attitude, subjective norm, and PBC were all significant predictors of intention to initiate social interactive technology use on smartphones while driving. Of the extended variables, moral norm and cognitive capture were significant predictors; however, anticipated action regret, anticipated inaction regret, and mobile phone involvement were not. Overall, in regard to the beta weights, attitude emerged as the strongest predictor and moral norm as the second strongest. Young drivers who believed that initiating social interactive technology while driving was an immoral behaviour were less likely to intend to engage in it. This result aligns with previous TPBbased research for the specific behaviour of sending text messages (Nemme & White, 2010). While not investigating the array of social interactive technologies in this previous study, texting falls within this broad category of smartphone applications. This result also supports the idea that moral norm may be a valuable predictor for illegal behaviours (Ajzen, 1991) because initiating social interactive technology use while driving is most likely to take place in the illegal hand-held mode (Rudin-Brown et al., 2013). Indeed, moral norm was found to be a significant predictor of young drivers concealing their texting (Gauld et al., 2014), thereby supporting this proposition. In addition, as past research based upon the behaviours of calling and texting has suggested that initiating communications while driving is perceived as particularly risky (e.g., Shi et al., 2016), it is possible that many young drivers feel a heightened sense of moral responsibility when they are driving with passengers. Neither form of anticipated regret (action nor inaction) was a significant predictor of intention to initiate social interactive technology use while driving. Participants who reported higher feelings of anticipated regret for initiating social interactive technology (i.e., anticipated action regret) were not less likely to intend to do so while driving. This finding is surprising given the illegal nature of the behaviour and the increase in crash risk. However, as social interactive technology is usually accessed in hand-held mode, it is possible that individuals are concealing their use and are, therefore, making enforcement difficult (Gauld et al., 2014; Rudin-Brown et al., 2013). Consequently, young drivers may perceive that apprehension is unlikely, thereby reducing their feelings of anticipated action regret. For the measure of anticipated inaction regret, participants who reported higher feelings of regret for not initiating social interactive technology were not found to be more likely to intend to engage in this behaviour. This result does not support the idea presented by Gauld et al. (2014) who suggested that young drivers may feel more anticipated regret for not staying in touch with their friends. Previous qualitative research (e.g., Gauld et al., 2016), however, has found that many young drivers are aware of the dangers associated with distracted driving so their perception of the risk may have, in this case, outweighed the possible feeling of anticipated inaction regret.

181

4.2. Monitoring/reading In accordance with the TPB, subjective norm and PBC emerged as a significant predictor of intention to monitor/read social interactive technology on smartphones while driving. Inconsistent with prior TPB research in the area of general mobile phone use while driving (i.e., calling and texting), attitude was not a significant predictor of intention to monitor/read social interactive technology (e.g., Gauld et al., 2014; Nemme & White, 2010). Of the three behaviours, monitoring/reading was found to be the most prevalent in the current study (see Table 2) and the mean for the attitude score was highest for monitoring/reading. This suggests that young drivers in this study felt the most positive about engaging in it, possibly because they don't believe it is as risky as initiating and responding (Shi et al., 2016). Despite this perception, recent experimental evidence has suggested that simply hearing a notification is as distracting as actually interacting with the phone (Stothart et al., 2015). Of the additional predictors, cognitive capture and moral norm emerged as significant. Anticipated action regret, anticipated inaction regret, and mobile phone involvement were not. Overall, cognitive capture had the highest beta weight and moral norm as the second highest. As for initiating behaviour and for previous studies (e.g., Gauld et al., 2014 which investigated concealed texting), the more a young driver perceived monitoring/reading communications on social interactive technology as immoral, the less likely they were to report an intention to engage in this behaviour. Other studies (e.g., Shi et al., 2016) have found that monitoring/reading is perceived as a less risky behaviour than initiating and responding and it follows, therefore, that feelings of anticipated action regret may be low as found in the current study. While included as an exploratory variable, the results with regard to anticipated inaction regret were contrary to predictions given young drivers have reported that an advantage of monitoring/reading social interactive technology while driving is being able to keep up to date with friends’ plans (Gauld et al., 2016). As the reported risk perception associated with monitoring/reading is less than initiating and responding (e.g., Shi et al., 2016), it is possible that this behaviour is simply less motivated by affect. 4.3. Responding As expected, attitude, subjective norm, and PBC were all significant predictors of intention to respond to social interactive technology on smartphones while driving. Of the additional predictors, anticipated action regret and cognitive capture were significant predictors; however, anticipated inaction regret, moral norm, and mobile phone involvement were not. PBC and subjective norm emerged as the variables with the equally strongest beta weights. The significance of anticipated action regret for responding in the current study was expected. In addition to the heightened crash risk associated with responding as well as the possibility of police prosecution, qualitative research has suggested that one of the disadvantages of responding is the unknown nature and length of the incoming communication (Gauld et al., 2016). If, for example, the communication is emotionally-based or contains bad news, there is the greater potential for distraction (Briggs, Hole, & Land, 2011) and the possibility of feeling regret may be heightened. The expectation that anticipated inaction regret would be a significant predictor of intention to respond due social pressure (e.g., Waddell & Wiener, 2014) was not supported in the current study. It is possible that this finding is due to the array of social interactive technologies investigated in this study which may be associated with different expected response timeframes (e.g., emails have a

182

C.S. Gauld et al. / Computers in Human Behavior 76 (2017) 174e183

longer acceptable response timeframe than text messages). For moral norm, while the beta weight indicated a negative association with intention (as for initiating and monitoring/reading), it was not significant as anticipated action regret was a stronger predictor. 4.4. Mobile phone involvement The expected result that participants who scored higher on mobile phone involvement would be more likely to engage with social interactive technology while driving did not emerge for any of the three behaviours. Items in the MPIQ (Walsh et al., 2010) refer to mobile phones in a general manner, as opposed to social interactive technology on smartphones specifically (e.g., ‘I often think about my mobile phone when I am not using it’) and this discrepancy may have impacted the results. If the measure was valid, this result did not support previous TPB-based studies where mobile phone involvement was found to be a significant predictor of mobile phone use while driving (specifically calling and texting) over and above the standard TPB constructs (Gauld et al., 2014; White et al., 2012). 4.5. Cognitive capture The current study supported the exploratory hypothesis for cognitive capture and found that there was a positive association between greater self-reported experiences of cognitive capture and intentions to initiate, monitor/read, and respond to social interactive technology while driving. This finding is especially the case for monitoring/reading where cognitive capture was found to be the most strongly associated with intention. The mean score on the cognitive capture scale, however, was low (i.e., it was a mean score of 2.14 which fell between the scale responses of ‘rarely’ and ‘not often’; SD ¼ 1.13). Although it appeared that most participants did not report experiencing cognitive capture, it should be noted that, due to the very nature of the construct (i.e., almost complete focus on the smartphone interaction at the expense of the driving task), drivers are often unaware of their resulting poor performance on the primary task (Hyman et al., 2010; Simons, 2000; Wickens & Alexander, 2009). Although found to be a significant predictor of intention to engage in all three behaviours, the predictive ability of this concept in the current study may be somewhat conservative. Given that this is a newly-constructed measure, future research should investigate this concept further. 4.6. Practical implications As laws regarding smartphone use while driving are difficult to enforce, other countermeasures, such as public education messages, are required as part of a multi-pronged approach to address this issue. Different predictors were found for the three behaviours of initiating, monitoring/reading, and responding, suggesting that unique message content may be required to target these three behaviours. A message targeting initiating behaviour, for example, may highlight moral norm by emphasising the illegal nature of the behaviour and the importance of adhering to the road rules. A message targeting monitoring/reading may focus on the concept of cognitive capture and the associated neglect of the driving task and the increase in driver error (e.g., lane wandering). A message targeting responding behaviour could emphasis the regret a young driver may feel (i.e., anticipated regret) if they respond to a communication, perhaps in relation to injuring their passengers. In addition to these suggestions, young drivers’ perception that monitoring/reading is not as risky as initiating and responding should be challenged, given a recent study found that simply hearing a notification was just as distracting as interacting with the

phone (Stothart et al., 2015) and monitoring/reading was reported as being the most prevalent in the current study (see Table 2). A public education message, for example, could depict a young driver monitoring/reading their smartphone in slow-moving traffic and inadvertently running into the car in front of them, or a young driver could choose not to read a notification and safely stop in time when a dog runs out onto a suburban road. Please note that the content proposed are suggestions only and all messages would need to be evaluated to determine their potential persuasiveness. 4.7. Strengths and limitations To the authors' knowledge, this study was the first theoreticallydriven investigation of young drivers' intentions to engage in the three behaviours of initiating, monitoring/reading, and responding to the range of social interactive technology on smartphone (beyond calling an texting) among young drivers. The study's focus on young drivers who are the highest owners of smartphone and the highest users of the social interactive technology capabilities, reinforce the study's high degree of practical applicability. The main limitation of this study was that it did not include a follow-up behaviour measure. The use of self-report measures for illegal behaviours may have caused some participants to respond according to social desirability. Future studies should include more males as the sample in this study was predominantly female2 (77%). Females are less likely to be involved in road crash (Queensland Government, 2016) and are less likely to be apprehended for mobile phone use while driving in the Australian state of Queensland (Department of Transport and Main Roads, 2016). In addition, most of the participants in this study were university students (79%). University students may be more educated than the general population, particularly regarding the dangers of smartphone use while driving, thereby potentially limiting the generalisability of the findings. 5. Conclusion This study provides a unique investigation of smartphone use and the three distinct behaviours of initiating, monitoring/reading, and responding to the array of social interactive technology that is accessible on such devices by young drivers. The majority of young drivers now own smartphones and many are accessing these additional capabilities while driving (e.g., Facebook). This study adds to the extant literature and extends on previous research that utilised the TPB to investigate psychosocial predictors of general mobile phone use (i.e., talking and/or texting) while driving. Importantly, the present study examined the impact of cognitive capture, a newly developed construct, on initiating, monitoring/ reading, and responding to social interactive technology. The challenges associated with enforcing the laws regarding smartphone use while driving (e.g., drivers concealing their use) highlights the importance of developing other countermeasures such as public education messages. Results of this study could help inform the content of future messages. In the extended TPB models, the results of this study provide strong support for the predictive utility of the standard TPB constructs for the deliberate behaviours of initiating and responding. It also provides strong support for the inclusion of the additional predictors of cognitive capture for the intention to engage in all three behaviours, moral norm for intentions to initiate and

2 While the sample was predominantly female, independent samples t-tests for the standard TPB measures of intention, attitude, subjective norm, and perceived behavioural control did not reveal any significant differences between the genders.

C.S. Gauld et al. / Computers in Human Behavior 76 (2017) 174e183

monitor/read, and anticipated action regret for intention to respond. Importantly, in the current study, further support is given to emerging evidence that initiating, monitoring/reading, and responding may be distinct behaviours and for the need to treat them as such in subsequent research. References AAMI. (2012). Young driver index 2012. Retrieved from www.aami.com.au/sites/ default/files/fm/news/Young%20Driver%20Index%22012_12_14.pdf. AAMI. (2015). AAMI targets social stigma to tackle distracted driving. Retrieved from https://www.aami.com.au/media-centre/aami-targets-social-stigma-tackledistracted-driving.html. Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. In J. Kuhl, & J. Beckmann (Eds.), Action control: From cognition to behaviour (pp. 11e39). Berlin: Springer-Verlag. Ajzen, I. (1991). The theory of planned behavior. Organizational Behaviour and Human Decision Processes, 50(2), 179e211. http://dx.doi.org/10.1016/07495978(91)90020-T. Ajzen, I., & Sheikh, S. (2013). Action versus inaction: Anticipated affect in the theory of planned behavior. Journal of Applied Social Psychology, 43(1), 155e162. Armitage, C., & Conner, M. (2001). Efficacy of the theory of planned behaviour: A meta-analytic review. British Journal of Social Psychology, 40, 471e499. http:// dx.doi.org/10.1348/014466601164939. Atchley, P., Atwood, S., & Boulton, A. (2011). The choice to text and drive in younger drivers: Behavior may shape attitude. Accident Analysis and Prevention, 43(1), 134e142. http://dx.doi.org/10.1016/j.aap.2010.08.003. Australian Bureau of Statistics. (2015). Australian demographic statistics. December, 2014. Retrieved from http://www.abs.gov.au/ausstats/[email protected]/mf/3101.0. Briggs, G. F., Hole, G. J., & Land, M. F. (2011). Emotionally involving telephone conversations lead to driver error and visual tunnelling. Transportation Research Part F: Traffic Psychology and Behaviour, 14(4), 313e323. Charlton, S. (2009). Driving while conversing: Cell phones that distract and passengers who react. Accident Analysis and Prevention, 41, 160e173. Conner, M., & Armitage, C. (1998). Extending the theory of planned behaviour: A review and avenues for further research. Journal of Applied Social Psychology, 2(15), 1429e1464. http://dx.doi.org/10.1111/j.1559-1816.1998.tb01685.x. Crundall, D., Bains, M., Chapman, P., & Underwood, G. (2005). Regulating conversation during driving: A problem for mobile telephones? Transportation Research Part F, 8, 197e211. Department of Infrastructure and Regional Development. (2014). Road trauma Australia 2014 statistical summary. Retrieved from http://bitre.gov.au/ publications/ongoing/files/Road_trauma_Australia_2014_statistical_summary_ N_ISSN.pdf. Department of Transport and Main Roads. (2016). Mobile phone infringements. Elliott, M., & Armitage, C. (2009). Promoting drivers' compliance with speed limits: Testing an intervention based on the theory of planned behaviour. British Journal of Psychology, 100(1), 111e132. http://dx.doi.org/10.1348/ 000712608X318626. Fishbein, M., & Ajzen, I. (2009). Predicting and Changing Behavior: The Reasoned Action Approach. New York: Psychology Press. Gauld, C., Lewis, I., & White, K. M. (2014). Concealing their communication: Exploring psychosocial predictors of young drivers' intentions and engagement in concealed texting. Accident Analysis & Prevention, 62, 285e293. http:// dx.doi.org/10.1016/j.aap.2013.10.016. Gauld, C., Lewis, I., White, K. M., & Watson, B. (2016). Key beliefs influencing young drivers' engagement with social interactive technology on their smartphones: A qualitative study. Traffic Injury Prevention, 17(2), 128e133. http://dx.doi.org/ 10.1080/15389588.2015.1047014. Godin, G., Conner, M., & Sheeran, P. (2005). Bridging the intention-behaviour 'gap': The role of moral norm. British Journal of Social Psychology, 44(4), 497e512. Haque, M. M., Ohlhauser, A. D., Washington, S., & Boyle, L. N. (2013). Examination of distracted driving and red light running: Analysis of simulator data. In Paper presented at the 92nd annual meeting of transportation research board, Washington DC, USA. Haque, M. M., & Washington, S. (2013). Effects of mobile phone distraction on drivers' reaction times. Journal of Australasian College of Road Safety, 24(3),

183

20e29. Horvath, C., Lewis, I., & Watson, B. (2012). Peer passenger identity and passenger pressure on young drivers' speeding intentions. Transportation Research Part F: Traffic Psychology and Behaviour, 15(1), 52e64. http://dx.doi.org/10.1016/ j.trf.2011.11.008. Hunton, J., & Rose, J. (2005). Cellular telephones and driving performance: The effects of attentional demands on motor vehicle crash risk. Risk Analysis, 25, 855e866. Hyman, I. E., Jr., Boss, S. M., Wise, B. M., McKenzie, K. E., & Caggiano, J. M. (2010). Did you see the unicycling clown? Inattentional blindness while walking and talking on a cell phone. Applied Cognitive Psychology, 24(5), 597e607. McKnight, A. Scott, & McKnight, A. James (1993). The effect of cellular phone use upon driver Attention. Accident Analysis and Prevention, 25(3), 259e265. Nemme, H. E., & White, K. M. (2010). Texting while driving: Psychosocial influences on young people's texting intentions and behaviour. Accident Analysis and Prevention, 42(4), 1257e1265. http://dx.doi.org/10.1016/j.aap.2010.01.019. Norman, P., & Conner, M. (2006). The theory of planned behaviour and binge drinking: Assessing the moderating role of past behaviour within the theory of planned behaviour. British Journal of Health Psychology, 11(1), 55e70. http:// dx.doi.org/10.1348/135910705X43741. NRMA. (2012). Staying connected while driving can be wheely distracting. Retrieved from http://www.nrma.com.au/driven-distraction-mobile-phones. Queensland Government. (2016). Driver demographics. Retrieved from https://data. qld.gov.au/dataset/crash-data-from-queensland-roads/resource/dd13a8892a48-4b91-8c64-59f824ed3d2c. Rudin-Brown, C. M., Young, K., & Lenne, M. (2013). Melbourne drivers' observed use of mobile phones: Could there be unintended consequences of partial bans? In M. Regan, & T. Victor (Eds.), Human factors in road and rail transport: Driver distraction and inattention: Advances in research and countermeasures (pp. 313e326). Farnham, Great Britain: Ashgate Publishing Ltd. Shi, J., Xiao, Y., & Atchley, P. (2016). Analysis of factors affecting drivers' choice to engage with a mobile phone while driving in Beijing. Transportation Research Part F: Traffic Psychology and Behaviour, 37, 1e9. http://dx.doi.org/10.1016/ j.trf.2015.12.003. Simons, D. J. (2000). Attentional capture and inattentional blindness. Trends in Cognitive Sciences, 4(4), 147e155. Smith, A. (2015). US Smartphone use in 2015. Retrieved from http://www. pewinternet.org/2015/04/01/us-smartphone-use-in-2015/. Stothart, C., Mitchum, A., & Yehnert. (2015). The attentional cost of receiving a cell phone notification. Journal of Experimental Psychology: Human Perception and Performance. http://dx.doi.org/10.1037/xhp0000100. Telstra. (2014). Cross-platform consumers: Media multi-tasking and multi-screening in today's mobile landscape. Retrieved from http://www.telstra.com.au/aboutus/ download/document/telstra-cross-platform-consumers.pdf. Ververs, P. M., & Wickens, C. D. (2000). Designing head-up displays (HUDs) to support flight path guidance while minimizing effects of cognitive tunneling. Human Factors and Ergonomics Society Annual Meeting Proceedings, 44(13), 45e48. Waddell, L. P., & Wiener, K. K. K. (2014). What's driving illegal mobile phone use? Psychosocial influences on drivers' intentions to use hand-held mobile phones. Transportation Research Part F: Traffic Psychology and Behaviour, 22, 1e11. http:// dx.doi.org/10.1016/j.trf.2013.10.008. Walsh, S., White, K. M., Cox, S., & Young, R. (2011). Keeping in constant touch: The predictors of young Australians' mobile phone involvement. Computers in Human Behavior, 27(1), 333e342. Walsh, S., White, K. M., & Young, R. (2009). The phone connection: A qualitative exploration of how belongingness and social identification relate to mobile phone use amongst Australian youth. Journal of Community & Applied Social Psychology, 19(3), 225e240. http://dx.doi.org/10.1002/casp.983. Walsh, S., White, K. M., & Young, R. (2010). Needing to connect: The effect of self and others on young people's involvement with their mobile phones. Australian Journal of Psychology, 62(4), 194e203. White, K. M., Walsh, S., Hyde, M., & Watson, B. (2012). Connection without caution?: the role of mobile phone involvement in predicting young people's intentions to use a mobile phone while driving. Journal of the Australasian College of Road Safety, 23(1), 16e21. Wickens, C. D., & Alexander, A. L. (2009). Attentional tunneling and task management in synthetic vision displays. The International Journal of Aviation Psychology, 19(2), 182e199. http://dx.doi.org/10.1080/10508410902766549.