ArticlePDF Available

Identification of pre-accident situations faced by French cyclists in urban environments: An on-road study

June 2017
Le travail humain 80(2):187-234

DOI:10.3917/th.802.0187

Authors:

Jordan Navarro

Lumière University Lyon 2

Samuel Aupetit

ERGO-CENTRE

Show all 5 authorsHide

The increasing number of cyclists, and the introduction of new infrastructure has created new ways for cyclists and other road users to interact, which may impact road safety. Consequently, our aim was to contribute to a better understanding of accidents involving cyclists through an in-depth, pre-accident analysis based on an activity-centred approach. Various pre-accident situations, judged as risky by twenty cyclists, were collected and classified as a function of their relative frequency of occurrence. The analysis found that a third of these situations involved interactions with pedestrians or other obstacles. Then, the most frequent situations are related to: (a) interactions with motorists either entering or leaving a parking space; (b) motorists who stepped in front of a cyclist when turning right; (c) cars entering the road from the right at an intersection, and (d) motorists squeezing cyclists off the road when overtaking. An analysis of situations where cyclists' visibility was impaired found similar results except for situations (c) and (d) where cyclists' visibility was very rarely engaged, and which both appeared to be associated to deliberate motorist behaviours. These findings point out the most frequent pre-accident situations reported by cyclists, and highlight the need for their better integration, as a new type of road user, in urban areas. © Presses Universitaires de France. Tous droits réservés pour tous pays.

Autres usagers de la route impliqués dans les situations de pré-accident à Paris et Lyon. Pourcentage (%) du nombre de situations reportées. Document téléchargé depuis www.cairn.info-Université Louis Lumière Lyon 2-159.84.74.48-17/07/2017 12h00. © Presses Universitaires de France Document téléchargé depuis www.cairn.info-Université Louis Lumière Lyon 2-159.84.74.48-17/07/2017 12h00. © Presses Universitaires de France

…

Other road users implied in the pre-accident situations reported in Paris and Lyon urban areas. Percentage (%) of the number of situations reported.

…

Categorisation of all pre-accident situations reported ordered by frequency. tABleAu 2. Catégorisation de l'ensemble des situations de pré-accident rapportées classées de la plus fréquente à la moins fréquente.

…

Categorisation of pre-accident situations reported with cyclists visibility engaged ordered by frequency. tABleAu 3. Catégorisation des situations de pré-accident rapportées associées à un défaut de visibilité des cyclistes et classées de la plus fréquente à la moins fréquente.

…

Category by category repartition of all pre-accident situations reported related to: motorists' failure to detect cyclists (sensorial and/ or cognitive conspicuity), cyclists inaccurate expectations and over- optimism, or motorists' or pedestrians' deliberate behaviours. tABleAu 5. Répartition catégorie par catégorie de l'ensemble des situations de pré- accident rapportées selon : l'incapacité des automobilistes à détecter les cyclistes (conspi- cuité sensorielle et/ou conspicuité cognitive), les attentes erronées des cyclistes et un excès d'optimisme, ou un comportement délibéré des automobilistes ou piétons.

…

Figures - uploaded by Jordan Navarro

Content may be subject to copyright.

Content uploaded by Jordan Navarro

Content may be subject to copyright.

IDENTIFICATION OF PRE-ACCIDENT SITUATIONS FACED BY

FRENCH CYCLISTS IN URBAN ENVIRONMENTS: AN ON-ROAD

STUDY

Jordan Navarro, Samuel Aupetit, Gaël Allain, Christophe Hilaire, Joceline Rogé

Presses Universitaires de France | « Le travail humain »

2017/2 Vol. 80 | pages 187 à 207

ISSN 0041-1868

ISBN 9782130788645

Article disponible en ligne à l'adresse :

--------------------------------------------------------------------------------------------------------------------

http://www.cairn.info/revue-le-travail-humain-2017-2-page-187.htm

--------------------------------------------------------------------------------------------------------------------

Pour citer cet article :

--------------------------------------------------------------------------------------------------------------------

Jordan Navarro et al., « Identification of pre-accident situations faced by french

cyclists in urban environments: an on-road study », Le travail humain 2017/2 (Vol.

80), p. 187-207.

DOI 10.3917/th.802.0187

--------------------------------------------------------------------------------------------------------------------

Distribution électronique Cairn.info pour Presses Universitaires de France.

La reproduction ou représentation de cet article, notamment par photocopie, n'est autorisée que dans les

limites des conditions générales d'utilisation du site ou, le cas échéant, des conditions générales de la

licence souscrite par votre établissement. Toute autre reproduction ou représentation, en tout ou partie,

sous quelque forme et de quelque manière que ce soit, est interdite sauf accord préalable et écrit de

l'éditeur, en dehors des cas prévus par la législation en vigueur en France. Il est précisé que son stockage

dans une base de données est également interdit.

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 186 / 240

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 187 / 240

Le Travail Humain, tome 80, n°2/2017, 187-205

EMPIRICAL STUDIES

RECHERCHES EMPIRIQUES

IDENTIFICATION OF PRE-ACCIDENT SITUATIONS

FACED BY FRENCH CYCLISTS IN URBAN

ENVIRONMENTS: AN ON-ROAD STUDY

IDENTIFICATION ET CLASSIFICATION

DES SITUATIONS DE PRÉ-ACCIDENT RAPPORTÉES

PAR LES CYCLISTES FRANÇAIS EN ZONE

URBAINE : UNE ÉTUDE EN CONTEXTE NATUREL

by/par Jo r D A n nA v A r r o 1, sA M u e l Au p e t i t 2, gA ë l Al l A i n 3,

ch r i s t o p h e hi l A i r e 2 & Jo c e l i n e ro g é 4

SUMMARY

The increasing number of cyclists, and the introduction of new infrastructure

has created new ways for cyclists and other road users to interact, which may

impact road safety. Consequently, our aim was to contribute to a better under-

standing of accidents involving cyclists through an in-depth, pre-accident analy-

sis based on an activity-centred approach. Various pre-accident situations, judged

as risky by twenty cyclists, were collected and classied as a function of their rela-

tive frequency of occurrence. The analysis found that a third of these situations

involved interactions with pedestrians or other obstacles. Then, the most frequent

situations are related to: (a) interactions with motorists either entering or leav-

ing a parking space; (b) motorists who stepped in front of a cyclist when turning

right; (c) cars entering the road from the right at an intersection, and (d) motor-

ists squeezing cyclists off the road when overtaking. An analysis of situations

where cyclists’ visibility was impaired found similar results except for situations

appeared to be associated to deliberate motorist behaviours. These ndings point

out the most frequent pre-accident situations reported by cyclists, and highlight the

need for their better integration, as a new type of road user, in urban areas.

Keywords: cyclist, on-road studies, pre-accident.

1. Corresponding author:

Laboratoire d’étude des mécanismes cognitifs, Institut de psychologie, université Lyon-II,

5 avenue Pierre Mendès-France, 69676 Bron Cedex, France – email : jordan.navarro@univ-

lyon2.fr

2. Dédale, 21 rue de Fécamp, 75012 Paris, France – email : ergo-centre@hotmail.com

3. Mémoire et marketing, Le chanay, France – email : gallain@memoire-marketing.com

4. IFSTTAR-TS2-LESCOT, 25 avenue François Mitterrand, 69675 Bron Cedex, France

– email : joceline.roge@ifsttar.fr

Identiﬁcation of pre-accident situations faced by cyclists

Jordan Navarro

205

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

188 Jordan Navarro et al.

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 188 / 240

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 189 / 240

RÉSUMÉ

Le nombre grandissant de cyclistes et les évolutions des aménagements routiers

associés induisent de nouvelles interactions entre cyclistes et autres usagers de la

route. Compte tenu des enjeux sécuritaires, l’objectif poursuivi ici était de mieux

comprendre les situations accidentogènes impliquant les cyclistes via une analyse

détaillée de ces situations avec une méthodologie ergonomique fondée sur l’analyse

de l’activité. Les différentes situations de pré-accident jugées comme risquées par

vingt cyclistes lors de leur pratique habituelle ont été collectées et classées selon leur

fréquence d’occurrence relative. Les résultats indiquent que près d’un tiers des situa-

tions de pré-accident rapportées résultent de l’interaction entre cyclistes et piétons.

Après quoi et par ordre de fréquence, les situations étaient liées à des interactions

avec des automobilistes (a) entrant ou quittant le stationnement, (b) interférant

avec les cyclistes en tournant à droite, (c) interférant avec les cyclistes en venant

de la droite à une intersection, (d) ou encore dépassant les cyclistes de trop près.

Lorsque l’intérêt s’est porté plus spéciquement sur les situations où la visibilité

des cyclistes était engagée, des résultats similaires ont été obtenus exception faite

des catégories (c) véhicule venant de la droite et coupant la trajectoire du cycliste

et (d) dépassant les cyclistes de trop près qui tendent à disparaître. Ces situations

(c & d) étaient majoritairement associées à des comportements délibérés des auto-

mobilistes. Dans leur ensemble, les résultats permettent de classer par fréquence

d’apparition les situations de pré-accident rapportées par les cyclistes et pointent la

nécessitée d’une meilleure intégration des usagers cyclistes en zone urbaine.

Mots-clés : Cycliste, contexte naturel, analyse située, pré-accident.

INTRODUCTION

Cyclists’ number on the roads of several French big cities such as Paris

and Lyon is increasing (Papon & De Solère, 2010). Public authorities keep

going on encouraging people to ride bikes for several reasons including

sustainable development, reducing trafc jams, reducing noise pollution in

urban areas, and ghting against sedentariness. Bike use in big urban areas

has been supported since the early 2000’s. Infrastructures improvements

in favour of cyclists such as bike paths, bike lanes, contraow bike lanes

and shared lanes (between public bus, taxis and cyclists) have been created

in several French cities including Paris and Lyon. If those improvements

aims to favour safe bike use and tend to do so (Forester, 1994), they

might also lead to a decrease of attention cyclists pay to other road users

(de Lapparent, 2005). The increasing number of cyclists and the related

infrastructure evolutions came with new interactions between cyclists and

other road users that might impact on road safety. Following this idea, a

high proportion of cyclists crashes seems to be related to drivers failure

to detect cyclists (Wood, Lacherez, Marszalek, & King, 2009). Even if

cyclists crashes are underreported (Amoros, Martin, & Laumon, 2006),

those implying another motorist are more severe (Amoros, Chiron, Thélot,

& Laumon, 2011). Cyclists are considered as vulnerable road users as they

are physically exposed directly in case of crash.

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 188 / 240

189Identication of pre-accident situations faced by cyclists

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 189 / 240

Several studies focused on accidents records in order to estimate the risks

associated with bike riding (e.g. Aultman-Hall & Hall, 1998; Aultman-Hall

& Kaltenecker, 1999; Minikel, 2012). For instance, light conditions and

infrastructure are known to impact on risky situations. More than a third of

motorist/cyclist crashes occurred under low visibility conditions (at night,

twilight or daybreak). A disproportionate ratio taking into account the

few cyclists riding at those times of the day (Lacherez, Wood, Marszalek,

& King, 2013). Regarding infrastructures, it was observed that Australian

cyclists in Brisbane feel more fear than Danish cyclists in Copenhagen

(Chataway, Kaplan, Nielsen, & Prato, 2014). According to the authors, the

differences were due to cyclists’ more adapted infrastructures and a better

integration of cyclists as road users in Copenhagen. Those studies aimed at

describing crashes in a variety of situations but did not intend to understand

in details the reported crashes. If a detailed methodologies to do so exists

(e.g. Ferrandez, 1995), very few studies focused on detailed crashes analysis

implying cyclists and trying to understand what happened from road users

perspective. Two studies however pointed out that many cyclists/motorists

accidents occurred when the motorist turned right stepping into the cyclist

located to the right of the vehicle (Räsänen & Summala, 1998; Summala,

Pasanen, Räsänen, & Sievänen, 1996). While turning right motorists often

only look to the left (Summala et al., 1996), explaining why only 11 %

of motorists perceived the cyclist before the collision. Cyclists were more

aware of the motorist presence, as 68 % perceived the motorist before

the collision. Among the cyclists that saw the vehicle, 92 % declared they

thought that the driver would yield as prescribed in the Highway Code.

In 37 % of the crashes analysed, neither the motorist nor the cyclist had

realized the crash risk, in the remaining situations the motorist (27 %), the

cyclist (24 %) or both (12 %) tried to avoid the crash (Räsänen & Summala,

1998). Infrastructure changes at bicycle crossings were showed to modify

both cyclists and motorists behaviours even in situations where the formal

change does not apply (Räsänen, Koivisto, & Summala, 1999).

The rationale followed in the current study was to analyse cyclists

reported risky situations from their own perspective based on their usual

daily cycling activity. Hereafter those situations will be referred as pre-

accident situations. Pre-accidents situations correspond to the situations

faced prior an actual collision, with a breakdown of the situation followed

by an emergency situation (Ferrandez, 1995). The methodology to do so

refers to “activity-centered ergonomics” (Ombredane & Faverge, 1955;

Theureau, 2003). The approach was previously applied successfully to

road safety to analyse motorcycle riders’ (Aupetit, Riff, Buttelli, & Espié,

2013). The methodology stand on real word activities analyse with minimal

intervention of the experimenters. In the current study, cyclists were given

no other instructions than to ride their bike as they usually do. The analysis

is performed by means of face-to-face interviews where participants are

confronted to videos traces of their own cycling activity to explore their

behaviours in the context of appearance. Although time consuming, the

best interest of this methodology is to offer an in-deep analysis taking

into account technical, social and cultural environments. Following this

method pre-accident situations will not be explored objectively organizing

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

190 Jordan Navarro et al.

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 190 / 240

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 191 / 240

crashes data according to the situation of occurrence but rather subjectively

based on cyclists’ comments on their own behaviours. In other words,

this methodological framework offers the possibility to analyse the same

cyclist safety concerns from another point of view. This new perspective

is not meant to be objective but to enrich accident analysis understanding

by exploring the main protagonists actions. This methodology is also

well suited to analyse variation in practices due to road users’ cultural

differences. In the current study, cyclists from two different French cities

were considered, the rational was that pre-accident situations and their

relative frequency might be different depending on the city considered as

suggested by previous studies (Chataway et al., 2014).

As suggested by previous results, crashes involving cyclists might be

related to a problem of cyclists visibility (Summala et al., 1996). A survey

leaded on 622 motorists and 838 cyclists showed that the main source of

accident between motorists and cyclists is linked to motorists not viewing

cyclists before the crash (Wood et al., 2009). Cyclists’ lack of visibility might

be explained by a different perception of the reality by motorists and cyclists.

Indeed, the distance at which cyclists estimate being visible (100.3m) is twice

bigger than the one estimated by drivers (48.3m). Additional data collected

on a test track indicated drivers difculties to detect cyclists, especially at

night (Wood et al., 2010). It also appears that drivers and cyclists do not

have the same representation of cyclists’ visibility. 88 % of cyclists judged

that drivers do not visually seek for potential cyclists’ presence but only 66 %

of drivers. Conversely, only 57 % of cyclists declared being hardly visible

in the trafc while 77 % of the drivers think so (Wood et al., 2009). All

together cyclists assume to be perceived earlier and more often than they

actually are by drivers. This gap between cyclists and drivers perception

may be attributed to four different factors: (a) the fact that drivers look in

the direction of the cyclist but do not perceive it (“looked-but-failed-to-

see phenomenon” ; Brown, 2005) and therefore related to a lack of cyclists

sensorial conspicuity (Engel, 1971, 1974; Wulf, Hancock, & Rahimic, 1989),

(b) a lack of cyclist appearance expectation from drivers perspective, because

biking in urban areas is relatively recent (Clarke, Ward, Bartle, & Truman,

2007) and therefore related to a problem of cognitive conspicuity (Hancock,

Wulf, Thom, & Fassnacht, 1990), (c) cyclists inaccurate expectations, such

as drivers yielding (Räsänen & Summala, 1998), (d) cyclists optimism

regarding the risk of accident. Cyclists are indeed optimistic regarding the

risk of being involved in an accident (King, Wood, Lacherez, & Marszalek,

2012). And for a given situation, cyclists perceive less risks than drivers

(Chaurand & Delhomme, 2013).

Within this context, our aim was to contribute to a better understanding

of accident involving cyclists through in deep pre-accident analysis based on

an activity-centred approach. The main objective of the current experiment

was to collect the different pre-accident situations judged as risky by cyclists

during their regular cycling activity and to class them according to their relative

frequency of occurrence. Another objective was to measure how cyclists’

perceived lack of visibility is a key factor in pre-accident situations appearance.

Finally, the type pavement and other road users implied in pre-accident

situations have been considered to guide future infrastructure development.

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 190 / 240

191Identication of pre-accident situations faced by cyclists

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 191 / 240

I. METHOD

i.1. pA r t i c i p A n t s

20 cyclists (7 females and 13 males), aged from 24 to 63 years (mean

age = 38.32 years ± 9.7), took part in the experiment. Table 1 shows each

participant gender, residence city, age, declared frequency of cycle use,

riding distance and time per day. To those data, the actual distance tracked

during the experiment was added as well as the number of risky situations

reported. All cyclists had normal or corrected-to-normal vision. Each

received a nancial compensation of 200 euros for their participation. The

local ethics committee approved the experiment.

i.2. nA t u r A l i s t i c o n -r o A D e x p e r i M e n t

The cyclists participating in the experiment were asked to ride their bike

as they usually do without any constraints apart from wearing the provided

helmet equipped with a video camera switched on. All cyclists’ travels

were therefore recorded by the video camera for the whole duration of the

experiment (i.e. four complete weeks). That way real life cycle journeys

data (including familiar journeys following regular routes at regular times)

with minimal intrusion from experimenters were collected.

The cyclists have been selected in the two biggest French urban areas:

Paris and Lyon urban agglomerations. Those two cities were the two rst

cities where large-scale public bike systems were proposed, in Lyon in 2005

and in Paris in 2007 and where cyclists are always more numerous (Papon

& De Solère, 2010).

So as to cover a large variety of weather and light conditions, the

experiment was spread on period going from March to December.

Five cyclists were experienced in March/April, ve in May/June, ve in

September/October and ve in November/December.

i.3. DA t A c o l l e c t i o n

Three types of data were collected: (1) video recordings of all cyclists’

travels, (2) logbooks to keep track of risky situations faced by cyclists’ day

by day and (3) self confrontation interviews to as to understand in details

the risky situations logged by cyclists.

I.3.1. Video recording

Participants were asked to record all bike trips using a video camera that

was offering a high quality recording and a wide-angle view (HD 1080p and

170° of eld of view, Garmin VIRB Elite®). In order to minimize interferences

between the video camera and the cycling activity helmets equipped with

camera xing device were provided to participants. The helmet and the

camera were adjusted to each participant so as to ensure both participants’

comfort and safety and a good video recording of the visual scene.

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

192 Jordan Navarro et al.

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 192 / 240

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 193 / 240

tA B l e 1: Participants’ main characteristics

tA B l e A u 1. Caractéristiques principales des participants à l'étude

Gen-

der City

Age

(years)

Declared

frequency

of cycle

use (days

per week)

Declared

riding

distance

per day

(km)

Declared

riding

time

per day

(minutes)

Distance

tracked

during the

experiment

(km)

Number

of risky

situations

reported

(n)

F Paris 24 Every day 10 40 330 13

M Paris 44 Every day 20 45 200 15

F Paris 33 Every day 10 30 180 16

M Paris 48 5 or 6 days 60 170 480 27

M Paris 49 5 or 6 days 46 150 960 18

M Paris 37 5 or 6 days 49 110 920 18

M Paris 26 Every day 25 90 510 13

M Paris 43 5 or 6 days 15 60 310 32

F Paris 42 5 or 6 days 26 90 520 15

M Paris 27 5 or 6 days 20 50 450 20

F Lyon 30 Every day 20 70 440 15

F Lyon 29 4 or 5 days 3 15 60 10

M Lyon 63 Every day 20 90 560 10

M Lyon 35 Every day 15 45 280 6

M Lyon 26 Every day 15 50 300 15

F Lyon 51 4 or 5 days 20 60 450 14

F Lyon 39 4 or 5 days 5 15 80 13

M Lyon 41 4 or 5 days 5 15 80 10

M Lyon 38 4 or 5 days 5 15 65 6

M Lyon 41 5 or 6 days 3 15 20 1

M 38.3 5.9 19.6 61.3 359.8 14.4

SD 9.7 1 15.3 43.3 256.9 6.8

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 192 / 240

193Identication of pre-accident situations faced by cyclists

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 193 / 240

I.3.2. Logbook

Participants were provided a logbook and instructed to note all the

situations judged as risky for them in it. Cyclists were also instructed to

complete the logbook at least once a day. For each situation reported on

the logbook cyclists had to detail a title, the date, time and travel starting

and end point. To that information, a drawing of the situation completed

with a short description was asked. The main objective of the logbook was

to identify the risky situations among all cyclists’ video recorded travels

and to guide the ensuing analysis.

I.3.3. Self-confrontation interviews

Self-confrontation interviews were run in face-to-face with every

participant individually at the end of each experimental week. All

participants faced four interviews of about half an hour each (depending

on the number of risky situations logged). Those interviews have focused

on (1) a detailed description of the risky situations reported in the logbook

and (2) self-confrontation based on the videos recordings of those risky

situations.

In practice, the cyclist and the experimenter looked at the computer

screen where the different risky situations reported by the cyclist were

played. The experimenter paused the recording as many times as required in

order to let the cyclist put words on the activity carried out in the presented

video. The principle of this method is to facilitate participants’ description

of their own activity by reconstructing step by step, the perceptive and

decision-making processes implied in the recorded situation (Theureau,

2003, 2010; Vermersch, 1994).

When cyclists commented videos, the experimenter encouraged them

to focus on the watched situation when necessary. The idea was to keep

the cyclist in the dynamic context of the situation previously experienced

so as to reach physical and cognitive actions actually performed during the

recorded situations. These techniques allow the exploration of both explicit

and implicit aspects of physical or mental actions. The interviewer guided

cyclists to put their experience into words. This method aimed at collecting

verbal description on feelings, sensations, internal states, and thoughts that

were experienced by the cyclist for each situation considered (Aupetit,

Gallier, Riff, Espié, & Delgehier, 2016; Aupetit et al., 2013; Vermersch,

1994).

i.4. pr o c e D u r e

Cyclists were recruited via advertisement on social networks and

cycling associations. Before inclusion in the experiment, all volunteers

completed a short questionnaire about cycling habits (see collected data

in Table 1). Only cyclists riding their bike 4 or 5 time a week or more

were included.

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

194 Jordan Navarro et al.

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 194 / 240

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 195 / 240

After inclusion, the experimenter met each cyclist face to face to present

the experiment and the material (i.e. video camera and logbook use). Each

cyclist was then ready to start the experiment itself for four consecutive

weeks.

An individual meeting was set every week so as to run self-confrontation

interviews. All participants took part to four self-confrontation interviews,

except the last participant presented in table 1 who stopped the experiment

after week 1.

i.5. DA t A A n A l y s i s

The type of pavement cyclists were riding on when the pre-accident

situations occurred was analysed in terms of frequency. The presence and

type of other road users implied in the pre-accident situations was also

analysed in terms of frequency.

Similar pre-accident situations were regrouped so as to offer a

categorization of the reported situations and their relative frequencies. Two

experimenters performed the categorization process independently. The

categorizations obtained were then compared and common decision made

in case of disagreement. This categorization process was conducted once

including all reported situations and once including only the situations

where the visibility of the cyclist was engaged based on videos and

interviews.

Pearson’s chi-squared tests were used for statistic analysis. The level of

signicance of p<0.05 was used.

II. RESULTS

All in all, 287 pre-accident situations were collected (187 in Paris and

100 in Lyon).

ii.1. ty p e o f p A v e M e n t i n t h e p r e -A c c i D e n t s i t u A t i o n s

The different types of pavement cyclists were riding on when pre-

accident situations occurred were regrouped in seven categories listed in

Figure 1. Those seven different types of pavement observed were not

equally distributed (Paris: c2 (6) = 449.92, p<0.001; Lyon: c2 (6) =208.42

p<0.001). The distributions observed in Paris and in Lyon were not

signicantly different (Pearson c2 (30) = 35, p=0.242). For both cities,

most of the pre-accident situations happened on regular shared pavement,

which represent more than half of the situations (M = 52%), much more

than off-road cycle paths (M = 15.5%) and bike lanes (M = 14%) and even

more than bus lanes (M = 8%), cycle path (M = 5.5%), contraow bike

lanes (M = 3.5%) and sidewalks (M = 1.5%).

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 194 / 240

195Identication of pre-accident situations faced by cyclists

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 195 / 240

Figure 1. Type of pavement cyclists were riding

on when pre-accident situations occurred in Paris and Lyon urban areas.

Percentage (%) of the number of situations reported.

Figure 1. Type de chaussée sur laquelle se trouvaient les cyclistes

lors des situations de pré-accident à Paris et Lyon.

Pourcentage (%) du nombre de situations reportées.

ii.2. ot h e r r o A D u s e r s i M p l i e D i n t h e p r e -A c c i D e n t s i t u A t i o n s

The other road users implied in pre-accident situations occurred were

regrouped in seven categories in

Figure 2. Other road users implied were not equally distributed (Paris:

c2 (6) = 318.33, p<0.001; Lyon: c2 (6) = 172.02, p<0.001). The distributions

observed in Paris and in Lyon were not signicantly different (Pearson

c2 (20) = 22.75, p=0.301). On the average car drivers were implied in more

than half of the pre-accident situations (M = 54.5%) and pedestrians in

about a quarter (M = 25.5%). The remaining fth of the situations was

spread between cyclists, truck and bus drivers, powered two-wheelers and

no other road user implication.

Figure 2. Other road users implied in the pre-accident situations reported in Paris and Lyon

urban areas. Percentage (%) of the number of situations reported.

Figure 2. Autres usagers de la route impliqués dans les situations de pré-accident à Paris et Lyon.

Pourcentage (%) du nombre de situations reportées.

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

196 Jordan Navarro et al.

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 196 / 240

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 197 / 240

ii.3. cA t e g o r i z A t i o n o f r e p o r t e D p r e -A c c i D e n t s i t u A t i o n s

II.3.1. All situations included

Fourteen different pre-accident categories were obtained by pooling

similar situations (see Table 2). Those fourteen categories were not equally

distributed (Paris: c2 (13) = 212.52, p<0.001; Lyon: c2 (13) = 37.07,

p<0.001). The distributions observed were not signicantly different

between Paris and Lyon (Pearson c2 (54) = 55.61, p=0.414). If all data are

available in Table 2, only the six categories that reached more than 5 % of

all reported situations on the average between the two urban areas will be

presented in detail. Among those six categories, very similar frequencies

were obtained for half of pre-accident situations and descriptive differences

must be mentioned for the other half.

tA B l e 2. Categorisation of all pre-accident situations reported ordered

by frequency.

tA B l e A u 2. Catégorisation de l'ensemble des situations de pré-accident rapportées

classées de la plus fréquente à la moins fréquente.

Category

number Category name Paris Lyon Mean

Pedestrian crossings or obstacles on bicycle facilities

32 % 33 % 32.5 %

2 Vehicle entering or leaving parking 3% 25% 14%

3 Vehicle turning right stepping into cyclists’ way 11% 5% 8%

4Vehicle coming from the right and stepping

into cyclists’ way 7% 8% 7.5%

5 Vehicle squeezing cyclists while overtaking 7% 8% 7.5%

6Other road users changing direction while

riding between lanes in trafc jams 11 % - 5.5 %

7 Conict with unpowered two-wheelers - 9 % 4.5 %

8 Opening doors of parked vehicles 4 % 4 % 4 %

9 Unadapted infrastructure - 8 % 4 %

Conict with other road users while crossings lanes

8 % - 4 %

11 Oncoming vehicle turning left stepping into

cyclists way 5 % - 2.5 %

12 Observed cyclist errors and violations 4% - 2 %

13 Conict with an oncoming road user in

contraow bike lane 4 % - 2 %

14 Conict in shared lane with bus or taxi 4% - 2 %

Categories 1, 4 and 5 appeared as similar between Paris and Lyon.

Pedestrian crossings or obstacles on bicycle facilities (category 1) were the

most frequently observed situations with about a third of the situations

reported. Categories 4 and 5 (i.e. vehicle coming from the right and

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 196 / 240

197Identication of pre-accident situations faced by cyclists

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 197 / 240

stepping into cyclists’ way and vehicle squeezing cyclists while overtaking

respectively) were much less frequent (7.5 % on the average). Even if

categories number 2, 3 and 6 descriptively appeared as quite different

between Paris and Lyon, no signicant differences were observed.

II.3.2. Situations with cyclists’ visibility engaged

151 (96 in Paris and 55 in Lyon) pre-accident situations among

the 287 collected were identied as related to cyclists’ visibility during

interviews. It is not an objective visibility measure, but rather cyclists’

point of view of their own visibility on the situations reported. When it

was explicit in the self-confrontation interviews that at least a contributing

factor to the pre-accident situation was due to a lack of cyclist visibility for

the other road user, the situation was classied as a “situation with cyclists’

visibility engaged”. The same category numbers were used.

Thirteen of the fourteen pre-accident categorises previously

described were found when focusing only on situations where cyclists’

visibility was engaged (see Table 3). Those thirteen categories were not

equally distributed (Paris: c2 (12) = 143.24, p<0.001; Lyon: c2 (12)

31.84, p<0.002) and the distribution observed was not signicantly

different between Paris and Lyon (Pearson c2 (45) = 55.25, p=0.141).

tA B l e 3. Categorisation of pre-accident situations reported with cyclists

visibility engaged ordered by frequency.

tA B l e A u 3. Catégorisation des situations de pré-accident rapportées associées à un défaut

de visibilité des cyclistes et classées de la plus fréquente à la moins fréquente.

Category

number

Category name Paris Lyon Mean

Pedestrian crossings or obstacles on bicycle facilities

37 % 42 % 39,5 %

2 Vehicle entering or leaving parking 4% 36% 20,0%

3Vehicle turning right stepping into cyclists’ way 18% 4% 11,0%

7 Conict with unpowered two-wheelers - 13 % 6,5 %

8 Opening doors of parked vehicles 6 % 5 % 5,5 %

6Other road users changing direction while

riding between lanes in trafc jams 10 % - 5,0 %

4Vehicle coming from the right and steeping into

cyclists’ way 8 % - 4 %

5 Vehicle squeezing cyclists while overtaking 4% - 2,0 %

Conict with other road users while crossings lanes

4 % - 2,0 %

13 Conict with an oncoming road user in

contraow bike lane 3 % - 1,5 %

14 Conict in shared lane with bus or taxi 3% - 1,5 %

11 Oncoming vehicle turning left stepping into

cyclists way 2 % - 1,0 %

12 Observed cyclist errors and violations 1% - 0,5 %

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

198 Jordan Navarro et al.

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 198 / 240

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 199 / 240

Compared to all pre-accident situations, the same three rst categories in

term of average relative frequencies were found in the same order. If all

data are available in ween Paris and Lyon, no signicant differences were

observed.

Table 3, only the ve categories that reached more than 5 % of reported

situations on the average between the two urban areas will be presented in

detail. Among those ve categories, very similar frequencies were obtained

in Paris and Lyon for two of pre-accident situations and descriptive

differences must be mentioned for the three others.

Categories 1 and 8 appeared as very similar between Paris and Lyon.

Pedestrian crossings or obstacles on bicycle facilities (category 1) were

the most frequently observed situations with about 40 % of the situations

reported. Opening doors of parked vehicles (category 8) was also found in

both cities but much less frequently (5.5 %). Even if categories number 2,

3 and 7 appeared as descriptively quite different between Paris and Lyon,

no signicant differences were observed.

III. DISCUSSION

When looking at all pre-accident situations reported by cyclists’, it

appears that a third of the risky situations collected implied interactions

with pedestrians or obstacles located on bicycle facilities. The following

most frequent situations were related to interactions with motorists either

entering or leaving parking, stepping into cyclists’ way while turning right

or coming from the right at an intersection, or motorists squeezing cyclists

while overtaking. When focusing on situations where cyclists’ visibility was

engaged similar most frequent situations were observed except for vehicle

coming from the right at an intersection and steeping into cyclists’ way

and vehicle squeezing cyclists while overtaking that appeared related as

motorists deliberate behaviours. Instead, two other situations were found

quite frequent, conict between unpowered two-wheelers and opening

doors of parked vehicles. Whatever the agglomeration considered, more

than half of the pre-accident situations were located on regular shared

pavement and more than half of those situations implied a car driver, and

about a quarter pedestrians.

A more detailed presentation of the most frequently collected situations

will help to understand precisely what happened in those situations. The

main situation reported by cyclists in the current study was related to

pedestrian crossings, and not to other motorists. To our knowledge, no study

specically focused on cyclists/pedestrian conicts, possibly because their

number is under-estimated due to a lack of report by public forces and

because their physical and economical consequences are not as important as

with motorists/cyclists accidents. Another possible explanation is that many

pre-accident cyclists/pedestrians occur but only few accidents results from

these situations. This would be explained by relatively low displacement speed

compared to powered vehicles, as a consequence both cyclists and pedestrian

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 198 / 240

199Identication of pre-accident situations faced by cyclists

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 199 / 240

can slow down and change directions quickly. Pedestrians’ crossings were

regrouped with obstacles on bicycle facilities in the same category (category

number 1). Obstacles located on the bicycle facilities were added to that

category because they implied similar cyclists’ behaviours: cyclists had to

skirt and/or brake. Pedestrian crossings represent most of the situations

reported in this category (about 80 %). In many occasions the pedestrians

did not paid attention to cyclists’ presence because interacting with their

phone or seeking for noisy vehicles only for instance. Almost all the obstacles

were vehicles moving or parked on bicycle facilities, others obstacles consist

in objects on bicycle facilities (e.g. tree branches or trashes).

The second most frequent situation is related to a vehicle entering or

leaving parking. In most of those situations, the motorist did not perceive

the cyclist because his/her attention was focused on the parking manoeuver

or on its insertion in the trafc. Possibly due to a greater number of parking

possibilities in Lyon, the relative frequency of this category was higher in

Lyon than in Paris. The third most frequent situation occurred when the

motorist turned right stepping into cyclist way at the right of the vehicle,

a situation already analysed (Räsänen & Summala, 1998; Summala et al.,

1996). The relative frequency of this situation increased when looking

only to the pre-accident situations where cyclist visibility was engaged

(from 8 % to 11 %), indicating that visibility is indeed a key factor in that

context. However, in several occasions participants declared that motorists

perceived the cyclist presence but forced their way obliging the cyclist to

brake.

To provide a more detailed view on the sources of the pre-accident

situations the different situations were classied according to the different

causes identied in the literature (Table 4). As all the reported data, this

classication is based on cyclists’ subjective description of the considered

situations. A high proportion of cyclists crashes seems to be related to drivers

failure to detect cyclists (Wood et al., 2009). This failure might be due to:

(a) the fact that drivers or other road users look in the direction of the cyclist

but do not perceive it (“looked-but-failed-to-see phenomenon” ; Brown,

2005) and therefore related to a lack of cyclists sensorial conspicuity (Engel,

1971, 1974; Wulf et al., 1989) (b) a lack of cyclist appearance expectation

from drivers or other road user perspective, because biking in urban areas is

relatively recent (Clarke et al., 2007), and therefore related to a problem of

cognitive conspicuity (Hancock et al., 1990). Based on cyclists’ perspective

only it was not possible to differentiate sensorial and cognitive conspicuity,

consequently those two causes were regrouped under the same category

referred as other road users failure to detect cyclists (Table 4 second

column). To those other road users failures, pre-accident situations could

also be caused by: (c) cyclists inaccurate expectations, such as drivers yielding

(Räsänen & Summala, 1998) and (d) cyclists optimism regarding the risk

of accident. Based on cyclists’ self-confrontations analysis it appeared that

those two accident causes were linked and therefore hardly differentiable,

consequently they were regrouped (Table 4 third column). Finally in

addition to the previous four categories documented in the literature, a fth

cause appeared in the current study (e) motorists’ or pedestrians’ deliberate

behaviours (Table 4 fourth column). That is to say motorists or pedestrians

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

200 Jordan Navarro et al.

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 200 / 240

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 201 / 240

that did perceive correctly the cyclist, should have given priority to the

cyclist according to the Highway Code and had the capacity to do so, but

deliberately engaged a displacement that put the cyclist at risk.

Slightly more than half of the pre-accident situations were linked

to cyclists not being correctly perceived (sensorial and or cognitive

conspicuity). Surprisingly, motorists’ and occasionally pedestrians’

deliberate behaviour caused about a third of the pre-accident situations. In

those situations, motorists’ break the rules and outing common courtesy

to gain way. The remaining situations were due to cyclists’ inaccurate

expectations and/or over-optimism. Our data are in line with previously

reported results (Wood et al., 2009). A high proportion (more than half in

the current study) of cyclists’ pre-accidents are related to motorists’ failure

to detect cyclists. Additionally our data point out motorists’ deliberate

behaviours to gain way over cyclists as another important source of pre-

accident situations. Along with cyclists’ inaccurate expectations and excess

of optimism, pre-accident situations seem to be related to difculties in

sharing the road.

tA B l e 4. Repartition of pre-accident situations reported related to:

motorists’ failure to detect cyclists (sensorial conspicuity (a) and/or

cognitive conspicuity (b)), cyclists inaccurate expectations (c) and over-

optimism (d) or motorists’ or pedestrians’ deliberate behaviours (e).

tA B l e A u 4. Répartition des situations de pré-accident rapportées selon : l'incapacité

des automobilistes à détecter les cyclistes (conspicuité sensorielle (a) et/ou conspicuité

cognitive (b)), les attentes erronées des cyclistes (c) et un excès d'optimisme (d), ou un

comportement délibéré des automobilistes ou piétons (e).

Location

Sensorial

conspicuity (a)

and/or cognitive

conspicuity (b)

Cyclist inaccurate

expectations (c) and/

or over-optimism (d)

Motorist’s or

pedestrians’

deliberate

behaviours (e)

Paris 52 % 15 % 33 %

Lyon 55 % 13 % 32 %

Mean 53,5 % 14 % 32,5 %

A more detailed category per category presentation of these results

offers more precise insights so as to improve safety depending on the

context. Only the pre-accident categories of 15 or more occurrences (more

that 5 %) are presented in Table 5. Those categories (from 1 to 6) represent

75 % of all reported pre-accident situations. All categories but number 6

present an unbalanced repartition with more than 60 % of pre-accident

situations related to (1) sensorial and/or cognitive conspicuity for cate-

gories 1, 2 and 3, or (2) motorists’/pedestrians’ behaviours for categories

4 and 5. Cyclists’ conspicuity is a critical issue involved in an average of

69 % of pre-accidents categories 1, 2 and 3. Those three categories repre-

senting themselves about 55 % of all pre-accident situations collected. So

as to improve noticeably cyclists’ safety in those situations, conspicuity

must be improved. Sensorial conspicuity may be improved with the use of

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 200 / 240

201Identication of pre-accident situations faced by cyclists

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 201 / 240

reecting gear for biking for instance. Besides cognitive conspicuity should

be improved by prevention campaigns to inform other road users in parti-

cular motorists’ while falling in categories 2 and 3 and pedestrians while

crossing roads. Some work should also be undertaken in order to change

motorists’ and pedestrians’ attitudes toward cyclists to reduce the amount

of deliberate behaviours that sometimes cause pre-accident situations in

particular while overtaking cyclists or by stepping into cyclists’ path (cate-

gories 4 and 5). A way to do so is to improve motorists’ road rule know-

ledge. Indeed low level of road rule knowledge is associated with poor

attitudes toward cyclists (Rissel, Campbell, Ashley, & Jackson, 2002). Pre-

accidents occurring while riding between lanes in trafc jam are not parti-

cularly related to a specic cause identied.

tA B l e 5. Category by category repartition of all pre-accident situations

reported related to: motorists’ failure to detect cyclists (sensorial and/

or cognitive conspicuity), cyclists inaccurate expectations and over-

optimism, or motorists’ or pedestrians’ deliberate behaviours.

tA B l e A u 5. Répartition catégorie par catégorie de l'ensemble des situations de pré-

accident rapportées selon : l'incapacité des automobilistes à détecter les cyclistes (conspi-

cuité sensorielle et/ou conspicuité cognitive), les attentes erronées des cyclistes et un excès

d'optimisme, ou un comportement délibéré des automobilistes ou piétons.

Category

number Category name

Sensorial

and/or

cognitive

conspicuity

Cyclist

inaccurate

expectations

and/or over-

optimism

Motorists’

pedestrians’

deliberate

behaviours

Category

weight /

all pre-

accident

situations

Pedestrian

crossings or

obstacles on

bicycle facilities

64 % 16.5 % 19.5 % 32.5 %

2Vehicle entering

or leaving parking 80% 0% 20% 14%

3Vehicle turning

right stepping

into cyclists’ way 62.5% 0% 37.5% 8%

Vehicle coming

from the right

and stepping into

cyclists’ way

28.5% 7% 64.5% 7.5%

5Vehicle squeezing

cyclists while

overtaking 15.5% 0% 84.5% 7.5%

Other road

users changing

direction while

riding between

lanes in trafc jams

45 % 30 % 25 % 5.5 %

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

202 Jordan Navarro et al.

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 202 / 240

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 203 / 240

Using a survey, differences in perceived safety of mixed trafc infra-

structure layouts and fear of trafc were found between different cities

located in different countries with different cycling habits, namely

Brisbane and Copenhagen (Chataway et al., 2014). In the current

study, no signicant differences were observed between Paris and Lyon.

Possibly because of the relatively small number of participants involved

to compare the two French towns or because French cyclists, drivers,

their habits and the infrastructure available in the two French towns

are not different enough. Some descriptive differences in the variety of

the situations recorded, more important in Paris than in Lyon, and fre-

quencies in a few situations tend to indicate that differences might exists.

Further experiments focusing on that point particularly are required in

order to conclude.

About half of the pre-accident situations happened on regular shared

pavement and the other half was distributed across infrastructures dedica-

ted to unpowered road users: off-road cycle paths (15.5 %) and bike lanes

(14 %), bus lanes (8 %), cycle path (5.5 %), contraow bike lanes (3.5 %)

and sidewalks (1.5 %). Although less present on urban areas than regular

shared pavement, it appeared that specic cycle equipment are regularly

used by cyclists and are a location where pre-accident situations are almost

as frequent as regular pavement. As a consequence specic attention should

be paid to those relatively new and expanding road pavements.

Regarding the type of road users involved in reported pre-accident situa-

tions, car drivers were implied in more than half of the pre-accident

situations (54.5 %) and pedestrians in about a quarter (25.5 %). Those two

road users’ types were clearly identied as the main sources of pre-accident

situations. If a high proportion of situations with drivers was expected,

the proportion of situations implying pedestrians was unexpectedly high.

This might be due to the fact that during the last decades, pedestrians

were more and more protected and they have now priority in most of the

situations and location. In addition pedestrian often choose to walk on bike

lane or cycle paths increasing potential conicts with cyclists.

No crash was recorded during the experiment, and the results collec-

ted only took into account cyclists’ perception of the situations. Obviously

pre-accident situations are more numerous than actual accidents (Sanders,

2015). Identity pre-accidents situations is highly valuable because it may

help to improve safety in a different way that accident analysis. However

pre-accidents might not necessarily be representative of accidents. In addi-

tion, cyclists known the journeys were recorded by the video camera. As

a consequence, it can not be excluded that cyclists might have changed

some of their usual behaviours during the experiment. Another weakness

of the methodology used is that experimenters had no control on cyclists’

activity, the ways they choose, speed, time of the time and so on. The situa-

tions collected should now be tested in more controlled conditions so as

to understand precisely the underpinning mechanisms. The evaluation of

several other variables would also be of particular interest in order to better

characterise pre-accident situations. In particular, the inuence cyclists’

age, gender and expertise as well as the season of the year and weather of

the day might impact on pre-accident situation reported. Another inter-

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 202 / 240

203Identication of pre-accident situations faced by cyclists

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 203 / 240

esting perspective in order to complement the reported data will be to

consider pre-accident situations involving cyclists’ and their visibility from

motorist and/or pedestrian perspectives.

IV. CONCLUSION

The data collected offered a new perspective on cyclists’ safety on

French agglomerations. Cyclists’ increasing number and the development

of specic infrastructures comes with new challenges in terms of road

safety, including risky situations involving pedestrians and cyclists

visibility. About half of the situations analysed in the experiment were

related to a problem of cyclists’ visibility. This issue might be related to a

weak cyclists expectance from motorists’ and pedestrians’ perspectives. In

addition drivers-cyclists are known to be much more likely than drivers

only to report safe driving behaviours related to sharing the roads with

cyclists and to have a better knowledge of road rules related to cycling

infrastructure (Johnson, Oxley, Newstead, & Charlton, 2014). Therefore,

cyclists’ individual differences in the pre-accident situations should be

further analysed in future experiments. Another perspective to improve

cyclists’ safety would be to identify the most risky areas as performed in

Stockholm for instance (Gustafsson & Archer, 2013). More immediately,

the pre-accident scenarios identied in this naturalistic study would be

used to conduct simulated safe and well-controlled experiments in order to

understand even more precisely why the situations reported occurred.

AcknowleDgeMents

This study was supported by the French association “Fondation

Sécurité Routière” (VISIBLE project).

REFERENCES

Amoros, E., Chiron, M., Thélot, B., & Laumon, B. (2011). The injury epidemiology

of cyclists based on a road trauma registry. BMC Public Health, 11(1), 653.

Amoros, E., Martin, J.-L., & Laumon, B. (2006). Under-reporting of road crash

casualties in France. Accident Analysis & Prevention, 38(4), 627–635.

Aultman-Hall, L., & Hall, F. L. (1998). Ottawa-Carleton commuter cyclist on-and

off-road incident rates. Accident Analysis & Prevention, 30(1), 29–43. http://doi.

org/10.1016/S0001-4575(97)00059-6

Aultman-Hall, L., & Kaltenecker, M. G. (1999). Toronto bicycle commuter safety

rates. Accident Analysis & Prevention, 31(6), 675–686. http://doi.org/10.1016/

S0001-4575(99)00028-7

Aupetit, S., Gallier, V., Riff, J., Espié, S., & Delgehier, F. (2016). Naturalistic study

of the risky situations faced by novice riders. Ergonomics, 59(8), 1109–1120.

http://doi.org/10.1080/00140139.2015.1120887

Aupetit, S., Riff, J., Buttelli, O., & Espié, S. (2013). Naturalistic study of rider’s

behaviour in initial training in France: Evidence of limitations in the educational

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

204 Jordan Navarro et al.

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 204 / 240

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 205 / 240

content. Accident Analysis and Prevention, 58, 206–217. http://doi.org/10.1016/j.

aap.2012.09.036

Brown, I. D. (2005). Review of the “Looked but Failed to See” Accident Causation

Factor (No. 60). Department for transport, London, UK.

Chataway, E. S., Kaplan, S., Nielsen, T. A. S., & Prato, C. G. (2014). Safety

perceptions and reported behavior related to cycling in mixed trafﬁc: A

comparison between Brisbane and Copenhagen. Transportation Research

Part F: Trafﬁc Psychology and Behaviour, 23, 32–43. http://doi.org/10.1016/j.

trf.2013.12.021

Chaurand, N., & Delhomme, P. (2013). Cyclists and drivers in road interactions:

A comparison of perceived crash risk. Accident; Analysis and Prevention, 50,

1176–84. http://doi.org/10.1016/j.aap.2012.09.005

Clarke, D. D., Ward, P., Bartle, C., & Truman, W. (2007). The role of motorcyclist

and other driver behaviour in two types of serious accident in the UK.

Accident; Analysis and Prevention, 39(5), 974–81. http://doi.org/10.1016/j.

aap.2007.01.002

de Lapparent, M. (2005). Individual cyclists’ probability distributions of severe/

fatal crashes in large French urban areas. Accident; Analysis and Prevention,

37(6), 1086–92. http://doi.org/10.1016/j.aap.2005.06.006

Engel, F. (1971). Visual conspicuity, directed attention and retinal locus. Visison

Research, (11), 563–576.

Engel, F. (1974). Visual conspicuity and selective background interference in

eccentric vision. Vision Research, (14), 459–471.

Ferrandez, F. (1995). L’ étude détaillée d’accidents orientée vers la sécurité primaire:

méthodologie de recueil et de pré-analyse. Paris: Presses de L’E.N.C.T.

Forester, J. (1994). Bicycle Transportation: A Handbook for Cycling Transportation

Engineers. Cambridge, MA, 2nd ed., MIT Press.ISBN 978-0-262-56079-5.

Gustafsson, L., & Archer, J. (2013). A naturalistic study of commuter cyclists in the

greater Stockholm area. Accident; Analysis and Prevention, 58, 286–98. http://

doi.org/10.1016/j.aap.2012.06.004

Hancock, P., Wulf, G., Thom, D., & Fassnacht, P. (1990). Driver workload during

differing driving maneuvers. Accident Analysis & Prevention, 22(3), 281–290.

Johnson, M., Oxley, J., Newstead, S., & Charlton, J. (2014). Safety in numbers?

Investigating Australian driver behaviour, knowledge and attitudes towards

cyclists. Accident; Analysis and Prevention, 70, 148–54. http://doi.org/10.1016/j.

aap.2014.02.010

King, M. J., Wood, J. M., Lacherez, P. F., & Marszalek, R. P. (2012). Optimism

about safety and group-serving interpretations of safety among pedestrians

and cyclists in relation to road use in general and under low light conditions.

Accident; Analysis and Prevention, 44(1), 154–9. http://doi.org/10.1016/j.

aap.2010.12.003

Lacherez, P. F., Wood, J. M., Marszalek, R. P., & King, M. J. (2013). Visibility-related

characteristics of crashes involving bicyclists and motor vehicles – Responses

from an online questionnaire study. Transportation Research Part F: Trafﬁc

Psychology and Behaviour, 20, 52–58. http://doi.org/10.1016/j.trf.2013.04.003

Minikel, E. (2012). Cyclist safety on bicycle boulevards and parallel arterial routes

in Berkeley, California. Accident; Analysis and Prevention, 45, 241–7. http://doi.

org/10.1016/j.aap.2011.07.009

Ombredane, A., & Faverge, J. (1955). L’analyse du travail (PUF, Paris).

Papon, F., & De Solère, R. (2010). Les modes actifs: marche et vélo de retour

en ville. La Revue, Commissariat Général Au Développement Durable-Service de

L’observation et Des Statistiques.

Räsänen, M., Koivisto, I., & Summala, H. (1999). Car Driver and Bicyclist Behavior

at Bicycle Crossings Under Different Priority Regulations. Journal of Safety

Research, 30(1), 67–77. http://doi.org/10.1016/S0022-4375(98)00062-0

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 204 / 240

205Identication of pre-accident situations faced by cyclists

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 205 / 240

Räsänen, M., & Summala, H. (1998). Attention and expectation problems in

bicycle-car collisions: an in-depth study. Accident; Analysis and Prevention, 30(5),

657–66. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/9678219

Rissel, C., Campbell, F., Ashley, B., & Jackson, L. (2002). Driver road rule

knowledge and attitudes towards cyclists. Australian Journal of Primary Health,

8(2), 66–69.

Sanders, R. L. (2015). Perceived trafﬁc risk for cyclists: The impact of near miss

and collision experiences. Accident; Analysis and Prevention, 75C, 26–34. http://

doi.org/10.1016/j.aap.2014.11.004

Summala, H., Pasanen, E., Räsänen, M., & Sievänen, J. (1996). Bicycle accidents

and drivers’ visual search at left and right turns. Accident Analysis & Prevention,

28(2), 147–153. http://doi.org/10.1016/0001-4575(95)00041-0

Theureau, J. (2003). Course-of-action analysis and course-of-action centered

design. Handbook of Cognitive Task Design, 55–81.

Theureau, J. (2010). Les entretiens d’autoconfrontation et de remise en situation

par les traces matérielles et le programme de recherche «cours d’action». Revue

D’anthropologie Des Connaissances, 42(2), 287–322.

Vermersch, P. (1994). L’entretien d’explicitation. Paris : ESF.

Wood, J. M., Lacherez, P. F., Marszalek, R. P., & King, M. J. (2009). Drivers’ and

cyclists’ experiences of sharing the road: incidents, attitudes and perceptions of

visibility. Accident Analysis & Prevention, 41(4), 772–6. http://doi.org/10.1016/j.

aap.2009.03.014

Wood, J. M., Tyrrell, R., Marszalek, R. P., Lacherez, P. F., Carberry, T., Chu, B.,

& King, M. J. (2010). Cyclist visibility at night: Perceptions of visibility do not

necessarily match reality. Journal of the Australasian College of Road Safety, 21,

56–60. Retrieved from http://eprints.qut.edu.au/38338

Wulf, G., Hancock, P., & Rahimic, M. (1989). Motorcycle conspicuity: an evaluation

and synthesis of inﬂuential factors. Journal of Safety Research, (20), 153–176.

Manuscrit reçu : mars 2016.

Accepté après révision par F. Darses : décembre 2016.

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 206 / 240

31 mai 2017 04:03 - Le travail humain n°2/2017 - Collectif - Le travail humain - 155 x 240 - page 207 / 240

Document téléchargé depuis www.cairn.info - Université Louis Lumière Lyon 2 - - 159.84.74.48 - 17/07/2017 12h00. © Presses Universitaires de France

Mechanisms underlying cognitive conspicuity in the detection of cyclists by car drivers

Article

Jul 2017

Objective: The aim of this study was to evaluate the visibility of cyclists for motorists in a simulated car driving task. Background: In several cases involving collisions between cars and cyclists, car drivers failed to detect the latter in time to avoid collision because of their low conspicuity. Method: 2 groups of motorists (29.2 years old), including 12 cyclist-motorists and 13 non-cyclist-motorists, performed a vulnerable road user detection task in a car-driving simulator. They had to detect cyclists and pedestrians in an urban setting and evaluate the realism of the cyclists, the traffic, the city, the infrastructure, the car driven and the situations. Cyclists appeared in critical situations derived from previous accounts given by injured cyclists and from cyclists' observations in real-life situations. Cyclist's levels of visibility for car drivers were either high or low in these situations according to the cyclists. Results: Realism scores were similar and high in both groups. Cyclist-motorists had fewer collisions with cyclists and detected cyclists at a greater distance in all situations, irrespective of cyclist visibility. Several mechanisms underlying the cognitive conspicuity of cyclists for car drivers were considered. Conclusion: The attentional selection of a cyclist in the road environment during car driving depends on top-down processing. Application: We consider the practical implications of these results for the safety of vulnerable road users and future directions of research.

A SYSTEMATIC REVIEW ON MOTORCYCLISTS' AGGRESSIVE BEHAVIOR ANALYSIS USING COMPUTATIONAL MODELS: CURRENT STATE, CHALLENGES, AND RECOMMENDATIONS

Article

Full-text available

Aug 2022

There is an increase in motorcycles traffic accidents, while the cause for such accidents has always been associated with aggressive driving behaviors. There has been considerable research attention on how to deal with such driving behavior that causes severe and fatal accidents from the academic perspective; these research works addressed technical, scientific, and social issues. This study systematically searches, reviews, and analyzes the literature associated with motorcycle accidents and driving behaviors. Between the years 2014 and 2021, the next four databases have been searched: ScienceDirect, Scopus, Web of Science, and IEEE Xplore. A total of 108 people were picked depending on certain inclusion and exclusion criteria. Approximately 68% (n=79/108) of the researchers looked at the challenges from a social science perspective, whereas 25% (n=26/108) concentrated on experimental research variables. Only 7% (=3/108) explored the development of Apps & systems. Finally, our contribution comprehensively analyses most of the articles by highlighting challenges associated with motorcycle behavior, motivations, and recommendations. In addition, provide potential research gaps in current studies that require further investigation.

A Comprehensive Review on the Behaviour of Motorcyclists: Motivations, Issues, Challenges, Substantial Analysis and Recommendations

Article

Full-text available

Mar 2022
Int J Environ Res Publ Health

With the continuous emergence of new technologies and the adaptation of smart systems in transportation, motorcyclist driving behaviour plays an important role in the transition towards intelligent transportation systems (ITS). Studying motorcyclist driving behaviour requires accurate models with accurate and complete datasets for better road safety and traffic management. As accuracy is needed in modelling, motorcyclist driving behaviour analyses can be performed using sensors that collect driving behaviour characteristics during real-time experiments. This review article systematically investigates the literature on motorcyclist driving behaviour to present many findings related to the issues, problems, challenges, and research gaps that have existed over the last 10 years (2011–2021). A number of digital databases (i.e., IEEE Xplore®, ScienceDirect, Scopus, and Web of Science) were searched and explored to collect reliable peer-reviewed articles. Out of the 2214 collected articles, only 174 articles formed the final set of articles used in the analysis of the motorcyclist research area. The filtration process consisted of two stages that were implemented on the collected articles. Inclusion criteria were the core of the first stage of the filtration process keeping articles only if they were a study or review written in English or were articles that mainly incorporated the driving style of motorcyclists. The second phase of the filtration process is based on more rules for article inclusion. The criteria of inclusion for the second phase of filtration examined the deployment of motorcyclist driver behaviour characterisation procedures using a real-time-based data acquisition system (DAS) or a questionnaire. The final number of articles was divided into three main groups: reviews (7/174), experimental studies (41/174), and social studies-based articles (126/174). This taxonomy of the literature was developed to group the literature into articles with similar types of experimental conditions. Recommendation topics are also presented to enable and enhance the pace of the development in this research area. Research gaps are presented by implementing a substantial analysis of the previously proposed methodologies. The analysis mainly identified the gaps in the development of data acquisition systems, model accuracy, and data types incorporated in the proposed models. Finally, research directions towards ITS are provided by exploring key topics necessary in the advancement of this research area.

Bicycle crash contributory factors: A systematic review

Article

Jan 2022
SAFETY SCI

There is a growing body of road safety research that seeks to identify crash contributory factors beyond road users, their vehicles, and the immediate road environment. Although cyclist safety represents a critical research area, this ‘systems thinking’ approach has received less attention in bicycle crash analysis. This article presents the findings from a systematic literature review which aimed to synthesise the peer reviewed literature regarding bicycle crash contributory factors (defined as factors which play a contributory role in bicycle crashes, as opposed to risk factors which are factors which may increase the probability of crashes). Crash contributory factors were extracted from included articles and mapped onto a systems thinking framework comprising seven hierarchical road transport system levels. The findings show that a majority of the included studies identified contributory factors relating to the road environment, cycling infrastructure, and cyclist and driver behaviour. No studies identified contributory factors outside of cyclists and road users, bicycles and vehicles, and the road environment and few specifically examined causal relationships between contributory factors. It is concluded that there are gaps in the knowledge base regarding the broader transport system features that play a role in bicycle crashes and how contributory factors interact to create crashes. We argue that more expansive research into the systemic factors involved in bicycle crashes is required and that initial work should focus on the development of new data sources and analysis methods.

Les entretiens d'autoconfrontation et de remise en situation par les traces matérielles et le programme de recherche « cours d'action »

Article

Full-text available

Sep 2010

Jacques Theureau

SELF-CONFRONTATION INTERVIEWS, RE-SITUATING WITH MATERIAL TRACES, AND THE ‘COURSE OF ACTION’ RESEARCH PROGRAM The development of the course-of-action research program has included both the creation of methods of data construction and the transformation of methods belonging to other research programs. Several aspects of the resulting methodology are still original today. The methods of self-confrontation and re-enactment through material traces of the activity are central. Each of these two methods is developed along two ways, one aiming at the controlled expression of actors’pre-reflective consciousness, the other aiming at both actors’participation in the analysis of their own activity and the pursuit with less control of the expression of actors’pre-reflective consciousness. The article presents the resulting four methods along with the ontological hypotheses and the epistemological hypotheses and principles which command their creation and their development.

Perceived traffic risk for cyclists: The impact of near miss and collision experiences

Article

Full-text available

Nov 2014
ACCIDENT ANAL PREV

Rebecca L. Sanders

Though the percentage of people bicycling for transportation rose during the last decade, with an average increase in bicycle commuting of 47% (Flusche, 2012), still only 1% of all U.S. trips are made by bike (Flusche, 2010). Research suggests that people’s concern regarding the risk of bicycling near traffic—namely the risk of being hit by a car—remain a significant barrier to widespread cycling. However, research has not disaggregated traffic risk to expose its many aspects and how they may affect bicyclists with differing skill levels, experiences, and behaviors. This study begins to address this gap in our understanding. Elaborating on results from an internet survey, this study examined various aspects of traffic risk among 406 potential and current bicyclists in the San Francisco Bay Area. The data indicate that perceived traffic risk negatively influences the decision to bicycle for potential and occasional bicyclists, although the influence decreases with cycling frequency. Additionally, cycling frequency seems to heighten awareness of traffic risk, particularly for cyclists who have experienced “near misses” or collisions. In particular, near misses were found to be (a) much more common than collisions and (b) more strongly associated than collisions with perceived traffic risk. The findings suggest that efforts targeting road user behaviors and roadway designs associated with these near misses could mitigate perceived and actual traffic risk for bicyclists, and thereby eventually help achieve higher cycling ridership.

Course-of-action analysis and course-of-action centered design

Article

Full-text available

Jun 2003

Jacques Theureau

Les modes actifs : marche et vélo de retour en ville

Article

Full-text available

Jan 2010

Francis Papon

Après une très forte baisse de 1982 à 1994, la part de la marche à pied et du vélo dans les déplacements quotidiens des individus se stabilise, marquant ainsi une inflexion de la tendance. La marche à pied représente 22,3 % des déplacements en semaine en 2008, en légère baisse par rapport à 1994. La part du vélo est stable à 2,7 %. Pourtant, le parc de vélos détenus par les ménages résidant en France métropolitaine poursuit sa progression : de 21,3 millions en 1994, le nombre de vélos d’adultes est passé à 26,7 millions en 2008. L’usage du vélo, plus masculin, se modifie : son utilisation augmente dans les villes centres des grandes agglomérations et diminue dans les zones rurales et périurbaines. Il est par ailleurs délaissé par les ouvriers et davantage prisé par les cadres. La marche, plus féminine, est typiquement urbaine. Elle est même devenue majoritaire dans les modes de déplacements parisiens.

L'entretien d'explicitation

Book

Jan 2014

Pierre Vermersch

Bicycle Transportation: A Handbook for Cycling Transportation Engineers

Book

Jan 1994

John Forester

Naturalistic study of the risky situations faced by novice riders

Article

Jan 2016
ERGONOMICS

This article sets out to identify the typical risky situations experienced by novice motorcyclists in the real world just after licensing. The procedure consists of a follow-up of six novices during their first two months of riding with their own motorbike instrumented with cameras. The novices completed logbooks on a daily basis in order to identify the risky situations they encountered, and were given face-to-face interviews to identify the context and their shortcomings during the reported events. Data show a large number of road configurations considered as risky by the riders (248 occurrences), especially during the first two weeks. The results revealed that a lack of hazard perception skills contributed to the majority of these incidents. These situations were grouped together to form clusters of typical incident scenarios on the basis of their similarities. The most frequent scenario corresponds to a lane change in dense traffic (15% of all incidents). The discussion shows how this has enhanced our understanding of novice riders’ behaviour and how the findings can improve training and licensing. Lastly, the main methodological limitations of the study and some guidelines for improving future naturalistic riding studies are presented. Practitioner Summary: This article aims to identify the risky situations of novice motorcyclists in real roads. Two hundred forty-eight events were recorded and 13 incident scenarios identified. Results revealed that a lack of hazard perception contributed to the majority of these events. The most frequent scenario corresponds to a lane change in dense traffic.

L'entretien d'explicitation en formation initiale et en formation continue

Book

Jan 1994

Pierre Vermersch

Safety in numbers? Investigating Australian driver behaviour, knowledge and attitudes towards cyclists

Article

Apr 2014

A key tenet of the safety in numbers theory is that as the number of people cycling increases, more drivers will also be cyclists and therefore will give greater consideration to cyclists when driving. We tested this theory in relation to self-reported behaviour, attitudes and knowledge in relation to cycling. An online survey was conducted of Australian drivers (n=1984) who were also cyclists (cyclist-drivers) and drivers who did not cycle (drivers). Cyclist-drivers were 1.5 times more likely than drivers to report safe driving behaviours related to sharing the roads with cyclists (95% CI: 1.1-1.9, p<0.01). Cyclist-drivers had better knowledge of the road rules related to cycling infrastructure than drivers; however knowledge of road rules related to bike lanes was low for both groups. Drivers were more likely than cyclist-drivers to have negative attitudes (e.g. cyclists are unpredictable and repeatedly overtaking cyclists is frustrating). Findings from this study highlight the need for increased education and awareness in relation to safe driving behaviour, road rules and attitudes towards cyclists. Specific recommendations are made for approaches to improve safety for cyclists.

Safety perceptions and reported behavior related to cycling in mixed traffic: A comparison between Brisbane and Copenhagen

Article

Mar 2014

This study explores the differences in safety perceptions and reported behavior of cyclists in mixed traffic between an emerging cycling city (Brisbane, Australia) and an established cycling city (Copenhagen, Denmark). Perceptions and reported behavior were retrieved from a custom-designed web-based survey administered among cyclists in the two cities. Elicited items concerned perceived risk of infrastructure layouts, fear of traffic, cycling while distracted, use of safety gear, cycling avoidance due to feeling unsafe, and avoidance to cycle in mixed traffic conditions. The data were analyzed with structural equation models. Results show that, in comparison with cyclists in Copenhagen, cyclists in Brisbane perceive mixed traffic infrastructure layouts as less safe, feel more fear of traffic, and are more likely to adopt cycling avoidance as a coping strategy. Results also show that cyclists in Copenhagen tend to use less helmets and to cycle more while being distracted.

Identification of pre-accident situations faced by French cyclists in urban environments: An on-road study

Abstract and Figures

Recommended publications

Road Crash Prediction Models: Different Statistical Modeling Approaches

Theoretical aspects of innovative markets organization in the food industry

Measuring street enclosure and its influence to human physiology through wearable sensors

Exploring the Safety Performance of Tram Roundabouts