Download pdf - Mobile telephones and other communication devices and ...€¦ · 4.3. Do hands-free and handheld mobile phones affect driving performance in a different manner? 4.4. What are drivers’

Mobile telephones and other communication devices and their impact on traffic safety A review of the literature Katja KircherChristopher PattenChrister Ahlström 2011

Mobile telephones and other communication devices and their impact on traffic safety A review of the literature 3

Quality review Internal peer review was performed on 2011-10-21 by Dr. Lena Nilsson. Dr. Katja Kircher has made alterations to the final manuscript of the report. The research director of the project manager Dr. Jan Andersson examined and approved the report for publication on 2011-10-27.

1. Abstract

2. Introduction

2.1. Literature search

2.2. Research methods

2.3. Driver distraction and inattention

2.4. Scope

2.5. Types of communication

3. Frame of Reference

4. Main Findings

4.1. In what way are drivers affected when talking or texting?

4.2. Does practice improve the ability to drive while handling a mobile phone?

4.3. Do hands-free and handheld mobile phones affect driving performance in a different manner?

4.4. What are drivers’ attitudes to mobile phone usage when driving?

4.5 How do drivers use their mobile communication devices?

4.6. Do drivers adapt their driving behaviour to device-usage?

4.7. How dangerous is it for drivers to talk or text?

4.8. Does a ban on handheld mobile phones work with respect to compliance?

4.9. Do bans have a positive effect on road traffic safety?

5. Summary

References

Annex


Table of Contents

PAge

1. Abstract

2. Introduction

2.1. Literature search

2.2. Research methods

2.3. Driver distraction and inattention

2.4. Scope

2.5. Types of communication

3. Frame of Reference

4. Main Findings

4.1. In what way are drivers affected when talking or texting?

4.2. Does practice improve the ability to drive while handling a mobile phone?

4.3. Do hands-free and handheld mobile phones affect driving performance in a different manner?

4.4. What are drivers’ attitudes to mobile phone usage when driving?


4.6. Do drivers adapt their driving behaviour to device-usage?

4.7. How dangerous is it for drivers to talk or text?

4.8. Does a ban on handheld mobile phones work with respect to compliance?

4.9. Do bans have a positive effect on road traffic safety?

5. Summary

References

Annex

61112131415151618182021222428303537404353

Mobile telephones and other communication devices and their impact on traffic safety A review of the literature6

1 Abstract The Swedish National Road and Transport Research Institute (VTI) was com-missioned by the Swedish government to compile research on mobile phone and other communication device usage whilst driving. To this end a literature search with focus on the most recent years was conducted, and the results are presented in question-and-answer form. Typical examples of distracted driving include mobile telephone use, daydreaming or eating a sandwich. Distracting activities can be subdivided into different modalities; visual (eyes off the road), cognitive/auditory (mind off the road) and manual distraction (hands off the wheel). Many real-life distractions involve multiple modalities. For example when someone calls you (auditory cue), you pick up and answer the phone (visual and manual) and engage in the conversa-tion (cognitive/auditory). The more one investigates how mobile phone use affects traffic safety the more one realises how complex this question really is. An abridged summary of the literature reviewed in this report is listed below:

• Driving performance is impaired by talking on a mobile phone in controlled laboratory, simulator and field studies.

• There is no evidence suggesting that hands-free mobile phone use is less risky than handheld use. Nevertheless, a large percentage of drivers think that hands-free sets are much safer to use than handheld sets.

• There is broad consensus that driving performance is impaired by visual-manual interactions with mobile phones, such as engaging in social media or sending text messages. The degree of the impairment is reflected by the time the distracting task takes to complete, the complexity of the task, the capabilities of the driver and the current traffic circumstances.

• Telephone use, both for calling and texting, is most frequent with younger drivers and high-mileage drivers. More men than women admit to phone usage while driving.

• Most EU countries require hands-free equipment for legal mobile phone use but none of them have a total ban. There appears to be a measurable compliance with the bans for the first year or so, but then frequencies of handheld use return to pre-ban levels. Detailed insurance claims-studies from USA suggest that there have been no insurance claim-benefits of the banning of handheld phones or the specific banning of text messaging.

Deterioration of driver performance due to mobile phone use is an established fact, but as of today we cannot draw any conclusions about the actual safety impact of telephone use in terms of crash frequency. Even though the traffic system is forgiving, things can go awry if the driver is distracted while something unexpected occurs.


Sammanfattning Statens väg- och transportforskningsinstitut (VTI) har fått i uppdrag av regeringen att sammanställa relevant forskning på området användande av mobiltelefon och annan kommunikationsutrustning under bilkörning. En litteraturgenomgång har utförts och resultaten presenteras i form av frågor med ett kort sammanfattande svar och ett mer utförligt svar för de som vill veta mer.

Förardistraktion kan vara att prata i telefon, att dagdrömma eller att äta en smör-gås samtidigt som man kör. Distraktion kan delas in i visuell distraktion (att titta bort från vägen), kognitiv/auditiv distraktion (att tänka på något annat än trafiken) och motorisk distraktion (att ta händerna från ratten). Distraherande uppgifter består ofta av flera av dessa komponenter. Exempelvis när någon ringer till dig (auditivt stimuli), du plockar upp telefonen och svarar (visuell och motorisk dist-raktion) och börjar prata (kognitiv och auditiv distraktion).

Ju mer man bryter ner problemet och fördjupar sig i hur mobiltelefoni påverkar trafiksäkerheten, desto mer förstår man hur komplex frågeställningen är. Det vi vet idag kan sammanfattas i följande punkter:

• Kontrollerade studier i labbmiljö, i körsimulatorer och i fältförsök visar att förarens prestationsförmåga försämras av att prata i telefon.

• Det finns ingenting som pekar på att det skulle vara mindre farligt att använda handsfree jämfört med en handhållen telefon medan man pratar. Ändå tror många förare att det är mycket säkrare att använda handsfree än handhållen telefon.

• Uppgifter som att skriva ett SMS eller slå ett telefonnummer, som kräver både visuell och motorisk uppmärksamhet, försämrar förarens prestations-förmåga. Hur mycket prestationsförmågan försämras beror på tiden det tar att slutföra uppgiften, hur svår uppgiften är, hur duktig föraren är samt den rådande trafiksituationen.

• Yngre förare och förare som tillbringar mycket tid bakom ratten tenderar att använda sin telefon till en högre grad än andra grupper. Män säger sig också använda sin telefon oftare än kvinnor.

• De flesta EU-länderna tillåter endast handsfree-telefoner, men det finns inget land som har ett totalförbud mot mobiltelefoni under körning. Effekten av en lag håller i sig i ungefär ett år, sedan återgår telefonianvänd- andet till samma nivåer som före lagstiftningen. Studier från amerikanska försäkringsbolag visar att försäkringsanspråken inte har minskat vare sig till följd av lagar mot handhållna telefoner eller till följd av lagar som för-bjuder SMS.

Det är ett faktum att mobiltelefonianvändande under körning försämrar förarens prestationsförmåga, men vi kan idag inte säga hur och om detta verkligen på-verkar trafiksäkerheten i termer av antal olyckor. Trafikmiljön är idag väldigt förlåtande, men om föraren är distraherad vid ett tillfälle då något oväntat händer så är sannolikheten högre att det slutar illa.


Zusammenfassung Das Staatliche Schwedische Straßen- und Verkehrsforschungsinstitut (VTI) wurde von der schwe-dischen Regierung beauftragt, Forschungsergebnisse zur Nutzung von Mobiltelefonen und anderen Kommunikationsmitteln im Straßenverkehr zusammenzustellen. Hierzu wurde eine Auswertung der aktuellen Literatur vorgenommen. Die Resultate werden als Fragen und Antworten präsentiert.

Typische Beispiele für Ablenkung beim Autofahren sind das Telefonieren, ein Brot zu essen oder Tagträumen nachzuhängen. Ablenkungen können die Aufmerksamkeit des Fahrers auf unter-schiedliche Weise beeinträchtigen: Sie können visueller, kognitiv-auditiver, oder manueller Natur sein, je nachdem ob der Blick oder die Gedanken vom Verkehrsgeschehen abschweifen oder die Hände vom Lenkrad genommen werden. Viele der üblichen Ablenkungen haben mehrere Ursachen. Wenn man zum Beispiel angerufen wird (ein auditives Signal), hebt man ab (visuelle und manuelle Beanspruchung), um sich dann dem Gespräch zu widmen (kognitiv-auditive Belastung).

Je mehr man sich damit beschäftigt, welchen Einfluss die Nutzung von Mobiltelefonen auf die Verkehrssicherheit hat, desto deutlicher wird es, wie komplex das Thema ist. Es folgt eine gekürzte Zusammenfassung der Literatur, die im Rahmen dieses Reports ausgewertet wurde:

• Kontrollierte Labor-, Simulator- und Feldstudien zeigen, dass die Fahrleistung beeinträchtigt wird, wenn man gleichzeitig telefoniert.

• Es gibt keine Hinweise darauf, dass Freisprecheinrichtungen das Risiko im Vergleich zu hand-gehaltenen Telefonen verringern. Trotzdem glaubt ein großer Prozentsatz der Autofahrer, dass die Benutzung von Freisprechanlagen sicherer ist als das Telefon beim Sprechen in der Hand zu halten.

• Es herrscht große Einigkeit, dass die visuell-manuelle Bedienung eines Mobiltelefons, wie zum Beispiel die Aktualisierung von Informationen in sozialen Netzwerken oder das Schicken von SMS, die Fahrleistung beeinträchtigt. Das Ausmaß der Beeinträchtigung hängt ab von der Dauer, die diese Tätigkeit in Anspruch nimmt, von der Komplexität der Aufgabe, den Fähigkeiten des Fahrers und der aktuellen Verkehrslage.

• Am häufigsten werden Mobiltelefone von jüngeren Fahrern und Vielfahrern sowohl zum Telefonieren als auch für SMS benützt. Mehr Männer als Frauen verwenden Mobiltelefone während der Fahrt.

• Die meisten EU-Länder schreiben die Anwendung von Freisprechanlagen vor, doch nirgendwo in Europa ist das Telefonieren beim Fahren vollständig verboten. Während des ersten Jahres nach Inkrafttreten der Vorschrift von Freisprechanlagen wurde ein Rückgang der Nutzung von handgehaltenen Telefonen verzeichnet, doch anschließend stieg die Verwendung dieser Telefone wieder auf das Niveau an, das vor der Gesetzgebung gemessen wurde. Studien, die auf Daten amerikanischer Versicherungen beruhen, zeigen, dass die Versicherungsansprüche nicht gesun-ken sind, nachdem Vorschriften für Freisprechanlagen und SMS-Verbote in Kraft getreten sind.

Es ist erwiesen, dass die Fahrleistung bei gleichzeitiger Verwendung eines Mobiltelefons sinkt, dennoch können wir aufgrund der heutigen Datenlage nicht eindeutig sagen, wie und ob die Verkehrssicherheit, bezogen auf die Anzahl der auftretenden Unfälle, wirklich beeinträchtigt wird. Auch wenn die heutzutage herrschenden verkehrstechnischen Rahmenbedingungen Fahrfehler ver-zeihen, kann dennoch etwas passieren, wenn der Fahrer abgelenkt ist und etwas Unvorhergesehenes geschieht.


Resumen El gobierno sueco ha solicitado al Instituto Nacional de Investigación de Transportes y Carreteras de Suecia (VTI, por sus siglas en sueco) la realización de un estudio acerca de la utilización del telé-fono móvil, así como de otros dispositivos de comunicación, durante la conducción. Con este objetivo, se ha realizado una búsqueda de toda la bibliografía, referente a los últimos años, relacionada con este tema. Los resultados se presentan en forma de preguntas y respuestas.

Entre los ejemplos típicos de distracciones al volante se encuentran utilizar el teléfono móvil, pensar en ”nuestras cosas” (conducción subconsciente) o comerse un bocadillo. Las actividades que provocan distracciones se pueden subdividir en varias modalidades, a saber, distracciones visuales (apartar la vista de la carretera), distracciones cognitivas/auditivas (no prestar atención a la conducción) y dist-racciones manuales (soltar las manos del volante). Varias de estas modalidades están presentes en muchas de las distracciones de la vida cotidiana. Por ejemplo, cuando alguien nos llama (distracción auditiva), cogemos el teléfono (distracción visual), descolgamos (distracción manual) y entablamos una conversación (distracción cognitiva/auditiva). Cuanto más se investiga acerca del impacto de la utilización del teléfono móvil en la seguridad vial, más nos damos cuenta de la complejidad del asunto. A continuación, se realiza un resumen abreviado de la bibliografía consultada durante la realización de este estudio:

• Según estudios controlados, realizados tanto en laboratorios como en simuladores, y estudios de campo, el rendimiento al volante se ve limitado al hablar por el teléfono móvil.

• No hay pruebas que sugieran que la utilización de un teléfono móvil en modo manos libres sea menos peligrosa que el hecho en sí de cogerlo para responder la llamada. Sin embargo, un gran porcentaje de conductores considera que los dispositivos de manos libres son mucho más seguros.

• Existe un gran consenso acerca de que la interacción visual y manual con el teléfono móvil, como podría ser interactuar en redes sociales o enviar mensajes de texto, limita el rendimiento de la conducción. El grado de limitación está determinado por el tiempo que se emplea en com-pletar dicha actividad, la complejidad de la tarea, la habilidad del conductor y las circunstancias reales del tráfico.

• La utilización del teléfono móvil, ya sea para llamar o para enviar mensajes, es más frecuente en conductores jóvenes y de larga distancia y son más los hombres, en comparación con el número de mujeres, quienes reconocen utilizar el teléfono mientras conducen.

• En muchos países de la Unión Europea la utilización de un dispositivo de manos libres legal es obligatoria. Sin embargo, no existe una prohibición total en ninguno de ellos. Parece existir un cumplimiento medible de estas prohibiciones durante el primer año; sin embargo, pasado este período de tiempo, la frecuencia de la utilización de los teléfonos vuelve a los niveles previos a la prohibición. Estudios detallados realizados en EE. UU. acerca de las reclamaciones a las compañías aseguradoras, demuestran que no se han obtenido beneficios derivados de la prohi-bición de los teléfonos de mano o de la prohibición específica de enviar mensajes de texto.

La disminución del rendimiento del conductor debido a la utilización del teléfono móvil es un hecho comprobado. Sin embargo, a fecha de hoy, no es posible sacar conclusiones acerca del impacto real de su utilización en la seguridad en términos de frecuencia de accidentes. Aunque la utilización de sistemas de información de tráfico esté permitida, la situación podría complicarse si el conductor se distrae mientras tiene lugar un suceso inesperado.


Résumé Le VTI, l’institut suédois des transports et du réseau routier national, a été chargé par le gouverne-ment suédois de compiler les recherches effectuées sur l’usage du téléphone portable et des autres moyens de communication pendant la conduite d’un véhicule. À cette fin, une recherche documen-taire axée sur les années les plus récentes a été effectuée, dont les résultats sont présentés dans un formulaire de questions et réponses.

Les exemples typiques de distraction au volant incluent notamment l’usage d’un téléphone portable, la rêverie éveillée ou la consommation d’un sandwich. Les activités génératrices de distraction peu-vent être divisées en différentes modalités : distraction visuelle (regard hors de la route), cognitive/auditive (esprit hors de la route) et manuelle (mains hors du volant). De nombreuses distractions de la vie réelle impliquent plusieurs modalités à la fois. Par exemple, lorsque quelqu’un vous appelle (signal auditif), vous saisissez le téléphone et répondez à l’appel (distraction visuelle et manuelle) et engagez une conversation (distraction cognitive/auditive). Plus l’étude des effets sur la sécurité de l’usage du téléphone portable au volant progresse, plus la complexité réelle du problème apparaît. Voici un résumé des points principaux de la documentation consultée pour ce rapport :

• La maîtrise de la conduite est compromise lors de l’usage d’un téléphone portable dans les études réalisées en laboratoire de contrôle, sur un simulateur et sur le terrain.• Rien ne prouve que l’utilisation d’un téléphone portable en mode mains libres est moins dangereuse qu’en mode manuel. Néanmoins, un grand nombre de conducteurs pense que les kits de téléphone mains libres sont beaucoup plus sûrs que les téléphones à main.• Il existe un large consensus sur le fait que la maîtrise de la conduite est compromise par les interactions manuelles et visuelles avec les téléphones portables, comme d’engager une con versation ou d’envoyer des messages. Le degré de dangerosité dépend du temps que dure l’interaction distractive, de la complexité de l’interaction, des capacités du conducteur et des conditions en cours de circulation routière. • L’utilisation du téléphone au volant, qu’il s’agisse d’appels ou d’envois de message, est plus fréquente chez les jeunes conducteurs et les conducteurs ayant un fort kilométrage. Plus d’hommes que de femmes admettent qu’ils utilisent leur téléphone en conduisant un véhicule.• La plupart des pays européens requièrent un kit mains libres pour l’usage légal d’un téléphone portable au volant, mais aucun d’entre eux n’impose une interdiction totale. Il semble qu’une conformité mesurable avec les interdictions existe pendant plus ou moins la première année d’application mais, ensuite, la fréquence d’utilisation du téléphone à main revient au niveau antérieur aux interdictions. Les études détaillées des déclarations de sinistre aux États-Unis suggèrent qu’il n’y a pas eu moins de sinistres après une inter diction des téléphones à main au volant, ni d’une interdiction spécifique de l’envoi de messages.

La dégradation de la maîtrise du conducteur due à l’usage du téléphone portable au volant est un fait établi mais, à ce jour, aucune conclusion ne peut être tirée sur l’impact réel sur la sécurité de cet usage en termes de fréquence des accidents. Même si les conditions de circulation sont bien contrôlées, les choses peuvent se dégrader si le conducteur est distrait par un événement imprévu.


2 Introduction The Swedish National Road and Transport Research Institute (VTI) was com-missioned by the Swedish Government to compile relevant research on mobile phone and other communication device usage whilst driving. The compilation of scientific research also includes relevant analyses of compliance with legisla-tion restricting mobile phone use and its effect on road traffic safety from countries primarily in the EU.

We chose to present our findings as answers to a set of questions that demarcate a framework with respect to communication device impact on driving and traffic safety. Some of those questions are derived directly from the government com-mission, while others contribute to explaining necessary background. • In what way are drivers affected when talking or texting?• Does practice improve the ability to drive while handling a mobile phone?• Do hands-free and handheld mobile phones affect driving performance in a different manner?• What are drivers’ attitudes to mobile phone usage when driving?• How do drivers use their mobile communication devices?• Do drivers adapt their driving behaviour to device-usage?• How dangerous is it for drivers to talk or text?• Does a ban on handheld mobile phones work with respect to compliance?• Do bans have a positive effect on road traffic safety?

To each question we provide one short answer, which summarises the key points, and one longer answer, which includes a description of the methods commonly used in the area, references to the literature and a more detailed discussion.


2.1 Literature searchA literature search of the databases Scopus, IRTD, TRIS, Web of Science and VTI’s library catalogue TRAX produced more than 400 research papers on the influence of modern communication devices on driver performance and traffic safety. The keywords used and the years of publication included in the search are presented in Table 1. The TRAX catalogue was searched over a longer publishing period to also include older seminal papers, while the search in the larger databases was restricted to newer material. The obtained body of litera-ture was the starting point for the present report. Additionally, we included seminal research from earlier years and relevant references that turned up by reviewing the literature.

Most papers were concerned with making telephone calls, some focussed on texting, while only very few were concerned with other types of communication.

Table 1. Overview of searched databases together with the keywords used and

the included publication time span. Different spellings and endings of keywords

were included in the search.

DATAbASe

KeyWoRDS

yeAR oF PubLICATIon

TRAX (library catalogue at VTI)

Mobiltelefon, mobile phone, mobile telephone, cell phone, cellphone, cell telephone, cellular phone, cellular telephone, car phone, car telephone, carphone, automobile phone, auto-mobile telephone, wireless telephone, wireless phone, texting, text message, text messaging

1990–now (mid 2011)

ITRD

as TRAX, in combination with safety or driving/driver

1990–now (mid 2011)

TRIS, Web of Science and Scopus

as TRAX, in combination with driver, driving, accident, crash, collision, hazard, danger, risk, safety, safe, behaviour, attention, attentive, alert, alertness, awareness, vigilance, distraction, secondary task, dual task, mental load, workload, stress, reaction, perception, cognition, human factor, hmi, error, law, regulations, directive, legislation, legal, ban, banning, rule, specification, recommendation, traffic act, prohibition

TRIS: 2000–nowWoS: 2003–nowScopus: 2005–now


2.2 Research methods

In the reviewed research numerous research methods were employed in order to answer different types of research questions. In Annex II the main research methods are described briefly, together with a short comment on how the resulting data can be used. We would like to stress that a basic understanding of the potential and limitations of different methodological approaches is paramount for making educated and critical interpretations of the available data.

There is not one method that can answer all research questions on the topic. The strongest answers are obtained when the results from different methodologi-cal approaches provide a consistent picture of the problem at hand. A combination of methods is necessary to provide detailed information about different aspects of the problem. When the goal is to increase road safety, the ultimate goal is to reduce the number of fatalities, injuries and also property damage. However, looking at crash statistics alone is not always the best solution to assess the effect of certain safety measures, partly because crashes are rare, which is why it takes a long time to accumulate sufficient data for statistical analyses, and partly because many other factors may change during the data acquisition period, which is why it is difficult to attribute possible changes in crash statistics to the safety measure and nothing else. Therefore, many studies use “surrogate” measures, that is, performance indicators that are assumed to be related to safe and unsafe driving, and therefore ultimately to crashes. These performance indicators can stem from measured driving behaviour such as speed choice or the variation in lateral position, they can come from observations such as the frequency of overtaking manoeuvres or telephone usage, and they can be deducted from interviews and questionnaires such as opinions and acceptance, and so on.


2.3 Driver distraction and inattention There are many different definitions of driver distraction, but most of them have in common that the driver’s attention is diverted away from what is critical for safe driving [for reviews of the literature see e. g. 62, 98, 97]. Opinions differ as to whether driving related behaviour, for example looking into the mirror instead of straight ahead, or the other way round, should be subsumed under distraction. Similarly, being “lost in thought” or daydreaming as well as sleepiness is included in the inattention concept by some but not all researchers. Some researchers consider distraction and inattention to be synonymous, while others have other opinions. There is a debate whether a driver is inattentive when the amount of attention paid is not sufficient to drive safely, or when he or she engages in anything not related to driving. Each definition has its own drawbacks, and the fact that many different definitions exist and are used in parallel makes comparisons between different studies difficult.

One major problem with most definitions is the fact that “safe driving” is included as a benchmark, as this inva-riably introduces a hindsight bias. In hindsight we know where the unexpected event occurred, but in order to use the concept of inattention in a meaningful way we should be able to tell beforehand which amount of attention is needed where.

Here we review papers which use many different defi-nitions, some of them explicitly stated, others just used implicitly; therefore we cannot provide one common defi-nition of driver distraction for the whole report. Where necessary we provide the definition used in the individual study, otherwise we report the actual facts. For further reading on the latest attempts to define and classify driver inattention and distraction we refer to Regan, Lee and Young [97], Regan, Hallet and Gordon [98] and Hurts, Angell and Perez [56].

We would like to make a few important statements about distraction before we proceed:

• First and foremost, distraction is very much an integral part of being human – without the innate capa- bility of quickly redirecting one’s attention to

important stimuli we would not have survived as humans.• Some things, like loud noises or a flashing light in the periphery, “grab” our attention in a physiological way we cannot avoid, while


in other matters, like picking up a CD that dropped to the floor, we have a choice.• Distraction, in contrast to intoxication or sleepiness, can be fleeting and transient – an attentive driver may be distracted in the next second, and attentive again shortly after, while these changes do not occur as fast for either intoxication or sleepiness.• During a typical trip there are many periods during which drivers have spare attentional capacity, depending on the traffic situation, their current capabilities and other factors.• Distraction can be difficult to observe and especially to reconstruct, and self-reports of distraction are often inaccurate by their very nature, as distraction periods without a surprising outcome often go unnoticed even by the one distracted.

2.4 ScopeIn this report we focus on distraction caused by mobile telephones and other communication devices. All other secondary task engagement, like eating, grooming, attending to children, reading, the use of navigation devices, day-dreaming, smoking, putting on makeup etc., is not considered and only mentioned in passing, in case that it was integral to the reviewed study. We make one exception to this rule for communication with passengers, as it is often set in comparison with mobile telephone calls. We urge the reader, however, to keep in mind that communication devices only represent one small portion of the stimuli that can catch the driver’s attention.

We excluded studies on distraction mitigation systems, even though they could also be applied to in-vehicle communication systems. However, we briefly present two studies concerned with workload managers.

2.5 Types of communicationA plethora of different communication functions can be carried out with the help of SmartPhones and similar devices. Communication is omnipresent in our lives and also a big issue while driving a vehicle. In Annex I an overview of different functions and their coupling to different sensory modalities relevant to driving is presented. All functions have a cognitive component, as talking, reading and writing are inherently cognitive.


3 Frame of ReferenceTo drive a vehicle is a complex and dynamic task. In order to maintain a safe traffic system a number of requirements are put on road planners, lawmakers, vehicle manufacturers, road maintenance enterprises and, of course, the driver. Often the focus is on the driver, being at the “sharp end” of things. Whether the drivers manage to interact with each other and/or their surroundings will determine whether things go well or not. The driver continuously needs to take in and process information and is often required to be proactive instead of reactive. Redundancies are built into the traffic system, and drivers may make up for each other’s mistakes, but it is necessary that the attention resources that are necessary to resolve the current situation are available.

The relationship between driver behaviour, telephone usage, crashes and the potential effects of a law on telephone usage is complex, as indicated in the schematic overview presented in Figure 1. The arrows in the figure represent a direct causal influence.

The driver’s capabilities and limitations, both physical and resulting from the gained experience, determine how well a driver can perform in relation to what is required for safe driving, with and without concurrent telephone use. The driver’s capabilities and limitations are also likely to influence his or her attitude and opinion on driving in general, and on telephone use while driving in particular. The driver’s attitude/opinion influences how, when, and in which situations a driver is willing to use a telephone, while the driver’s performance will determine how well it actually goes when the driver has decided to use the telephone. At the same time, the own and possibly other’s device usage will influence a driver’s attitude towards telephone use. If and how the telephone is used is the factor that determines the resulting crashes. A driver might for example exhibit very bad performance while on the phone, but be aware of this limitation, which leads him or her never to use the phone while driving, there-fore no crashes will result from phone use for this particular driver. Another


driver may be much better in handling both the driving situation and concurrent telephone use, but overestimate his or her capabilities, resulting in telephone use in too complex situations, which then can lead to crashes. Own and other’s crashes that are related to telephone use are likely to influence a driver’s opinion on telephone usage in general.

One way to influence this relationship is to pass a law on telephone usage. This law comes about due to crashes that are associated with telephone usage, and due to observed actual usage. The public opinion may also influence the law. The law itself, probably in conjunction with enforcement, will influence the driver’s opinion on telephone usage, which will influence usage in turn, which can then have an impact on the crash rate.

Regulation by law is, however, not the only possibility to change the picture. New technology will for example influence both driver performance and the driver’s attitude/opinion, and education can influence the driver’s attitude/opinion as well as improve the driver’s capabilities to handle certain situations. Additional measures are concernable.

Figure 1. Schematic overview of the complex relationship between different

aspects of the traffic system in relation to a law concerning communication

device usage. Observe that a law cannot influence the frequency of crashes in

a direct, but only in an indirect manner.

Capabilities and limitations(experience, visual activity, ...)

Capabilities and limitations(experience, visual activity, ...)

Driver performance

Crashes

Actual usage Law

Attitude/opinion


4 Main FindingsIn this section the questions are presented together with concise, summarising answers, followed by longer answers. Those include both a section on methods that are customarily employed to investigate the question at hand, and a more detailed overview of the literature on the topic, together with references to the actual studies.

4.1 In what way are drivers affected when talking or texting?

MeTHoD

The effect of driver distraction is usually measured in a controlled experimen-tal setting. The idea is to isolate the distracting event as much as possible so that its effect can be measured without confounding factors. The high level of control comes with a cost; the drivers are not allowed to choose when and where to engage in the distracting activity, e. g. the phone call, and they are very much aware of being monitored and may thus not behave or react as they normally would [74]. If the experiment is run in a lab or in a driving simulator, it is also necessary to consider the fact that there is no actual risk involved. In other words, the experiments allow us to compare how text messaging or phone conversations affect your driving when all other factors are kept con-stant, but since we don’t know if, when and where you would engage in the conversation it is not possible to use the results to predict real-world crashes. There are naturalistic driving studies that offer great realism, but since such studies cannot be controlled, we can never be sure if the measured consequence of a telephone conversation is due to the conversation itself or some external factor such as weather, surrounding traffic and the mood of the driver.

Long AnSWeR

The research community agrees that mobile phone conversations increase reaction time. This holds true for handheld and hands-free, in simulators and in the field, and for information processing tasks and normal conversation [e. g. 13, 53]. Many drivers consequently have to brake harder to avoid a safety critical event [44]. They may compensate for their impairment by increasing their safe-ty margins in terms of decreased speed [117, 129] and increased distance to the vehicle in front of them [108, 109, 107, 129]. However, there are also results indicating that no such compensation occurs [3, 100], and behavioural changes can be interpreted as loss of control instead of as compensation [e. g. 6].

[Mobile phone conversation increases reaction times and missed events. In addition, texting and dialling require the driver to look at the device and away from the road leading to reduced vehicle control and an increase in missed events.

[

SHoRT AnSWeR


The visual behaviour changes during mobile phone conversations and other cognitive tasks such that drivers look more straight ahead and neglect scan-ning the periphery, including the mirrors and the speedometer [4, 46, 77, 88, 99, 125]. This is reflected in a reduction of the variation in gaze direction [94, 125]. Other visual functions requiring eccentric fixations, such as glances to speedometer and to interior mirrors [46, 94] and even safety critical visual search for traffic at street crossings [41] and anticipatory glances when entering a curve [69] have been shown to decrease or even get left out under cognitive load. Further results show that it is more difficult to remember and interpret objects that have been looked at during a mobile phone conversation [6, 88, 110, 107, 122]. This is commonly referred to as the “looked-but-failed-to-see” phenomenon.

The visual behaviour during texting, dialling and other tasks that require the driver to look at the device is characterised by more frequent glances away from the road, sometimes of very long duration [54, 87, 102]. Looking away from the road leads to increased reaction times [25], missed events [54, 68] and decreased control of the vehicle [21, 25, 54].

The extent of how much the driver is affected by making calls or texting varies with a number of factors such as age, experience, driving conditions and how emotionally engaging the conversation is. Older drivers’ performance is impai-red by irregular use of mobile phones or by deterioration of cognitive abilities whereas novice drivers are more likely to take risks and be unaware of danger [17]. Experienced drivers commit fewer driving infractions [61] and are better able to detect hazards [68], but no matter the experience level, hazard detection is delayed [106, 108] and driving performance is worsened [61]. Conversational complexity and emotional content of the conversation has been found to affect


reaction times, the amount of driving errors and visual scanning behaviour [10, 26, 89]. Further research is necessary in order to understand how these factors, and additional factors such as sleepiness and intoxication, interact with driver distraction.

4.2 Does practice improve the ability to drive while handling a mobile phone?

MeTHoD

Whether practice improves actual dual task performance has to be investigated under controlled circumstances, like in a simulator or controlled field studies, such that performance in comparable situations can be assessed. Practice can either be operationalised by a number of training runs, or, probably more exter-nally valid, by controlling for people’s experience with telephone use and also with driving. In the latter case, however, one has to be aware that proficient and non-proficient users might also differ systematically with respect to other characteristics that might influence performance.

Long AnSWeR

Generally the literature on learning effects of telephone usage and driving is limited. Being well practised as a driver will help release attention resources and being familiar with the manual handling of your mobile phone will also require fewer attentional resources than being unfamiliar with the phone [89]. This means that more capacity becomes available for dual or multi-tasking, for example using a mobile phone. This does not necessarily mean that it is safe to do so, because less focus on the surrounding traffic will entail an increased risk of missing safety critical events if they should occur.

Driving simulator studies have suggested that practice can improve the concur-rent execution of a telephone task and driving [105] and also that practice of iPod use in combination with driving show improvements over time [16]. It has however been argued that these experiments probably measure that the drivers become better at driving a driving simulator rather than becoming better at concurrently handling a secondary task [18]. Using a more stringent experi-

[The small body of available data indicates that practice of simultaneous driving and telep-hone use has at best a very limited effect. There seems to be some agreement, however, that more experienced drivers have better strategies to handle telephone usage.

[

SHoRT AnSWeR


[ Mobile phone conversations during driving have a negative impact on driving performance regardless of whether a handheld or a hands-free phone is used.

[

mental setup where driving simulator practice was controlled for, Cooper and Strayer [18] found that participants with a greater real world experience of telephone usage while driving did not perform better than those participants who were not used to mobile phone use while driving. These findings are in line with an early on-road study, which did not find any dual-task learning for measures related to vehicle control either [11].

For future research it appears to be at least as important to investigate whether drivers learn to choose opportune situations for placing their calls as it is to investigate whether actual performance during a call can be improved. This would presuppose that experience in driving rather than in handling a telephone can make a difference in overall performance.

4.3 Do hands-free and handheld mobile phones affect driving performance in a different manner?

MeTHoD

In order to investigate whether handheld and hands-free phone usage influences driver behaviour differently, controlled studies have to be used. When the ques-tion is whether the two phone types actually are used in different ways in traffic, observational studies are a suitable method (see also Section 4.5).

Long AnSWeR

In Section 4.1 we concluded that phone conversations lead to longer reaction times. This is the case regardless of phone type. In fact, there are no significant differences in reaction times when comparing hands-free with handheld phone conversations [12, 57, 89, 109, 117]. Also when investigating mobile phone records of individuals involved in crashes resulting in property damage or hos-pitalisation, no safety advantages have been found when using hands-free phones as compared to handheld phones [75, 95]. Some studies suggest that drivers adjust their speed and headway when using a handheld phone, but not when using a hands-free phone [3, 12, 88, 117]. This may indicate that drivers compensate for the deteriorating effects of mobile phone use when using a

SHoRT AnSWeR


handheld phone but neglect to do so when using a hands-free phone [13, 57], possibly because a handheld phone serves as a reminder of being engaged in something else than driving.

It may be argued that talking to a passenger is as distracting as talking over the phone. However, a passenger can provide an additional pair of eyes if needed, and a passenger also tends to slow the rate of the conversation in difficult driving conditions [15, 24, 72]. This behaviour of conversation modulation has not been found during mobile phone conversations.

Whether handheld and hands-free telephones are used in a different manner is dealt with in Section 4.5.

4.4 What are drivers’ attitudes to mobile phone usage when driving?

MeTHoD

Drivers’ attitudes can only be accessed by asking them directly. Self-reports of driver behaviour are relatively reliable, but special attention has to be paid to the instructions and the procedure of how the self-reports are collected. Anonymous responses and settings in which individuals cannot be singled out, instructions stressing the importance of honest answers, and the use of lie scales for controlling socially desirable response tendencies statistically are all

[

[Drivers are generally aware of the risks involved with texting and talking while driving. However, perceived social norms, attitude, perception of control and believed importance of the call are more indicative of phone use than perceived risk.

SHoRT AnSWeR


effective techniques for improving the reliability of self-reports (Lajunen & Summala, 2003).

Long AnSWeR

Drivers generally believe that texting or talking on a mobile phone while driving is dangerous. In the EU, 76 per cent thought that driving while talking on a handheld phone was

a major safety problem, while only 26 per cent said so about driving while talking on a hands-free phone, 22 per cent of all respondents did not consider driving while talking on a hands-free phone to be a problem at all [32]. The discrepancy between attitudes towards perceived risk of handheld and hands-free mobile phone use [33, 124, 132], despite comparable decrements in driving performance, is clearly an erroneous deduction by the public when one system is banned on safety grounds but the other is not.

There are a number of explanations for drivers’ propensity to use their mobile phone while driving. One reason is that there is a time benefit with talking on the phone while driving [123]. Another reason is that most drivers believe that they are more capable of handling nomadic devices while driving than other drivers [58, 121, 127]. In general, people believe that texting or talking on a mobile phone while driving is dangerous, but not all are aware of their decreased performance [71]. Even if they were, the perception of risk is a weak predictor of behaviour [70]. Instead it is the driver’s attitude to telephone usage [81, 101], their perceived control over the situation [101], normative beliefs [123] and the perceived importance of the call [14, 80] that determine whether a driver is willing to initiate a text message or a phone call.

A survey in the US established that about half of the respondents objected to a law that would ban all mobile phone use for all drivers, while the remaining half supported such a law [2]. Around 80 per cent of all interviewees supported a texting ban, which can be contrasted to 24 per cent of the respondents admitting having sent a text or e-mailed while driving in the last month. In general, drivers who have engaged in a certain activity view this activity as being less risky [124, 127]. On the other hand, drivers who had been involved in incidents where telephone use was a contributing factor were more likely to be in favour of a ban [127]. Personal history obviously affects people’s opinion.

A Swedish report from 2004 summarises a national questionnaire-based survey, a small literature search and two focus groups on mobile phone usage and driver attitudes [115]. Already in 2004 especially younger drivers reported text messaging while driving, and frequent handheld telephone use. Drivers reported behavioural


adaptation like slowing down, choosing suitable traffic situations or keeping calls short when on the telephone, but also errors especially on the tactical level, for example missing exits, getting lost or missing traffic signal changes. Results from the questionnaire showed that drivers who used their telephones extensively were also more likely to drive without seat belt. There are indica-tions that telephone use while driving can be part of a general attitude towards risk taking and rule violation [5, 27, 118], which would mean that a certain sub-group of drivers is more prone to exhibiting risky behaviours.


MeTHoD

Three different types of methods are employed to study the prevalence of com-munication device usage in normal traffic. One is direct observation on site, where trained observers basically watch the passing traffic and note whether each passing driver uses a telephone or not. Usually the observers work during daytime and close to a junction where speeds are lower, which facilitates encod- ing. Handheld usage is much easier to identify than hands-free usage, which makes the data generalisable for handheld device usage prevalence at daytime in a given location.

Another method is to ask drivers about their device usage, e. g. via question-naires or telephone polls. As with all other studies, it is important to decide beforehand who to study, which determines to which population the results can be generalised. When asking about sensitive issues like telephone usage while driving it is important to keep a possible “social desirability bias” in mind, that is, that drivers answer as they think they are expected to as good citizens, while they actually do something else. Even if drivers actually want to answer honestly it can be difficult to remember exactly how much one uses the tele-phone, which can make the data less reliable than observed behaviour.

A third method type, called naturalistic driving studies, has emerged over the recent years, in which private vehicles are instrumented and used as usual by their owners over a longer time period, e. g. a year. Data can be logged either continuously or event based, which will give more or less restricted access to what drivers actually do in real traffic. Samples from those data allow con-

[[Around 2–6 per cent of all drivers use mobile phones at any given moment. Most of them

use handheld devices. younger males are most likely to both call and text. Frequent drivers and drivers alone in their car are more likely to use their telephone while driving.

SHoRT AnSWeR


clusions about actual communication device usage patterns in real driving. As not all data can be reviewed due to time and budget limitations, it is very important to be aware of how the studied data snippets are selected.

Long AnSWeR

With the extremely fast market penetration of mobile communication devices and the quickly changing spectrum of the usage of those devices, studies on usage while driving may quickly become outdated.

Numerous polls and observational studies have been conducted on telephone usage while driving. In most studies only handheld phone use is reported, because it is more readily observable. The majority of studies stem from the United States and Canada, even though some European data can be found. In addition, some naturalistic driving data on telephone usage exist.

Prevalence values indicate which percentage of drivers talks on a telephone at a given point in time. In the US, observational studies are made annually in intersections during daytime, which provides good estimates for the develop-ment over time. These studies show that the prevalence initially increased but has stabilised at about 6 per cent in the recent years [36–38, 83, 90, 104]. The data were obtained from 50 000 vehicles in 1 500 randomly selected intersections through-out the US. Similar results have also been found during night-time [126].

While most observation studies occur at intersections, as lower speeds allow for an easier identification of telephone users, there is one notable exception from 2001, which reanalysed around 40 000 high-quality digital photographs of drivers on the New Jersey Turnpike [60]. The results can be compared to the national survey conducted in 2002 [36], where mobile phone usage was recorded


during daytime in intersections in lower-speed areas. Mobile phone use was 1.5 per cent on the New Jersey Turnpike as compared to 3–4 per cent in lower-speed areas. Mobile phone use on the New Jersey Turnpike was less frequent at night and on weekends. Mobile phone usage was positively associated with lower speeds and drivers that did not carry passengers. An observational study from Australia shows a prevalence of around 2 per cent for handheld phone use [114].

When using data from naturalistic studies for prevalence estimates of telep-hone usage it should be appropriate to look at random samples. These do not really exist, however, as baseline samples in many cases are selected such that they match cases in which safety critical events were recorded. An older study, which approximated the desired sampling method, found that around 30 per cent of drivers used their mobile phones during the selected trip samples, and they did so for 1.5 per cent of the driving time [111]. Estimates from the newer studies indicate a mobile phone usage during around 4–9 per cent of the driving time, including both handheld and hands-free telephones [48, 63, 86].

Hands-free sets are less common than handheld telephones with more than half of the drivers using solely or mostly handheld phones, whereas 13–37 per cent mostly used hands-free sets [2, 8]. In two observational studies 0.5–1 per cent of the drivers were seen using hands-free devices at any given time [83].

Surveys, naturalistic studies, and to a certain extent observational studies indicate that phone use differs across driver type:

• Some drivers use their phones much more frequently than others, but in recent years the number of drivers who report to use their phone fairly often/regularly/during the last week had risen to around 35 per cent from 20 per cent in 2001 [2, 120]. Out of 1100 Danish truck drivers more than 99 per cent reported in a questionnaire that they used their mobile telephones while driving [116].• Both surveys and observational studies agree that younger people are more likely to be on the phone while driving than older people


are [e. g. 5, 8, 36–38, 58, 67, 90, 92, 112, 116, 120, 126]. • Polls repeatedly find that male drivers indicate more telephone use while driving than female drivers do [1, 2, 5, 8, 120], while in observational studies females are consistently overrepresented in telephone usage [e. g. 83, 104, 126]. One study from Australia reports a higher frequency of phone usage for males in 2006, while no difference was found in 2002 [114].• Results show that higher mileage drivers also are more frequent phone users [58, 116]. Given that men still drive more than women

do, this would corroborate the male overrepresentation of telephone use found in the surveys. No data were found on the issue, but it might be possible that males are more likely to use hands-free kits, which might help explaining the higher handheld phone usage by females.• Time of day was also found to be associated with phone use, with lowest usage rates in the early morning and an almost monotonous increase over the day, peaking at around 17.00 [1]. Telephone usage was also found to be higher during weekdays than during weekends [104].• The presence of passengers was found to influence the frequency of mobile phone use. When no passenger was present drivers were three times more likely to use a phone than drivers with two or more passengers [1].

Data on prevalence for texting, e-mailing and the like are more fragmentary. Generally, the prevalence for texting seems to be lower than for phone calls. Visible device manipulation lay between 0.6–1.0 per cent in observational studies [83], while a questionnaire revealed that 30 per cent of the responding 320 teenagers with a driving licence had texted during their most recent trip [85]. It was more likely that text messages were sent in reply than in order to initiate a text conversation. A telephone survey in the US in 2009 found that 13 per cent of all 1 219 responding drivers reported to text, 6 per cent reported to e-mail and 4 per cent reported other internet usage while driving [8].

In naturalistic studies not enough texting has been observed to allow reliable prevalence estimates. Those two studies for which texting was reported at all were made on professional truck drivers [48, 86], which means that caution should be applied when generalising to all drivers. In one study texting was detected in 3 out of 205 582 baselines, and in 90 out of 37 708 safety critical events, including crashes and near-crashes [48]. The other study reported text-ing to have occurred in 31 out of 4 452 safety critical events, while the activity was logged 6 times in 19 888 baseline instances [86]. In the former study each


of the reviewed video clips lasted for twelve seconds, while in the latter study the clips were six seconds long. Naturally, clip duration influences the likelihood to detect a certain behaviour within a specific clip.

4.6 Do drivers adapt their driving behaviour to device-usage?

MeTHoD

To investigate whether drivers adapt when and where to use their telephone usage in real traffic, observational studies of normal driving are very important. With instrumented vehicles it is possible to follow one driver over time and in varying traffic environments, while road-side observations provide a picture for one location, but with a large number of drivers. Questionnaires and interviews allow conclusions about drivers’ attitudes. In order to assess adaptation strate-gies on a tactical level, given that a phone is used, more controlled studies are called for, as it is important to investigate driving in the same situation with and without concurrent phone use. These studies typically take place in a dri-ving simulator, on a test track or in controlled experiments in real traffic.

Long AnSWeR

Adaptation to telephone usage can occur on many different levels. While they are all touched upon within other sections in this text, a short summary of the different possibilities to adapt is provided here.

Drivers can choose if, when and where to use a phone while driving. They can even refrain from driving or choose a different route when they know they have to make a call. The latter aspect has not been investigated as far as we know, but it has been shown that drivers tend to choose less demanding situations to make calls [e. g. 31], that professional drivers are better at this than drivers of private vehicles [48, 64, 86], and that drivers may refrain more from using their phone when high visibility enforcement is in place [20], or alternatively switch to hands-free [8] or use the phone in a more discreet manner [50, 51].

To which extent drivers adapt their phone usage to the situation is also depen-dent on the driver type. As discussed in Section 4.4 attitudes towards phone use differ between drivers, which has an impact on how phones are used.

[Drivers choose when and where to use their telephones, and also during a call there are indications for compensatory driving behaviour. How successful the adaptation strategies are depends largely on the driver’s capabilities and motivations, but also on the predictabi-lity of the situation.

[

SHoRT AnSWeR


For example, an observational study found an association between being on the telephone and not coming to a complete stop at a stop sign [52]. While it is tempting to interpret the result such that drivers are distracted and conse-quently miss the sign, it is also possible to assume that drivers who generally take higher risks are both more likely to talk on the phone and not to stop at stop signs.

Furthermore, drivers can adapt their behaviour during an actual phone call. In one questionnaire study about half of the interviewed teenage drivers reported trying to keep conversations short, and to wait until it feels safe to read or write a text message [85]. Many more studies were devoted to how drivers adapt their driving behaviour while on the phone. Discussions are ongoing whether behavioural changes are conscious adaptations or indications of a loss of control or both. How driver behaviour is affected by concurrent phone use is discussed in Section 4.1.

Typical for more controlled studies is that the drivers’ tactical behavioural span is limited by the instructions given for the experiment. Drivers might be required to follow a certain vehicle, not to overtake, or similar. This is done to keep the setting alike between different participants, and such that variables like speed and lateral position are not influenced. However, there are a few exceptions of studies where drivers had more freedom of choice on the tactical level of driving behaviour. In one simulator study on hands-free phones drivers were allowed to drive as they wished on a multilane roadway, and they could change lanes at will [19]. No increase in following distance was found, but drivers changed lanes less frequently in the medium- and high-flow conditions when on the telephone, and those lane changes that occurred were less safe in terms of distance to the vehicle approaching from behind. These findings indicate also that traffic flow may be hampered in high-flow conditions when a substantial number of drivers is on the phone at the same time.

Workload managers might support the driver in the choice of when to use or not to use the telephone. This might be especially helpful for incoming calls and texts. One simulator study [128] and one field study [91] showed promising results, indicating that driving performance improved while controlled commu-nication could be maintained.

Last but not least, not only the driver can adapt to the situation. It is a com-mon misconception that a conversation on the telephone is equivalent to talking with an adult, sober and traffic experienced, passenger. It has been shown in several studies, however, that passengers with traffic knowledge adapt their conversation patterns to the traffic situation at hand, and therefore can help in regulating the driver’s workload, while this is usually not the case for telephone calls [e. g. 15, includes further references to the topic, 24, 72, 76]. Therefore, providing means that allow an adaptation on the side of the conversation part-ner might be beneficial.


4.7 How dangerous is it for drivers to talk or text?

MeTHoD

The method that has traditionally been used to answer this type of question is the so-called epidemiological study, which uses either observational data or data collected from sources like hospitals, insurance companies, telephone companies. Crash database research has been used to obtain percentages of distraction related crashes. More recently, naturalistic driving studies have also been used as basis for crash estimates. Still, crash numbers are small even in the largest naturalistic studies, which makes estimates unsure and susceptible to confounding factors. Common to all these study types is that they are based on observations without active experimental manipulation. This means that it is not possible to draw causal conclusions from this type of data. Even though it can be shown that variables are related to each other, it cannot be deducted that one is the cause for the other.

Both for epidemiological studies and naturalistic driving studies odds ratios are a common way of reporting results. Especially for a higher prevalence of the outcome, odds ratios do not approximate relative risk very well, which is a potential source for misinterpretations, as we humans are used to thinking in terms of relative risk. Odds ratios always overstate the effect size in comparison to relative risk. For further information on odds ratios the reader is referred to Davies, Crombie and Tavakoli [22]. It has to be noted, though, that an odds ratio of 1 means that the two compared situations do not differ from each other.

The two approaches that are used most widely are case-control and case-cross-over. In both cases, the frequency of for example mobile phone use is compared for events (e. g. crashes) and non-events (e. g. normal driving). In the case-control approach, the data from the events and non-events stem from different drivers, while in the case-crossover approach the data are taken from the same drivers.

[It depends on the situational context and the driver’s capabilities. Recent naturalistic re-search on commercial truck drivers indicates that the occurrence of crashes, near-crashes and safety critical events is correlated with handling the telephone (dialling, texting), but not with just talking on a phone. This does not mean, however, that talking on the phone while driving is safe in all situations and for all drivers. It is more likely to be at least in part a sign that drivers are capable of to self-regulating their telephoning behaviour.

SHoRT AnSWeR ]


Long AnSWeR

Crash database studiesCrash database studies have been used in different ways to analyse the contri-bution of mobile phone usage to crashes. One way is to get a feeling for the overall prevalence of distraction/telephone usage in crashes. In 2008 16 per cent of all traffic fatalities were attributed to distraction of some kind [82]. This number was based on what could be found in police reports, where the use of mobile phones, navigation systems and personal digital assistants as well as other types of distraction were subsumed under “driver related factors”. However, other crash database research estimates 22 per cent of all crashes to be related to mobile phones [30].

Another possibility is to compare crashes in which mobile phone usage occurred with other crashes for which no mobile phone usage is registered for differences. Such a comparison based on data from North Carolina (1996–2000) found that mobile phone related crashes were more unlikely to lead to serious or fatal injuries, that they were twice as likely to be rear-end crashes, and that they were more likely to occur during midday/afternoon than other crashes. Furthermore, drivers involved in mobile phone related crashes were more likely to drive SUVs, to have committed a violation, and to have driven straight on [55]. A slightly more recent study from Missouri confirmed that distraction-related crashes in general tended to be rear-end, but that mobile phone related crashes were more likely to be angular crashes, while other electronic device use had a higher association with single vehicle crashes [34].

There are some issues with crash databases. Firstly, not all crashes are reported with certain crashes being systematically underreported. Secondly, there may be a reporting bias with respect to mobile phone usage. Third, if the behaviour under normal driving conditions is not known, then there is no way to set crash data in relation to normal driving.

epidemiological studiesSo-called epidemiological studies try to amend this latter concern by comparing the prevalence of mobile phone use in crashes with prevalence in comparable situations during normal driving (baseline). Two widely cited studies, both with case-crossover design, combined telephone records obtained from phone companies in Australia with crash data and comparable driving periods in the days before the crash. In one study crashes with property damage only from an area without telephone ban were investigated [95], while the other study was concerned with injury crashes followed by hospital attendance in an area where handheld phone use was banned [75]. Both studies arrived at an estimated odds ratio of 4 for phone use in crash events. In a later comment this was considered to be on the conservative side, meaning that the actual odds ratio could be higher [96]. Even though cited widely, the studies received a fair amount of criticism for how the baseline periods were selected [40, 73, 131]. A different


selection of baseline periods resulted in odds ratios that did no longer show a significant increase of crash likelihood for mobile phone use [131].

Results obtained when analysing air bag deployment crashes in comparison to baseline driving with respect to phone use employing the GM promoted hands-free OnStar system [130] showed that there was no association between phone use and crashes. Critics have identified a number of confounding variables [9], however, and the results are still debated [see letters to the editor in the journal Accident Analysis and Prevention, 2009, Vol 29, and 131].

Instead of analysing actual crashes and corresponding baselines Bhargava and Pathania put crash likelihood in relation to an increase in phone calls placed in the US, which is observed on weekdays immediately after the lower off-peak rates take effect [7, updated in 2010]. No corresponding relative rise in crashes during the critical time periods could be found in their study.

Laberge-Nadeau et al. [66] tried yet another method to arrive at estimates of telephone involvement in crashes by sending out 175 000 questionnaires to Canadian driving licence holders. The response rate was around 20 per cent. Questions concerned driving habits, opinions about distracting activities and crashes within the last 24 months. The results showed that mobile phone users had a 38 per cent higher relative crash risk both for injury collisions and for collisions of all types. However, the relative risk sank to the level of non-users when mileage and driving habits were taken into account. This indicates, just as was found in other studies, that frequent drivers also are more frequent mobile phone users.

This short overview of influential studies already shows that odds ratio esti-mates vary widely between different studies, and even within one and the same study if different factors are taken into account. Elvik, who reviewed twelve epidemiological studies, found odds ratios to vary between 1 (no difference to baseline) and slightly above 4 [28]. Elvik explains this big variation partly by the difficulty to obtain good exposure data for phone usage in baseline driving. The different methods used to assess baseline prevalence vary widely in quality, leading to very divergent odds ratio estimates. Therefore, clearly, better base-line exposure data are needed, but correct exposure values for events are just as necessary.

naturalistic driving studiesIt should be expected that naturalistic driving studies can fill this gap. Typically, in those studies the drivers’ behaviour is logged continuously over a long period of time. The first study of this kind was the 100-car study con-ducted by Virginia Tech Transportation Institute (VTTI), where 100 passenger vehicles were instrumented with log equipment and cameras and then used for day-to-day driving [23, 79]. Similar studies with professional truck drivers followed [45, 86]. Further, data logged by the driver monitoring system DriveCam were analysed [47–49]. In the latter case data logging was event-based, that is, only clips of twelve seconds duration that were triggered by fast accelerations/dece-


lerations were saved. Some of those triggers identified actual safety critical events, while others were “false alarms”, which were subsequently treated as baseline clips.

For the passenger car study it was found that a large percentage of the safety critical events stemmed from a small percentage of drivers [23]. For data analysis phone usage was subdivided into “manipulating the phone” and “listening/talking on the phone”. When the passenger car data were analysed with respect to crashes/near-crashes with a case-control approach, dialling a handheld device obtained an odds ratio of 2.8, while talking/listening on a handheld device had an odds ratio of 1.3, which did not differ significantly from baseline driving [63]. In a re-analysis employing a case-crossover design dialling obtained an odds ratio of 2.1, while the odds ratio for listening/talking again was 1.3, this time significantly different from baseline [64].

For the commercial truck studies analyses were made with respect to “safety critical events” which comprise crashes, near-crashes, crash-related conflicts and unintentional lane deviations. The odds ratios for manually manipulating a mobile device lay between 3 and 4, while listening/talking did not have an increased odds ratio – for hands-free talking/listening the odds ratio was even found to be significantly below baseline [48, 86].


Only very limited data material was obtained with respect to texting. In the somewhat older passenger car study no texting was observed at all, while the commercial truck studies produced 37 texting events in 4 452 safety critical events and 19 888 baselines [86] respectively 93 texting events in 37 708 safety critical events and 205 582 baselines [48]. Even though the odds ratios found for texting based on those data (23.2 in the study by Olson [86] and 163 in the study by Hickman [48]) are impressive and widely reported in the media, they should be taken with a grain of salt, as the values are not very reliable due to their small numbers. This is also acknowledged by the authors themselves. It should also be kept in mind that data from commercial truck drivers do not necessarily generalise to the driving population at large.

Information on phone use is most likely more precise in naturalistic driving studies than in epidemiological studies, as it is coded from video observation. Additionally, eye glance behaviour can be coded based on video analysis – and in future studies most likely based on eye tracking data. This opens up new possibilities for analyses, as gaze behaviour can be set in relation to safety criti-cal events. Data from all reported naturalistic studies indicate that secondary tasks that entail more glancing away from the road in combination with manual interaction have higher odds ratios than tasks with lower visual and manual demands. Comparable analyses for cognitive load cannot be conducted, as there is presently no way to assess cognitive load by the data logged in naturalistic studies.

While it is tempting to interpret the data such that talking/listening on a phone while driving is not dangerous or even beneficial, this conclusion is not justi-fied. In a naturalistic setting drivers often choose when to engage in a secon-dary task, and it is very likely that they choose at least somewhat wisely. This would mean that they are more likely to use their phone in less demanding situations, which was confirmed by a follow-up to the Hickman-study [31]. Texting and dialling were rarely observed in more complex driving situations, and safety-critical events were few in general when task demands were low. Talking/listening on the phone was observed on all driving demand levels, and in none of them odds ratios indicated an increase in crash risk associated with this activity. There are several possible explanations for these results, most of which touch upon the human’s general ability for self-regulation. Drivers might either compensate or over-compensate for their actual device usage, or


they were just not overtaxed in the present situation. It is also possible that the researchers’ classification of situational complexity does not match with what is actually experienced by drivers.

Naturalistic and epidemiological studies do not allow conclusions about causal relationships between telephone usage and safety critical events, based on their methodological premise. It also becomes clear that a correct and unbiased base-line selection is paramount in order to obtain valid results. What these studies show is that drivers apparently are quite capable in dealing with the detrimen-tal effects that usage of both handheld and hands-free devices has on driving.

4.8 Does a ban on handheld mobile phones work with respect to compliance?

MeTHoD

Law compliance is mainly investigated via surveys and to a certain extent via observational studies. While the social desirability bias can be an issue with surveys, this method gives access to people’s opinions and attitudes. Due to practicality reasons observations are often limited to daytime and locations with low speeds, thus, they are most useful for reporting relative changes in phone use.

Long AnSWeR

The legislative or regulatory situation in the European Union is summarised in Table 2, which draws on the work done in an EC tendered study [59]. None of the twenty-seven EU countries have a total ban on mobile phone use whilst driving. Sweden is the only EU country without specific mobile phone laws requi-ring hands-free or other such equipment. The Swedish Road Traffic Ordinance does, however, require the driver in Chapter 2, Section 1 to pay sufficient atten-tion to driving, ”to avoid accidents, road users shall observe care and attention that the circumstances demand”. Chapter 3, Section 1 addresses issues such as health, alertness, soberness and concentration of the driver [59, 113].

[no, because any reductions in mobile phone use generally dissipate within a year or so of the new legislation being introduced. It is interesting to note that compliance with bans or tendencies to use mobile phones for talking and texting etc. is that drivers in highly regulated countries were just as likely to send text messages as drivers in countries with no bans at all.

SHoRT AnSWeR]


Table 2: Overview of the specific mobile phone legislation within the EU. The

specific legislation refers to a hands-free requirement [59]. Note that there are

no EU countries that have total bans on mobile phone usage.

Janitzek et al. [59] reported after elaborate comparative analyses of EU member state legislation and the monetary penalties of mobile phone and other nomadic device violations, that there was no correlation between the level of the mone-tary fine and the number of imposed fines per population. Janitzek et al. [59] also reported that the proportion of drivers committing offenses in countries

LegISLATIon ReQuIReS

HAnDHeLD PHone PRoHIbITeD IF

ReQuIReMenT To uSe

HAnDS-FRee ReQuIReD WHen uSIng

FoRbIDDen To uSeReQuIReMenTS ConCeRnIng

Cou

nTR

y

CoM

PLeT

e b

An

uSe

oF

HA

nD

S-FR

ee e

Qu

IPM

enT

eng

Ine

IS R

un

nIn

g

VeH

ICLe

IS M

oVI

ng

HeA

DSe

T/b

LueT

oo

TH

AD

DIT

Ion

ALL

y FI

XeD

PH

on

e

PH

on

e Fu

nCT

Ion

oTH

eR F

un

CTIo

n

TeXT

Ing

Fu

nCT

Ion

ALL

Fu

nCT

Ion

S TH

AT

InVo

LVe

Co

nTI

nu

ou

S H

An

DLI

ng

HeA

DPH

on

eS

LoCA

TIo

n o

F M

ou

nTI

ng

WA

y o

F FI

XIn

g

Austria x x x x

belgium x x x x x

bulgaria x x x x x

Cyprus x x x x x

Czech Republic x x x x x

germany x x x x x x

Denmark x x x x x

estonia x x x x

greece x x x x x x x

Spain x x x x

Finland x x x x x x

France x x x x x x

Hungary x x x x

Ireland x x x x

Italy x x x x x x

Lithuania x x x x x x

Luxembourg x x x x x x x

Latvia x x x x x

Malta x x x x

The netherlands x x x x x

Poland x x x x x

Portugal x x x x x x x

Romania x x x x

Sweden

Slovenia x x x x x x x

Slovakia x x x x x

united Kingdom x x x x x

Switzerland x x x x x x

Iceland x x x x


with specific mobile phone laws were generally much more likely to speed and not obey seat-belt laws. The proportion of drivers breaking mobile phone laws was 2–4 per cent; running red lights 1 per cent; not wearing a seat belt 12 per cent; drunk driving 1.5 per cent and speeding 40 per cent. Moreover, although drivers in the EU countries were knowledgeable about mobile phone legislation especially younger drivers were more likely to send text messages. The drivers in highly regulated countries were just as likely to send text messages as drivers in the countries with no legislation, suggesting that the laws had little impact on actual behaviour [58, 59].

In a Finnish observational study of handheld mobile phone use, after legislation was introduced in 2003, it was concluded that the positive effect of the ban had dramatically decreased within a year [93]. Moreover, self-reported involvement in mobile phone related incidents had not decreased despite the new legislation. Observational data from Portugal showed that the immediate effects dissipated quickly, and after approximately a year levels of handheld phone use were back to pre-ban figures [39]. In the UK, a similar observational study found that immediately after the introduction of a ban in 2003 handheld mobile phone usage dropped, only to increase to pre-ban levels after 18–24 months. The existing ban was made more stringent in 2007 with the monetary fine being doubled and three penalty points added to the offender’s driving licence. This created another dip in the survey data, only to see a steady increase in hand-held mobile phone usage afterwards [119].

In a New Zealand survey of drivers’ perception of risk it was found that the more a driver conversed (on a mobile phone), the more the perceived risk of this behaviour decreased [43]. The authors also found that drivers in their survey were supportive of a ban (in New Zealand) on mobile phone use provided they could still talk using a hands-free device. Risk perception regarding mobile conversations was generally low.

4.9 Do bans have a positive effect on road traffic safety?

MeTHoD

A widely used method for evaluating road traffic safety effects is the analysis of road crashes. Unfortunately crash data in Europe are somewhat patchy. There is, however, an abundance of insurance loss and crash data in the USA. Study

[no, not in terms of reduced number of crashes, possibly because the bans that are in place are only requirements for hands-free equipment and are not total bans. Moreover, there are many studies that suggest that hands-free usage is no better than handheld.

[

SHoRT AnSWeR


methods vary, but are generally based on large datasets and studies with economic methodology or trend statistics.

Long AnSWeR

What do we mean by road traffic safety effects? There are many ways in which safety can be measured, but here we have focused on crash statistics as a measure of road traffic safety effects. Unfortunately crashes tend to be underreported, and crash causation definitions may also be a matter of educated guesswork.

Janitzek et al. [59] conclude in their survey of EU drivers and their use of noma-dic devices (including mobile phones), that ’drivers’ propensity to engage in high-risk behaviour does not appear to correspond to the stringency of the legislation” (p. 108). Legislation on its own would appear to have limited potential if the goal is to change driver behaviour. In fact, Janitzek et al. [59] acknowledge the inadequacies of legislation in reducing/changing this type of driver behaviour and suggest other policy-driven activities (p.112, ibid.). Some of these activities could entail improved driver training, public information and equipment/technological restrictions on inappropriate use (e. g. specific locations, high speeds etc.).


Early cost-benefit analyses suggested that banning handheld mobile phone use in the US was not justified [42]. The authors also suggested that bans might merely lead drivers to shift their risky behaviour, thereby reducing the net bene-fit of the ban. This was later corroborated by insurance claim studies [50, 51] and US driver surveys [8]. The two insurance company studies investigated the effectiveness of mobile phone legislation; one covering handheld issues and the other covering texting-ban issues. The findings are based on pre- and peri-legis-lation comparisons as well as twinned comparisons with states without phone bans as control [50, 51]. The study on the effects of banning handheld mobile telephone use whilst driving had expected to find a noticeable reduction in the number of insurance claims. However, no benefits of handheld phone bans on collision claims were found for any of the investigated states or for any of the age groups within each state [50]. No reduction in insurance claims could be found in any state after the introduction of texting bans either [51]. Instead, a significant increase in collision claims was found in California. In a survey of US drivers’ self-reported compliance with texting laws, no effects of laws on texting frequency was reported for any state [8]. This could provide insight into the actual habits of drivers, who may be hiding their phones when texting just because of the bans.

Epidemiological studies of the effectiveness and possible benefits of legislation in the US states that had introduced handheld mobile phone bans suggest that an association between the bans and a reduced number of fatal crashes exists, however, the authors also note that there was a downward trend in fatal crashes that could be attributed to other factors [84, 103]. An additional study also factored in weather conditions and found that hands-free requirements reduce the number of fatalities, but only if the weather was inclement [65].

A naturalistic driving study of commercial truck drivers investigated the rela-tionship between US state law pertaining to mobile phone use, fleet policies for ditto and compliance thereof [47, 49]. Their conclusions were that the com-mercial drivers were more likely to comply with the fleet or company rules than state legislation. This compliance appeared to occur irrespective of the nature of the fleet rule or policy, i. e. a rule could either ban or allow handheld use or texting.

It is worth noting that in September 2011, the US federal governmental body for road traffic safety, National Transportation Safety Board (NTSB) called for the 50 states and the District of Columbia to prohibit the use of handheld and hands-free mobile phones for all drivers of commercial vehicles [78]. The basis of this recommendation comes from the investigation of a single fatal accident where tragically eleven people died (in a truck and a mini-bus). Most of those them died because they were not wearing seatbelts. It is currently unclear if this recommendation will be adopted by the US states.


5 SummaryThe available research informs us about the drivers´ attitude towards mobile phones while driving, the effect of legislation, the frequency of distractions on the road and the effects of distraction in experimental settings. However, there is little evidence on the effect of distractions on crash risk. The main findings from the literature reviewed in this report are summarised below:

• Driving performance is impaired by talking on a mobile phone in controlled laboratory, simulator and field studies. •There is no evidence to suggest that hands-free mobile phone use is less risky than handheld use. •The longer a driver talks the more the driver’s exposure increases, thus increasing risk. The more complex the conversation, the greater the level of cognitive distraction and thus, the greater the risk. Therefore short and simple conversations are less risky than prolonged engaging, interesting/complex conversations. •A large scale naturalistic (observational) study suggested that there was no association between safety critical events and truck drivers’ mobile phone conversations. Because the naturalistic driving study technique cannot by its very nature investigate causality, the result has led to much speculation. One plausible theory is that the truck drivers choose the time and the place to make the majority of their phone calls and even use mobile phones (and other communication devices) to abate boredom and fatigue.

• There is broad consensus that driving performance is impaired by visual-manual interactions with mobile phones and any other device or object whether vehicle/driving related or not. •The degree of the impairment is reflected by the time the distracting task takes to complete, the complexity of that task, the capabilities of the driver and the traffic circumstances for the interaction. •Engaging in social media, sending text messages etc. are likely to drastically increase the likelihood of driving unsafely, as these activi ties require both visual and cognitive attention.

• Telephone use, both for calling and texting, is most frequent with younger drivers, high-mileage drivers, and more men than women admit to phone usage while driving. •Most drivers believe that they are more capable of handling mobile phones while driving than other drivers. •A large percentage of drivers think that hands-free sets are much safer to use than handheld sets, even though there is no scientific evidence for this. • Communication device usage is not only steered by perception of risk, but also by perceived social norms and perception of control.

• Most EU countries require hands-free equipment for legal mobile phone use but none of them have a total ban.


•The specification of the EU country-bans varies greatly; defining what is meant by hands-free is not always obvious. •There appears to be a measurable compliance with the bans for the first year or so, but then frequencies of handheld use return to pre-ban levels. •Detailed insurance claims-studies from America suggest that there have been no insurance claim-benefits of the banning of handheld phones or the specific banning of text messaging.

Modern mobile phones can be used for conversation, messaging, browsing the internet, watch/listen to different media, navigation, reading electronic books, updating Facebook/Twitter and so on. The areas of usage are ever expanding and it is hard to predict tomorrow’s applications. This report focus on con-versing, which is a predominantly cognitive task, and texting, which is a cog-nitive-visual-manual task. This limitation is solely due to the lack of available research in other usage areas.

Even though a deterioration of driver performance is an established fact, both through theory and empirical methods, this does not allow conclusions about the actual safety impact in the real world. Some studies based on data from


the real world show that telephone usage has a negative impact on safety, while others only find an impact for the visual-manual subtasks. Some studies see more and some less law compliance, but no clear safety effect seems to result from this. Why are results so vague, why do some contradict each other, and why do we so often need to say “it depends”, when there exists such a large amount of research on the topic?

These questions cannot be answered by just producing even more research. As shown throughout this report, the issue is complex, with many interdependencies. Controlled studies can tell us something about behavioural chan-ges in a given situation and may allow us to tap into the drivers’ attitudes and cognitive load. However, we can never be sure how all this translates to the real world. Observational studies in real traffic, on the other hand, actually show what happens in the real world, but we do not know for sure why these things happen, and the sheer amount of data that is produced in naturalistic driving stu-dies is a challenge in itself. Crash statistics cannot get any better than the underlying crash reports, to date crash data are inadequate especially with respect to the occurrence of distracting activities [59]. Results obtained from a group of college students or from professional truck drivers may not be generalisable to the population at large.

What is called for is a huge concerted effort on many layers. Crash reporting should be done systematically, and great care should be taken to make the in-data as good as possible. Those studies that are undertaken, wether in the simulator or in the field, whether observational or experimental, whether in search of opinions or behavioural changes, should always be planned for maximum data quality and uniform frameworks.

Based on the data that exist so far we can make some educated guesses on why the observed performance decrements do not directly translate into an increased crash rate. The “missing link” could be explained by most drivers most of the time refraining from using mobile phones in more complex traffic situations, in particular when calls are self-initiated. Drivers generally choose their moments for engaging in non-driving related tasks. A caveat is that this does not apply to all drivers, all of the time. Sometimes, however, it goes awry anyway. Being distracted has a negative impact on a driver’s performance, there-fore the crash risk will increase. The bottom line is that incidents may occur if a distraction happens in the wrong place at the wrong time.

The findings in this study concur with other similar literature studies pertaining to mobile phones in traffic [29, 35].


References1. AAA Foundation for Traffic Safety, (2009). 2009 Traffic safety culture

index. AAA Foundation for Traffic Safety: Washington DC, USA.2. AAA Foundation for Traffic Safety, (2010). 2010 Traffic safety culture

index. AAA Foundation for Traffic Safety: Washington DC, USA.3. Alm, H. and Nilsson, L., (1995). The effects of a mobile telephone

task on driver behavior in a car following situation. Accident Analysis and Prevention. 27 (5): p. 707-715.

4. Atchley, P. and Dressel, J., (2004). Conversation limits the functional

field of view. Human Factors. 46 (4): p. 664-673.5. Beck, K. H., Yan, F., and Wang, M. Q., (2007). Cell phone users,

reported crash risk, unsafe driving behaviors and dispositions: A survey

of motorists in Maryland. Journal of Safety Research. 38 (6): p. 683-688.

6. Beede, K. E. and Kass, S. J., (2006). Engrossed in conversation: The

impact of cell phones on simulated driving performance. Accident Analysis and Prevention. 38 (2): p. 415-421.

7. Bhargava, S. and Pathania, V., (2007). Driving under the (cellular)

influence: The link between cell phone use and vehicle crashes.8. Braitman, K. A. and McCartt, A. T., (2010). National reported patterns

of driver cell phone use in the United States. Traffic Injury Prevention. 11 (6): p. 543-548.

9. Braver, E. R., Lund, A. K., and McCartt, A. T., (2009). Review of

“Real-world personal conversations using a hands-free embedded

wireless device while driving: Effect on airbag deployment crash

rates”. Insurance Institute for Highway Safety.10. Briggs, G. F., Hole, G. J., and 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): p. 313-323.

11. Brookhuis, K. A., de Vries, G., and de Waard, D., (1991). The effects

of mobile telephoning on driving performance. Accident Analysis and Prevention. 23: p. 309-316.

12. Burns, P. C., Parkes, A. M., Burton, S., Smith, R. K., and Burch, D., (2002). How dangerous is driving with a mobile phone?

Benchmarking the impairment to alcohol: Berkshire.13. Caird, J. K., Willness, C. R., Steel, P., and Scialfa, C., (2008). A meta-

analysis of the effects of cell phones on driver performance. Accident Analysis and Prevention. 40 (4): p. 1282-1293.

14. Charlton, S. G., (2004). Perceptual and attentional effects on drivers’

speed selection at curves. Accident Analysis and Prevention. 36 (5): p. 877-884.

15. Charlton, S. G., (2009). Driving while conversing: Cell phones that

distract and passengers who react. Accident Analysis and Prevention. 41 (1): p. 160-173.


16. Chisholm, S. L., Caird, J. K., and Lockhart, J., (2008). The effects of

practice with MP3 players on driving performance. Accident Analysis and Prevention. 40 (2): p. 704-713.

17. Collet, C., Guillot, A., and Petit, C., (2010). Phoning while driving I:

A review of epidemiological, psychological, behavioural and physi-

ological studies. Ergonomics. 53 (5): p. 589-601.18. Cooper, J. M. and Strayer, D. L., (2008). Effects of simulator practice

and real-world experience on cell-phone-related driver distraction. Human Factors. 50 (6): p. 893-902.

19. Cooper, J. M., Vladisavljevic, I., Medeiros-Ward, N., Martin, P. T., and Strayer, D. L., (2009). An investigation of driver distraction near

the tipping point of traffic flow stability. Human Factors. 51 (2): p. 261-268.

20. Cosgrove, L., Chaudhary, N., and Roberts, S., (2010). High visibility

enforcement demonstration programs in Connecticut and New York

reduce hand-held phone use. NHTSA: Washington, DC.21. Crisler, M. C., Brooks, J. O., Ogle, J. H., Guirl, C. D., Alluri, P., and

Dixon, K. K., (2008). Effect of wireless communication and enter-

tainment devices on simulated driving performance. p. 48-54.22. Davies, H. T. O., Crombie, I. K., and Tavakoli, M., (1998). When can

odds ratios mislead? British Medical Journal. 316 (7136): p. 316-989.23. Dingus, T. A., Klauer, S. G., Neale, V. L., Petersen, A., Lee, S. E.,

Sudweeks, J., Perez, M. A., Hankey, J., Ramsey, D., Gupta, S., Bucher, C., Doerzaph, Z. R., Jermeland, J., and Knipling, R. R., (2006). The

100-car naturalistic driving study, Phase II - results of the 100-car

field experiment. NHTSA: Washington, DC.24. Drews, F. A., Pasupathi, M., and Strayer, D. L., (2008). Passenger

and cell phone conversations in simulated driving. Journal of Experimental Psychology: Applied. 14 (4): p. 392-400.

25. Drews, F. A., Yazdani, H., Godfrey, C. N., Cooper, J. M., and Strayer, D. L., (2009). Text messaging during simulated driving. Human Factors. 51 (5): p. 762-770.

26. Dula, C. S., Martin, B. A., Fox, R. T., and Leonard, R. L., (2011). Differing types of cellular phone conversations and dangerous driving. Accident Analysis and Prevention. 43 (1): p. 187-193.

27. Eby, D. W., Kostyniuk, L. P., and Vivoda, J. M., (2003). Risky

driving: Relationship between cellular phone and safety belt use. Transportation Research Record. 1843 (Paper No 03-2807): p. 20-23.

28. Elvik, R. (2011). Effects on accident risk of using mobile phones:

Problems of meta-analysis when studies are few and bad. in TRB

Annual Meeting. Waschington DC.29. Expertise collective, (2011). Téléphone et sécurité routière. Inserm:

Paris.30. Farmer, C. M., Braitman, K. A., and Lund, A. K., (2010). Cellphone

use while driving and attributable crash risk. Insurance Institute for Highway Safety.


31. Fitch, G. M. and Hanowski, R. J. (2011). The risk of a safety-critical

event associated with mobile device use as a function of driving

task demands. in Second Conference on Driver Distraction and

Inattention. Gothenburg, Sweden.32. Gallup Organisation, (2010). Road safety: Analytical report. EC

Directorate-General Communication.33. Garabet, A., Horrey, W. J., and Lesch, M. F. (2007). Does exposure

to distraction in an experimental setting impact driver perception of

cell phone ease of use and safety? in Fourth International Driving

Symposium on Human Factors in Driver Assessment, Training and

Vehicle Design. Stevenson, Washington.34. Ghazizadeh, M. and Boyle, L. N., (2009). Influence of driver distrac-

tions on the likelihood of rear-end, angular, and single-vehicle crashes

in Missouri. Transportation Research Record. (No 2138): p. 1-5.35. GHSA, (2011). Distracted driving - What research shows and what

states can do. GHSA (Government Highway Safety Association): Washington, DC.

36. Glassbrenner, D., (2004). Cell phone use on the roads in 2002. National Highway Traffic Safety Administration: Washington DC, USA.

37. Glassbrenner, D., (2005). Driver cell phone use in 2005 – overall

results. NHTSA: Washington, DC.38. Glassbrenner, D. and Ye, T. J., (2007). Driver cell phone use in 2006

– overall results: Washington, DC.39. Godinho, M. B., (2007). Long-term effects of more punitive legisla-

tion concerning the use of mobile phone during driving in Portugal, in Fourth International Driving Symposium on Human Factors in

Driver Assessment, Training and Vehicle Design. p. 381-386.40. Goodman, M. J., Tijerina, L., Bents, F. D., and Wierwille, W., (1999).

Using cellular telephones in vehicles: Safe or unsafe? Transportation Human Factors. 1 (3-42).

41. Gruzdaitis, L., Karola, J., Kotiranta, H., Lamble, D., and Summala, H., (2004). Mental load and visual search at street crossings, in 3rd International Conference on Traffic and Transport Psychology: Nottingham, UK.

42. Hahn, R. W., Tetlock, P. C., and Burnett, J. K., (2000). Should you be

allowed to use your cellular phone while driving? Regulation. 23 (3): p. p. 46-55.

43. Hallett, C., Lambert, A., and Regan, M. A., (2011). Cell phone con-

versing while driving in New Zealand: Prevalence, risk perception

and legislation. Accident Analysis and Prevention. 43 (3): p. 862-869.44. Hancock, P. A., Lesch, M., and Simmons, L., (2003). The distrac-

tion effects of phone use during a crucial driving maneuver. Accident Analysis and Prevention. 35 (4): p. 501-514.

45. Hanowski, R. J., Perez, M. A., and Dingus, T. A., (2005). Driver

distraction in long-haul truck drivers. Transportation Research Part F: Traffic Psychology and Behaviour. 8 (6): p. 441-458.


46. Harbluk, J. L., Noy, Y. I., Trbovich, P. L., and Eizenman, M., (2007). An on-road assessment of cognitive distraction: Impacts on drivers’

visual behavior and braking performance. Accident Analysis and Prevention. 39 (2): p. 372-379.

47. Hickman, J. S. and Hanowski, R. J., (2010). Evaluating the safety

benefits of a low-cost driving behavior management system in com-

mercial vehicle operations. Virginia Tech Transportation Institute: Blacksburg, VA.

48. Hickman, J. S., Hanowski, R. J., and Bocanegra, J., (2010). Distraction in commercial trucks and buses: Assessing prevalence

and risk in conjunction with crashes and near-crashes. Virginia Tech Transportation Institute: Blacksburg, VA.

49. Hickman, J. S., Hanowski, R. J., Camden, M., and Alvarez, A., (2011). Comparison of a state cell phone law versus a fleet cell phone

policy using naturalistic data, in 2nd International Driver Distraction

and Inattention Conference: Gothenburg, Sweden.50. Highway Loss Data Institute, (2009). Hand-held cellphone laws and

collision claim frequencies, in Loss bulletin: Arlington, VA.51. Highway Loss Data Institute, (2010). Texting laws and collision claim

frequencies: Arlington, VA.52. Holtmeyer, K. K., Montazerolghaem, P., and Rowcliffe, S. A., (2008).

Missing the signs: The impact of cell phone use on driving perform-

ance. Sentience - The University of Minnesota Undergraduate Journal of Psychology. 1: p. 17-19.

53. Horrey, W. J. and Wickens, C. D., (2006). Examining the impact of

cell phone conversations on driving using meta-analytic techniques. Human Factors. 48 (1): p. 196-205.

54. Hosking, S. G., Young, K. L., and Regan, M. A., (2009). The effects

of text messaging on young drivers. Human Factors. 51 (4): p. 582-592.

55. Huang, H. F., Stutts, J. C., and Hunter, W. W., (2003). Characteristics

of cell phone-related motor vehicle crashes in North Carolina. p. 10-19.

56. Hurts, K., Angell, L. S., and Perez, M. A., (2011). The distracted

driver: Mechanisms, models and measurement. Reviews of Human Factors and Ergonomics. 7 (1): p. 3-57.

57. Ishigami, Y. and Klein, R. M., (2009). Is a hands-free phone safer

than a handheld phone? Journal of Safety Research. 40 (2): p. 157-164.

58. Jamson, S., (2011). Legislation has minimal impact on drivers’

in-vehicle use of nomadic devices, in Second Conference on Driver

Distraction and Inattention: Gothenburg, Sweden.59. Janitzek, T., Brenck, A., Jamson, S., Carsten, O. M. J., and Eksler, V.,

(2010). Study on the regulatory situation in the member states regard-

ing brought-in (i. e. nomadic) devices and their use in vehicles.


60. Johnson, M. B., Voas, R. B., Lacey, J. H., McKnight, A. S., and Lange, J. E., (2004). Living dangerously: Driver distraction at high

speed. Traffic Injury Prevention. 5 (1): p. 1-7.61. Kass, S. J., Cole, K. S., and Stanny, C. J., (2007). Effects of distrac-

tion and experience on situation awareness and simulated driving. Transportation Research Part F: Traffic Psychology and Behaviour. 10 (4): p. 321-329.

62. Kircher, K., (2007). Driver distraction: A review of the literature. VTI (Swedish National Road and Transport Research Institute): Linköping, Sweden.

63. Klauer, S. G., Dingus, T. A., Neale, V. L., Sudweeks, J., and Ramsey, D., (2006). The impact of driver inattention on near-crash/crash

risk: An analysis using the 100-car naturalistic driving study data. NHTSA: Washington DC. p. 226.

64. Klauer, S. G., Guo, F., Sudweeks, J., and Dingus, T. A., (2010). An

analysis of driver inattention using a case-crossover approach on

100-car data: Final Report. National Highway Safety Administration: Washington, DC, USA.

65. Kolko, J. D., (2009). The effects of mobile phones and hands-free

laws on traffic fatalities. B.E. Journal of Economic Analysis and Policy. 9 (1).

66. Laberge-Nadeau, C., Maag, U., Bellavance, F., Lapierre, S. D., Desjardins, D., Messier, S., and Saı ̈di, A., (2003). Wireless telephones

and the risk of road crashes. Accident Analysis & Prevention. 35 (5): p. 649-660.

67. Lamble, D., Rajalin, S., and Summala, H., (2002). Mobile phone use

while driving: Public opinions on restrictions. Transportation. 29 (3): p. 223-236.

68. Lee, S. E., Klauer, S. G., Olsen, E. C. B., Simons-Morton, B. G., Dingus, T. A., Ramsey, D. J., and Ouimet, M. C., (2008). Detection

of road hazards by novice teen and experienced adult drivers. Transportation Research Record. (2078): p. 26-32.

69. Lehtonen, E., Lappi, O., and Summala, H., (in press). Anticipatory

eye movements when approaching a curve on a rural road depend

on working memory load. Transportation Research Part F: Traffic Psychology and Behaviour.

70. Lerner, N. and Boyd, S., (2005). On-road study of willingness

to engage in distracting tasks. National Highway Traffic Safety Administration (NHTSA): Washington, DC.

71. Lesch, M. F. and Hancock, P. A., (2004). Driving performance during

concurrent cell-phone use: Are drivers aware of their performance

decrements? Accident Analysis and Prevention. 36 (3): p. 471-480.72. Maciej, J., Nitsch, M., and Vollrath, M., (in press). Conversing while

driving: The importance of visual information for conversation

modulation. Transportation Research Part F: Traffic Psychology and Behaviour.


73. Maclure, M. and Mittleman, M. A., (1997). Cautions about car tele-

phones and collisions. New England Journal of Medicine. 336: p. 501-502.

74. McCartt, A. T., Hellinga, L. A., and Braitman, K. A., (2006). Cell

phones and driving: Review of research. Traffic Injury Prevention. 7 (2): p. 89-106.

75. McEvoy, S. P., Stevenson, M. R., McCartt, A. T., Woodward, M., Haworth, C., Palamara, P., and Cercarelli, R., (2005). Role of mobile

phones in motor vehicle crashes resulting in hospital attendance:

a case-crossover study. British Medical Journal. 331 (7514): p. 428-430A.

76. Merat, N. and Jamson, A. H. (2005). Shut up I’m driving! Is talking

to an inconsiderate passenger the same as talking on a mobile tele-

phone? in Third International Driving Symposium on Human Factors

in Driver Assessment, Training and Vehicle Design. Rockport, Maine, USA.

77. Muttart, J. W., Fisher, D. L., Knodler, M., and Pollatsek, A., (2007). Driving without a clue: Evaluation of driver simulator performance

during hands-free cell phone operation in a work zone, in TRB

Annual Meeting. p. 9-14.78. National Transportation Safety Board, (2011). Highway accident

report: Truck-tractor semitrailer median crossover collision with

15-passengar van, Munfordville, KY, March 26, 2010, in National

Transportation Safety Board, Office of Public Affairs: Washington DC.

79. Neale, V. L., Klauer, S. G., Knipling, R. R., Dingus, T. A., Holbrook, G. T., and Petersen, A., (2002). The 100 car naturalistic driving study,

Phase I - experimental design. NHTSA: Washington, DC.80. Nelson, E., Atchley, P., and Little, T. D., (2009). The effects of per-

ception of risk and importance of answering and initiating a cellular

phone call while driving. Accident Analysis and Prevention. 41 (3): p. 438-444.

81. Nemme, H. E. and White, K. M., (2010). Texting while driving:

Psychosocial influences on young people’s texting intentions and

behaviour. Accident Analysis and Prevention. 42 (4): p. 1257-1265.82. NHTSA, (2009). An examination of driver distraction as recorded in

NHTSA databases: Washington, DC.83. NHTSA, (2010). Driver electronic device use in 2009: Washington,

DC.84. Nikolaev, A. G., Robbins, M. J., and Jacobson, S. H., (2010).

Evaluating the impact of legislation prohibiting hand-held cell

phone use while driving. Transportation Research Part A: Policy and Practice. 44 (3): p. 182-193.

85. O’Brien, N. P., Goodwin, A. H., and Foss, R. D., (2010). Talking and

texting among teenage drivers: A glass half empty or half full? Traffic Injury Prevention. 11 (6): p. 549-554.


86. Olson, R., Hanowski, R. J., Hickman, J., and Bocanegra, J., (2009). Driver distraction in commercial vehicle operations. USDOT, FMCSA: Washington, DC.

87. Owens, J. M., McLaughlin, S. B., and Sudweeks, J., (2011). Driver

performance while text messaging using handheld and in-vehicle sys-

tems. Accident Analysis and Prevention. 43 (3): p. 939-947.88. Parkes, A. M., Luke, T., Burns, P. C., and Lansdown, T., (2007).

Conversations in cars: the relative hazards of mobile phones. TRL: Berkshire.

89. Patten, C. J. D., & Kircher, A., & Östlund, J., Nilsson, L., & Svenson, O. (2006). Driver experience and cognitive workload in different

traffic environments. Accident Analysis and Prevention, 38; 887-894.90. Pickrell, T. M. and Ye, T. J., (2009). Traffic Safety Facts Research

Note: Driver electronic device use in 2008. National Highway Traffic Safety Administration.

91. Piechulla, W., Mayser, C., Gehrke, H., and König, W., (2003). Reducing drivers’ mental workload by means of an adaptive man-

machine interface. Transportation Research Part F: Traffic Psychology and Behaviour. 6 (4): p. 233-248.

92. Pöysti, L., Rajalin, S., and Summala, H., (2005). Factors influencing

the use of cellular (mobile) phone during driving and hazards while

using it. Accident Analysis and Prevention. 37 (1): p. 47-51.93. Rajalin, S., Summala, H., Pöysti, L., Anteroinen, P., and Porter, B. E.,

(2005). In-car cell phone use and hazards following hands free legisla-

tion. Traffic Injury Prevention. 6 (3): p. 225-229.94. Recarte, M. A. and Nunes, L. M., (2000). Effects of verbal and

spatial-imagery tasks on eye fixation while driving. Journal of Experimental Psychology: Applied. 6 (1): p. 31-43.

95. Redelmeier, D. A. and Tibshirani, R. J., (1997). Association between

cellular-telephone calls and motor vehicle collisions. New England Journal of Medicine. 336 (7): p. 453-458.

96. Redelmeier, D. A. and Tibshirani, R. J., (2001). Car phones and car

crashes: some popular misconceptions. Canadian Medical Association Journal. 164 (11): p. 1581-1582.

97. Regan, M., Lee, J. D., and Young, K., eds. (2008). Driver distraction.

Theory, effects and mitigation. CRC Press: Boca Raton, London, New York.

98. Regan, M., Hallett, C., and Gordon, C. P., (2011). Driver distrac-

tion and driver inattention: Definition, relationship and taxonomy. Accident Analysis and Prevention. 43 (5): p. 1771-1781.

99. Reimer, B., (2009). Impact of Cognitive Task Complexity on Drivers’

Visual Tunneling. Transportation Research Record. (2138): p. 13-19.100. Rosenbloom, T., (2006). Driving performance while using cell

phones: An observational study. Journal of Safety Research. 37 (2): p. 207-212.


101. Rozario, M., Lewis, I., and White, K. M., (2010). An examination of

the factors that influence drivers’ willingness to use hand-held mobile

phones. Transportation Research Part F: Traffic Psychology and Behaviour. 13 (6): p. 365-376.

102. Rudin-Brown, C. M., Young, K. L., Patten, C., Lenné, M. G., and Ceci, R., (2011). Driver distraction in an unusual environ-

ment: Effects of text messaging in tunnels, in Second International

Conference on Driver Distraction and Inattention: Gothenburg, Sweden.

103. Sampaio, B., (2010). On the identification of the effect of prohibiting

hand-held cell phone use while driving: Comment. Transportation Research Part A: Policy and Practice. 44 (9): p. 766-770.

104. Savolainen, P. T., Das, A., Gates, T. J., and Datta, T. K. (2011). Cell

phone use and crash patterns among Michigan drivers. in TRB

Annual Meeting. Washington DC, USA.105. Shinar, D., Tractinsky, N., and Compton, R., (2005). Effects of prac-

tice, age, and task demands, on interference from a phone task while

driving. Accident Analysis and Prevention. 37 (2): p. 315-326.106. Smahel, T., Smiley, A., and Donderi, D., (2008). The effects of cellular

phone use on novice and experienced driver performance: An on-

road study Human Factors and Ergonomics Society Annual Meeting Proceedings. 52 (23): p. 1910-1914.

107. Strayer, D. L., Drews, F. A., and Johnston, W. A., (2003). Cell phone-

induced failures of visual attention during simulated driving. Journal of Experimental Psychology-Applied. 9 (1): p. 23-32.

108. Strayer, D. L. and Drews, F. A., (2004). Profiles in driver distraction:

Effects of cell phone conversations on younger and older drivers. Human Factors. 46 (4): p. 640-649.

109. Strayer, D. L., Drews, F. A., and Crouch, D. J., (2006). A comparison

of the cell phone driver and the drunk driver. Human Factors. 48 (2): p. 381-391.

110. Strayer, D. L. and Drews, F. A., (2007). Cell-phone-induced driver

distraction. Current Directions in Psychological Science. 16 (3): p. 128-131.

111. Stutts, J. C., Feaganes, J., Rodgman, E. A., Hamlett, C., Meadows, T., Reinfurt, D. W., Gish, K., Mercadante, M., and Staplin, L., (2003). Distractions in everyday driving: Washington, DC.

112. Sullman, M. J. M. and Baas, P. H., (2004). Mobile phone use amongst

New Zealand drivers. Transportation Research Part F: Traffic Psychology and Behaviour. 7 (2): p. 95-105.

113. Svensk författningssamling, (2011). Trafikförordningen, Rixlex, Editor.

114. Taylor, D. M., MacBean, C. E., Das, A., and Rosli, R. M., (2007). Handheld mobile telephone use among Melbourne drivers. Medical Journal of Australia. 187 (8): p. 432-434.

115. Thulin, H. and Gustafsson, S., (2004). Mobile phone use while driving. VTI: Linköping, Sweden.


116. Troglauer, T., Hels, T., and Christens, P. F., (2006). Extent and

variations in mobile phone use among drivers of heavy vehicles in

Denmark. Accident Analysis and Prevention. 38 (1): p. 105-111.117. Törnros, J. E. B. and Bolling, A. K., (2005). Mobile phone use -

Effects of handheld and handsfree phones on driving performance. Accident Analysis and Prevention. 37 (5): p. 902-909.

118. Walker, L., Williams, J., and Jamrozik, K., (2006). Unsafe driving

behaviour and four wheel drive vehicles: observational study. British Medical Journal. 333 (7558): p. 71-73.

119. Walter, L., (2010). Seatbelt and mobile phone usage surveys: England

and Scotland 2009, U. Department for Transport, Editor.120. Vanlaar, W., Simpson, H., Mayhew, D., and Robertson, R., (2007).

The road safety monitor 2006. Driver distraction. Traffic Injury Research Foundation: Ottawa, Ontario, Canada.

121. Waylen, A. E., Horswill, M. S., Alexander, J. L., and McKenna, F. P., (2004). Do expert drivers have a reduced illusion of superiority? Transportation Research Part F-Traffic Psychology and Behaviour. 7 (4-5): p. 323-331.

122. Wester, A. E., Bocker, K. B., Volkerts, E. R., Verster, J. C., and Kenemans, J. L., (2008). Event-related potentials and secondary task

performance during simulated driving. Accid Anal Prev. 40 (1): p. 1-7.123. White, K. M., Hyde, M. K., Walsh, S. P., and Watson, B., (2010).

Mobile phone use while driving: An investigation of the beliefs

influencing drivers’ hands-free and hand-held mobile phone use. Transportation Research Part F: Traffic Psychology and Behaviour. 13 (1): p. 9-20.

124. White, M. P., Eiser, J. R., and Harris, P. R., (2004). Risk perceptions

of mobile phone use while driving. Risk Analysis. 24 (2): p. 323-334.125. Victor, T. W., Harbluk, J. L., and Engström, J. A., (2005).

Sensitivity of eye-movement measures to in-vehicle task difficulty. Transportation Research Part F: Traffic Psychology and Behaviour. 8 (2): p. 167-190.

126. Vivoda, J. M., Eby, D. W., St. Louis, R. M., and Kostyniuk, L. P., (2008). Cellular phone use while driving at night. Traffic Injury Prevention. 9 (1): p. 37-41.

127. Wogalter, M. S. and Mayhorn, C. B., (2005). Perceptions of driver

distraction by cellular phone users and nonusers. Human Factors. 47 (2): p. 455-467.

128. Wood, C. and Hurwitz, J., (2005). Driver workload management

during cell phone conversations, in Driving Assessment 2005: 3rd

International Driving Symposium on Human Factors in Driver

Assessment, Training, and Vehicle Design: Rockport, Maine, USA.129. Yannis, G., Papadimitriou, E., Karekla, X., and Kontodima, E.,

(2010). Mobile phone use by young drivers: Effects on traffic speed

and headways. Transportation Planning and Technology. 33 (4): p. 385-394.


130. Young, R. A. and Schreiner, C., (2009). Real-world personal conver-

sations using a hands-free embedded wireless device while driving:

Effect on airbag-deployment crash rates. Risk Analysis. 29 (2): p. 187-204.

131. Young, R. A., (2011). Driving consistency errors overestimate crash

risk from cellular conversation in two case-crossover studies, in Driving Assessment 2011: 6th International Driving Symposium on

Human Factors in Driver Assessment, Training, and Vehicle Design, I. C. University of Iowa, Editor: Olympic Valley - Lake Tahoe CA. p. 298-305.

132. Zhou, R., Rau, P.-L. P., Zhang, W., and Zhuang, D., (2011, in press). Mobile phone use while driving: predicting drivers’ answering inten-

tions and compensatory decisions. Safety Science.


ACTIVITy (WITH eXAMPLeS) VISuAL AuDIToRy MAnuAL CognITIVe

mobile telephone conversationhandheld, dialling/hanging up

yes yes yes yes

mobile telephone conversation handheld, talking

no yes yes yes

mobile telephone conversation hands-free, dialling/hanging up

yes (if not voice controlled)

yes yes (if not voice controlled) yes

mobile telephone conversation hands-free, talking

no yes no yes

sms/mms, incoming yes, when reading signal when message arrivesyes, holding phone, possibly

some keyingyes

sms/mms, going outyes, when writing no (possibly confirmation) yes, keying in message yes

posting on Twitter, Facebookyes, when writing no (possibly confirmation) yes, keying in message yes

internet browsing (incl. reading Twitter/ Facebook)

yes depends on pageyes, magnitude depends on page, at least holding

equipmentyes

video/youtubeyes yes, usually only when starting film yes

music/Spotifyno yes no not much

navigationyes yes only when programming some

e-mailyes signal when message arrives yes, keying in message yes

making notesyes no yes yes

real-time traffic informationyes possible when starting up some

e-booksyes no some yes

AnnexAnnex I. Overview of different functions that can be executed with a modern Smartphone and similar communication

devices and their coupling to different sensory modalities.


STuDy TyPe DeSCRIPTIon DATA uSAge

Field studies of naturalistic driving type

observations of drivers going about their daily routines in real traffic. usually continuous data collection, often in combination with video. usually over longer time period, up to a year and more.

Prevalence of certain behaviours (e. g. usage of telep-hones) in certain situations, by certain age, occupatio-nal or other groups. Association of those behaviours with crash, near-crash or incident involvement. Hypothesis generation. Long-term behavioural changes can be studied.no causal conclusions possible that a certain behaviour caused something else, e. g. a crash, or that a beha-vioural change was caused by a certain factor, as other unobserved background variables may very well have influenced the observed variables.

Field studies of FoT type As naturalistic studies, but with an additional manipulation. either the same driver uses a vehicle with and without a safety system, communication device, etc., or a randomly assigned group of drivers uses the safety system/ commu-nication device, while the other randomly assigned group does not use such a device.

Similar to naturalistic studies, but stronger case to base observed changes in behaviour between no-device-usage and device-usage on the actual presence of the device.

Field observations (outside vehicle)

observers, cameras or other sensors are posted in e. g. junctions or at traffic lights or other locations of interest and note driver behaviour by looking into the vehicles and/or measuring speeds, etc. Can be short- or long-term.

Prevalence and/or distribution of certain behaviours in certain locations, can be split into different conditions depending on situational variations, like weather etc.

Controlled field studies usually with instrumented vehicles in real traffic, often experimenter in car, instructions on where to drive and what to do while driving, comparison of behaviour in a given situation with and without a certain “treatment”, e. g. an additional task, participants are assigned to conditions, to prevent self-selection. usually rather short-term.

Causal influence of one factor on another, e. g. behavioural changes in a given situation due to the “tre-atment”, but no information on whether a participant would have exposed him/herself to the “treatment” voluntarily, except possibly via self-reports.Long-term behavioural changes usually not assessed.

Studies on dedicated test tracks

Similar to controlled field studies, but instead of real traffic a test track is used, which increases the possibilities to control the situation and allows somewhat more critical manoeuvres.

Similar to controlled field studies.

Driving simulator studies (dif-ferent fidelities)

More or less advanced driving simulators are used for full control over the situation, both with respect to traffic, wea-ther, road curvature, etc. Allows isolation of the factor to be manipulated. Possible to execute dangerous manoeuvres and to test setups that do not yet exist in reality.

Similar to controlled field studies, but often with richer and more accurate log data. This is traded off for a de-crease in “external validity”, that is, the knowledge how the data obtained in the more artificial situation reflects what actually would have happened in traffic.

Polls Participants are selected based on certain criteria and then interviewed via telephone, internet, questionnaire, etc. The goal is to get a representative sample of the population.

If participants are selected in a way that makes them representative, a good picture of what people in a cer-tain region say about a certain issue. Depending on the sensitivity of the issue this corresponds more or less to what their actual attitude is. “Polls” with self-selection, where anybody can answer, do not necessarily reflect the view of the population at large, but of the group that chose to answer.

Focus groups Around 5-8 participants are invited to discuss a certain topic for about 1-2 hours. The moderator usually tries to keep a low profile in the discussion.

no representativeness for the population at large can be assumed. Focus groups are often used to get new perspectives on a certain topic, which can then be discussed in a more formal way.

Interview studies usually longer interviews that are structured or semi-structured and are aimed at in-depth investigations of a certain issue with a smaller number of e. g. experts or people involved with the topic at hand.

Depending on the number of participants and par-ticipant selection generalisability of the findings is variable.

Crash database analyses Analysis of past crashes that have been inserted into a national or international crash database.

The quality of the analysis depends on the quality of the data recorded.

epidemiological studies observational studies that investigate the prevalence of a certain behaviour in one group (e. g. in crashes) and com-pare it to the prevalence of a certain behaviour in another group (e. g. in cases without crashes). In a case-control design the drivers with crashes are compared to drivers without crashes, while in a case-crossover design a driver with a crash is compared to him- or herself at another point in time without crash.

epidemiological studies, like other observational stu-dies, cannot prove causality, but indicate an association of a certain behaviour with e. g. crash risk. The strength of an epidemiological study depends on the number of cases (e. g. crashes) and controls (baseline episodes without crashes) included in the study. A correct selec-tion of baseline episodes is paramount to a meaningful result interpretation.

Annex II. Overview of different research methods and the associated data usage.

In WHAT WAy ARe DRIVeRS AFFeCTeD WHen TALKIng oR TeXTIng?Mobile phone conversation increases reaction times and missed events. In addition, texting and dialling require the driver to look at the device and away from the road leading to reduced vehicle control and an increase in missed events.

DoeS PRACTICe IMPRoVe THe AbILITy To DRIVe WHILe HAnDLIng A MobILe PHone?The small body of available data indicates that practice of simultaneous driving and telephone use has at best a very limited effect. There seems to be some agreement, however, that more experienced drivers have better strategies to handle telephone usage.

Do HAnDS-FRee AnD HAnDHeLD MobILe PHoneS AFFeCT DRIVIng PeRFoRMAnCe In A DIFFeRenT MAnneR?Cell phone conversations during driving have a negative impact on driving performance regardless of whether a handheld or a hands-free phone is used.

WHAT ARe DRIVeRS’ ATTITuDeS To MobILe PHone uSAge WHen DRIVIng?Drivers are generally aware of the risks involved with texting and talking while driving. However, perceived social norms, attitude, perception of control and believed importance of the call are more indicative of phone use than perceived risk.

HoW Do DRIVeRS uSe THeIR MobILe CoMMunICATIon DeVICeS? Around 2–6 per cent of all drivers use mobile phones at any given moment. Most of them use handheld devices. younger males are most likely to both call and text. Frequent drivers and drivers alone in their car are more likely to use their telephone while driving.

Do DRIVeRS ADAPT THeIR DRIVIng beHAVIouR To DeVICe-uSAge?Drivers choose when and where to use their telephones, and also during a call there are indications for compensatory driving behaviour. How successful the adaptation strategies are depends largely on the driver’s capabilities and motivations, but also on the predictability of the situation.

HoW DAngeRouS IS IT FoR DRIVeRS To TALK oR TeXT?It depends on the situational context and the driver’s capabilities. Recent naturalistic research on commercial truck drivers indicates that the occurrence of crashes, near-crashes and safety critical events is correlated with handling the telephone (dialling, texting), but not with just talking on a phone. This does not mean, however, that talking on the phone while driving is safe in all situations and for all drivers. It is more likely to be at least in part a sign that drivers have capabilities to self-regulate their telephoning behaviour.

DoeS A bAn on HAnDHeLD MobILe PHoneS WoRK WITH ReSPeCT To CoMPLIAnCe?no, because any reductions in mobile phone use generally dissipate within a year or so of the new legislation being introduced. It is interesting to note that compliance with bans or tendencies to use mobile phones for talking and texting etc. is that drivers in highly regulated countries were just as likely to send text messages as drivers in countries with no bans at all.

Do bAnS HAVe A PoSITIVe eFFeCT on RoAD TRAFFIC SAFeTy?no, not in terms of reduced number of crashes, possibly because the bans that are in place are only requirements for hands-free equipment and are not total bans. Moreover, there are many studies that suggest that hands-free usage is no better than handheld.

PHo

ToS

by

KA

TjA

KIR

CHeR

huvudkontor/head office

LInKÖPIng post/mail Se-581 95 LInKÖPIng tel +46 (0)13-20 40 00www.vti.se

SToCKHoLM post/mail boX 55685 Se-102 15 SToCKHoLM tel +46 (0)8-555 770 20

goTHenbuRgpost/mail boX 8072Se-402 78 goTHenbuRgtel +46 (0)31-750 26 00

boRLÄnge post/mail boX 920Se-781 70 boRLÄngetel +46 (0)243-44 68 60