P
A
Pede
of p
to th
beha
head
indic
fear
atten
repo
pede
cros
cros
sign
dow
K
J
Cent
SE-1
Swe
www
Pedestria
E
Abstract
estrians are e
edestrians w
heir involvem
aviour: cross
ds down (rath
cator of the (
of falling (F
ntion to cros
orts on a field
estrians in tw
ssing the road
sswalks were
nificant effect
wnward head
Keywords: C
JEL Codes: I
tre for Tra
100 44 Sto
eden
w.cts.kth.s
ans’ beha
Erel Avineri
David Shin
Yusak O.
exposed to ri
was studied as
ment in road
sing speed an
her than tow
(lack of) atte
FOF) among
s traffic, and
d study that c
wo sites (sign
d. The FOF o
e revealed by
ts on crossin
pitches durin
rossing beha
I00, O18, Z0
ansport Stu
ockholm
se
aviour in
fall
– University
nar - Ben Gur
Susilo – Roy
CTS Work
isks when cro
s a factor con
accidents. Th
nd head pitch
ards the traff
ention to cros
pedestrians,
d more attent
combined an
nalised and u
of pedestrian
y means of qu
ng speed, and
ng crossing.
aviour, FOF (
0
udies
cross wa
ling and a
y of the West
rion Univers
yal Institute
king Paper
ossing roads
ntributing to
his work exp
hes—the prop
ffic) when cro
ss-traffic. We
as it might b
tion to the pa
n observatory
unsignalised c
ns and other m
uestionnaire.
d FOF had a
(fear of fallin
alks: The
age
t of England
sity of the Ne
of Technolo
r 2011:18
in urban are
their exposu
plores two sp
portion of tim
ossing a road
e also explor
be associated
avement and
y technique w
crosswalks)
measures of
. Age and ge
significant e
ng), Age
e effects
, Bristol, UK
egev, Israel
gy (KTH)
eas. The cros
ure to risks on
pecific aspect
me pedestria
d. The last on
red the possib
d with slow w
their footstep
with short sur
were video r
pedestrian b
nder had the
ffect on the p
of fear o
K
ssing behavio
n the road an
cts of crossin
ans point thei
ne is used as
ible effect of
walking, less
ps. This pap
urvey. 203
recorded whi
behaviour at
e most
proportion o
of
our
nd
g
ir
an
f
s
er
ile
of
Author's personal copy
AccidentAnalysisandPrevention44 (2012) 30–34
ContentslistsavailableatScienceDirect
Accident
Analysis
and
Prevention
j ou rna l h o me pag e :w w w . e l s e v i e r . c o m / l o c a t e / a a p
Pedestrians’
behaviour
in
cross
walks:
The
effects
of
fear
of
falling
and
age
Erel
Avineri
a,∗, David
Shinar
b, Yusak
O.
Susilo
aaCentreforTransport&Society,UniversityoftheWestofEngland,FrenchayCampus,ColdharbourLane,BristolBS161QY,UK
bDepartmentofIndustrialEngineeringandManagement,BenGurionUniversityoftheNegev,Ben-GurionAvenue1,P.O.Box653,BeerSheva84105,Israel
a
r
t
i
c
l
e
i
n
f
o
Articlehistory: Received14June2010 Receivedinrevisedform 26November2010 Accepted27November2010 Keywords:
Crossingbehaviour FOF(fearoffalling) Age
a
b
s
t
r
a
c
t
Pedestriansareexposedtoriskswhencrossingroadsinurbanareas.Thecrossingbehaviourofpedestrians wasstudiedasafactorcontributingtotheirexposuretorisksontheroadandtotheirinvolvementin roadaccidents.Thisworkexplorestwospecificaspectsofcrossingbehaviour:crossingspeedandhead pitches—theproportionoftimepedestrianspointtheirheadsdown(ratherthantowardsthetraffic) whencrossingaroad.Thelastoneisusedasanindicatorofthe(lackof)attentiontocross-traffic.We alsoexploredthepossibleeffectoffearoffalling(FOF)amongpedestrians,asitmightbeassociatedwith slowwalking,lessattentiontocrosstraffic,andmoreattentiontothepavementandtheirfootsteps.This paperreportsonafieldstudythatcombinedanobservatorytechniquewithshortsurvey.203pedestrians intwosites(signalisedandunsignalisedcrosswalks)werevideorecordedwhilecrossingtheroad.The FOFofpedestriansandothermeasuresofpedestrianbehaviouratcrosswalkswererevealedbymeansof questionnaire.Ageandgenderhadthemostsignificanteffectsoncrossingspeed,andFOFhadasignificant effectontheproportionofdownwardheadpitchesduringcrossing.
© 2010 Elsevier Ltd. All rights reserved.
1. Introduction
Understandingandanalysingtheriskspedestriansareexposed to when crossing a road has been the subject of many stud-iesaddressingaspectsrelatingtothetraffic,roaddesign,traffic signals and road users’ behaviours. Thereis extensive research onpedestrianbehaviour andthe evaluationof safety measures for pedestrians at urban areas. Two specific aspects of cross-ing behaviour that are studied in this work are pedestrians’ crossingspeedandheadpitches—theproportionoftime pedes-trians have their heads down. Both measures are interesting because they might be associated with pedestrians’ age and the so-called ‘fear of falling’ (that is associated by itself with olderage) thathasnot beenstudiedin thecontext ofcrossing behaviour.
Theincreasingproportionof olderpeopleinthecommunity inindustrializedcountries(inmanyofthemthisproportionhas reached10%andaboveit),andtheincreaseintheirlevelof mobil-ityandphysicalactivity,make thesafetyofolderroadusersan increasinglycriticalissue.
Olderpeopleareseenasavulnerablegroupofroadusers.A widerange offactorshasbeenexaminedin thiscontext. Older people are those individuals who are most likely tobe
physi-∗ Correspondingauthor.Tel.:+441173283197;fax:+441173283002. E-mailaddress:Erel.Avineri@uwe.ac.uk(E.Avineri).
cal vulnerable(DfT, 2001;Musselwhite,2006).Theyexperience deteriorationinsensoryandcognitiveskills(Dunbaretal.,2004; Kovalchik et al., 2004;Salthouse, 1996),and a progressive loss offeelingindependent(Orimoetal.,2006).Someorallofthese factorsmighthaveaffectonthecrossingbehaviourofolder pedes-trians.
Observationaltechniqueshavebeenwidelyusedtounderstand crossingbehaviourandidentifyriskybehavioursofdifferentage groups(see,for example,Oxleyetal.,1997).Theeffectsof age-relatedattitudinalfactorsandtheircontributiontoroadcrossing behaviourhavebeenalsoaddressedintheliterature,butmostof theresearchhasfocusedontherisktakingattitudesofyounger adults(HollandandHill,2007;Parkeretal.,1992).Therehavebeen veryfewstudiesontheattitudesofolderpedestrianstowardsrisky behaviour.For anolderpedestriannot payingenoughattention tocrossingtrafficmightbedescribedasariskybehaviour.But,in general,olderpedestriansexhibitsaferbehaviourwhencrossing a street(seeforexample Harrell,1991).Thehypothesisofhigh levelofrisk-takingamongolderpedestrianscontradictsthegeneral observationthatrisktakingdecreaseswithage;olderroadusers tendtotakefewerrisksthanyoungerpeopleinmanydifferentroad safetycontexts,andinroadcrossinginparticular(HollandandHill, 2007).
One shortcoming of olderpeople is theirslower gait. Older people walk more slowly when crossing the road (Coffin and Morrall, 1995; Oxleyet al., 1997). Thus, thetime spent bythe pedestrianatcrossing aroad(so-called “thetimeofexposure”) 0001-4575/$–seefrontmatter © 2010 Elsevier Ltd. All rights reserved.
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Under 1 Yr
1-4 yrs 5-14 yrs15-24 yrs25-34 yrs35-44 yrs45-54 yrs55-64 yrs65-74 yrs75-84 yrs 85 and over
age group
percentage was attributed
Motor Vehicle Falls
Fig.1. CausesofaccidentaldeathintheUSbyagegroups:fallsvs.motorvehicles (basedonCDC,2002).
increaseswith ageand increases risk exposure(Lassarreet al., 2007).
Inadditiontotheirvulnerabilitytotraffic,peopleofallagesare exposedandvulnerabletofalls.Cummingetal.(2000)definefear offalling(FOF)asageneralconceptthatdescribeslowconfidence atavoidingfalls,coupledwiththegreaterfearoftheconsequences offalling.FOFisknowntobemultifactorialwith,ataminimum, physical,psychological,andfunctionalcomponents.Inolder peo-pleitisassociatedwithpoorerhealthstatusandfunctionaldecline, increasein restrictionofactivity,psychologicalfactors (suchas depression and anxiety), and decreasedquality of life (Legters, 2002).DiagnosingFOFhasbeenthesubjectof atleast28 stud-iesusingarangeofmeasurementmethods(seereviewinScheffer etal.,2008).
FOFhasbeenlargelyassociatedwitholdage.Amongpeopleage 60andolder,fallsaretheleadingcauseofinjurydeaths(Stevens, 2005).Morethanathirdofolderpeoplefalleachyear(Hausdorff etal.,1994),andaboutquarterofthosewhofallsuffermoderate tosevere injuries (Alexander et al., 1992).It is the third lead-ingcauseofdeath(aftermotor-vehicleaccidentsand‘unspecified non-transportaccidents’)intheU.S.(CDC,2002).Fig.1showsthe frequencyofmotor-vehicleaccidentsandfallsascausesofdeath asafunctionofage:withtheformerpeakingaroundage20and thelatterincreasingexponentiallywithagetothepointwherethe poseagreaterriskthanmotor-vehicleaccidentsatage75+.
FOFis a concernto12–65%of olderadults (seea review in
Legters,2002).Thus,itispossiblethatpedestrianswhoaremore likelytobeafraidoffallingmightwalkslowerthanother pedestri-answhoarenot.Itcouldbealsohypothesizedthattheyaremore likelytopayless attentiontocrosstraffic,andpaymore atten-tiontothepavementandtheirfootstepswhencrossingaroad—a behaviourthatmightcompromisetheirsafety.
ObservationalmeasuresofFOFmightprovideadditional expla-nationofcrosswalkbehaviour.Forexample,itcouldbeassociated withthespeedofwalking(whichisexpectedtodecreasewithFOF). FOFcouldalsobeassociatedwithapedestrian’slevelofattention tocross trafficduringcrosswalking: theperceived riskofbeing involved,asapedestrian,inatrafficaccidentmightbetraded-off withthepsychologicaleffectofFOF;pedestrianswhoareafraidof fallingmightpaylessattentiontothecrossingtraffic,andpaymore attentiontothepavementandtotheirfootsteps.
Itisdifficulttocapturepedestrians’eyemovementsinthefield environment.Therefore,downwardheadpitches–specificallythe proportionof time pedestrianshave theirheads pointed down (ratherthantowardsthetraffic)whencrossingaroad–issuggested asanindicatorofthe(lackof)apedestrian’sattentionto cross-traffic.Whilethecorrelationbetweeneyemovementsandhead pitcheshavenotbeenexploredinthecontextofcrossingbehaviour,
thereissomeevidencethattheyarerelatedorevencoordinatedin arangeofhumanandanimalbehaviours(Pratt,1981;Guittonand Volle,1987;Land,1992;Gilchristetal.,1997;RomoserandFisher, 2009).
Largesaccadesareaccompaniedbyrotationsofthehead(see
Guitton,1988forareview).Moreover,accordingtoSparks(1991), theeyesandheadofsubjectsmoveinthesamedirectiononlyuntil theeyesreachtheiroculomotorlimitoruntilthelineofsightis directedatthetarget.Then,usually,theheadcontinuestomove andtheeyesmoveintheoppositedirection,compensatingforthe stillongoingheadmovement.Itmightbethereforearguedthatin suchsituationsheadmovementscouldmakebeagoodindicator ofcognitiveandvisualattention.
Cross-traffic and otherroad hazardscould be situated both withinandbeyondtheoculomotorrangehumanhave(whichis about±55◦); thereforea certainfrequency of headmovements towardssuchtargetsduringcrossingwouldbeexpected.
Using afield studyof crossingbehaviour,we exploredwhat measuresofpedestrianbehaviourexplaintheircrossingspeedand proportionofdownwardheadpitches.Wehypothesizedthat cross-ingspeedisreduced,andproportionofdownwardheadpitches duringcrossingisincreased,withageandwithFOF.
2. Methods
2.1. Sitesandparticipants
Thesiteschosenforthisstudywereastandardsignalised cross-walkandastandardunsignalisedcrosswalk,bothlocatedincentral Tel-Aviv,Israel.
Pedestrians’ crossing speed and other aspects of crossing behaviourareaffectedbythewalkingenvironment,andby pedes-trians’ motivation and the purpose of the journey (Finnis and Walton,2008).Toensurethatthepopulationsofpedestriansand driversat bothsitesaresimilaraspossible,thetwocrosswalks chosenforthisstudywerelocatedlessthan50mfromeachother. Therearenodesignedelements(suchashumps)orenhanced featurestoassistdisabledpeopleatthesecrosswalks.Thewalking surfacesatbothsiteshadessentiallynogradient.Bothcrosswalks wereclearlysignedand marked(withzebracrossings). Mainte-nanceconditionsofthecrosswalksweregood(withnocracked pavements,potholes,etc.thatmightaffectcrossingbehaviour).The speedlimitinthisurbanareais40km/h.
Thewidthofthesignalisedcrosswalkis10m.Itcrossesa two-wayroad.Thelengthofthegreenlightphase(23s)shouldprovide mostpedestriansareasonableamountoftimetocross.Atan aver-agewalkingspeedof1.2m/s(commonlyusedtocalculatetimings at pedestrian crosswalks) it takes 5.8sto cross theroad. Most pedestrians,eventhosewhowalkatspeedslessthan1m/s,should havesufficienttimetocrosstheroadduringthegreenphase.The unsignalisedcrosswalkis6.6mwideone-waystreet.Bylaw, pedes-trianscrossingsuchroadhavepriorityoverroadtraffic(although notalldriversinTel-Avivcomplywiththislaw).Bothcrosswalks areveryclosetoalargesupermarketandtoamedicalcentrethat attractpeopleofawiderangeofages.
2.2. Procedureandanalysis
Thisstudyusedamixedapproach:anobservationaltechnique wasappliedtostudythecrossingbehaviourofpedestriansatthe twocrosswalks,andaface-tofacesurveywasheldwithpedestrians immediatelyaftercrossing.
Pedestrianscrossingatbothsiteswererandomlysampled.At bothsites,thevideocamerawaslocatedatanunobtrusivefixed location on one side of the street. In order to make surethat
Author's personal copy
32 E.Avinerietal./AccidentAnalysisandPrevention44 (2012) 30–34
Table1
Ageandgenderfrequenciesofthepedestriansatthetwosites.
Site Numberofpedestrians Gender Agegroup(years)
Male Female 18–35 36–64 65andover
Signalisedcrosswalk 102 39 63 23 37 42
Unsignalisedcrosswalk 101 29 72 26 42 33
Total 203 68 135 49 79 75
headpitchesarewellcaptured,onlypedestrianscrossingthestreet towardsthelocationofthecamerawereincludedinthesample. Inordertoanalysetheregularwalkingspeed,whennotcrossinga road,pedestrianswerevideorecordedwalkingonthesidewalk sec-tionnearthecrosswalk.Allobservationsweremadeonamid-week daybetween10amand2pm.Thereweregoodweatherconditions (brightsun,norain)throughoutthestudy.
Aftertheycrossedtheroad,face-to-faceshortinterviewswere heldwiththepedestrians.Itincludedquestionssuchasthe pedes-trian’sage,maritalstatus,andhowoftendoesthepedestriancross this road. Pedestrians wereasked about whethertheywere in ahurrywhen crossingtheroad,aboutwhethertheyhave been involvedin roadaccidents,and whethertheyhavevision prob-lems.Inaddition,pedestrianswereaskedtodescribetheirFOFon a5-pointLikertscale(possibleresponseswere:Iamnotafraidof, usuallyIamnotafraidof,Iamabitafraidof,Iamafraidof,andI ammuchafraidoffalling).
Thedataanalysedinthisstudyincludesonlythosepedestrians whoprovidedafullresponsetothequestionnaire,andwho pro-videdtheirconsenttobeincludedinthestudy.Theresponselevel wasabout60%.
3. Results
Table1providesthenumberofpedestriansobservedateach crossingasafunctionofageandgender.
3.1. Crossingspeed
The effects of age, gender and the crosswalk type (sig-nalised/unsignalised)onthewalkingspeedofpedestriansatthe two sites were examined witha linear regression model. Also examinedwerethepedestrians’responsestoquestionsonbeingin ahurry,beinginvolvedinroadaccidents,havingvisionproblems, andbeingafraidoffalls.Becausetheproportionofpedestrianswho reportedanylevelofFOFwasratherlow(about14%atallthree lev-elsofFOF,representingpedestrianswhoare“abitafraid”,“afraid” and“muchafraid”offalling)thisvariablewasdichotomized.Tests onmulticollinearitydidnotshowthatanyofthevariableswere significantlycorrelatedwithothers.
The beta weights of the regression model are presented in
Table2.As expected agehad a significanteffect on the cross-ing speed: the crossing speed of younger adults (18–35) was significantlyhigher than thecrossingspeed of 36–64years old Table2
Theregressionmodelforwalkingspeedatthecrosswalk.
B Sig. (Constant) 1.450 .000 Female(dummy) −0.74 .056 Age18–35 0.116 .013 Age65+ −0.238 .000 Inahurry 0.02 .959
Notinvolvedinroadaccident −0.070 .304
Afraidoffalling −0.075 .198
Unsignalisedcrosswalk 0.02 .979 Hasavisionproblem −0.013 .714
1 1.1 1.2 1.3 1.4 1.5 1.6 1.7
18-35 yrs 35-64 yrs 65 and over
age group
walking speed (meter/sec)
signalised crosswalk unsignalised crosswalk sidewalk
Fig.2. Averagewalkingspeedsatthesignalisedcrosswalk,unsignalisedcrosswalk, andthenearbysidewalkfordifferentagegroups.
pedestrians,who,inturnwerefasterthanolderpedestrians(65+). Theageeffectwasconsistentinbothcrossingsandinthe‘normal’ walkingspeedonthesidewalk,asillustratedinFig.2.
Otherthanage,themodelalsoshowsthatfemaleswalkslower than males,though therelationship ismarginally significantat ˛=10%.Noneoftheothervariableshasasignificanteffecton cross-ingspeed.Specifically,wewereunabletoacceptthehypothesis aboutcrossingspeedbeingassociatedwithFOF.
Interestingly, Fig. 2 alsoindicates that pedestrians’ speed is fastestonthesidewalkandslowestattheunsignalisedcrossing, butthesedifferencesarenotsignificant.
3.2. Theproportionofdownwardheadpitchesduringcrossing The variables associated with the proportion of downward head pitches at the two sites were studied through a linear regression model, and the results are presented in Table 3. The variablesincluded inthemodel were:crosswalktype (sig-nalised/unsignalised),age,gender,andtheresponsesofpedestrians toquestionsontheirindividualcrossing-relatedexperiences:being involved in road accidents, having vision problems, and being afraidoffalls.However,thevariablesrelatedtopedestrians’ expe-rience(madeinresponsetothesurveyquestions)turnedouttobe insignificant.Thiswasnotsurprising;whilethere wasnot mul-ticollinearity betweenvariables, theinclusion of agemade the influencesoftheexperience-relatedvariablesweaker.Inamodel thatincludedagetheonlyvariablethatwasmarginallysignificant (at˛=10%)wasFOF.Therefore,forthepurposeofunderstanding theimpactoftheindividualexperiencevariablesmighthaveonthe frequencyofheadpitches,theagefactorhasbeenexcludedfrom theanalysispresentedinTables3and4.
Table3
Theregressionmodelfordownwardheadpitchesduringcrosswalking.
B Sig.
(Constant) 0.117 .121
Female(dummy) 0.040 .295
Inahurry −0.030 .403
Notinvolvedinroadaccident 0.008 .908
Afraidoffalling 0.119 .030
Unsignalisedcrosswalk −0.009 .896
Table4
Theregressionmodelfordownwardheadpitchesatthemidsectionofthecrosswalk.
B Sig.
(Constant) −3.689 .500
Female(dummy) 0.939 .064
Inahurry −0.332 .416
Notinvolvedinroadaccident 1.259 .110
Afraidoffalling 1.113 .026
Unsignalisedcrosswalk 1.207 .125
Hasavisionproblem 0.215 .593
Table 3shows that theonlyvariablesignificantly associated withdownwardheadpitchwasFOF,withthoseadmittingtoFOF atsomelevelspendingmoretimelookingatthepavement(26.4% versus14.0%).
Foramoredetailedanalysisofthephenomenonoflookingat thepavementwhilecrossing,weanalysedthepedestrians’head pitchesatthreedifferentsegmentsofthecrosswalks:atthestart andattheendofthecrosswalk,wherepedestriansstepdownor upthesidewalk,and inthemiddle,i.e.,themainsectionofthe crosswalk.Table4showsthebinarylogitmodelresultofwhether therespondents’headfaces downatthemiddle section ofthe cross(1=headis down;0=otherwise).Again,FOFwastheonly significantvariableinthemodel(thoughtheproportionof down-wardheadpitchesseemtobehigheramongfemalepedestriansat ˛=10%).
Noneofthevariableswasfoundtobesignificantintheanalysis oftheproportionofdownwardheadpitchesatthefirstandthird sectionofthecrosswalk.Thisisprobablybecausethebeginning andendofthecrosswalkinvolveaperceptualandphysicalchange thatcompelsmostpeopletoattendtoit(regardlessofFOF).
4. Studylimitations
Therewereseverallimitationstothisstudy.Ithasbeenimplied that a pedestrian’s head pitches while crossing is indicative of his/herattentionleveltothecross-traffic,thepavements,his/her ownfootsteps,orotherobjectsandmovementsinthewalking envi-ronment.Howeverheadpitchesmightnotbethebestindicatorof thedirectioninwhichhervisualorcognitiveattentionisfocused (thoughthetwoaregenerallycorrelated).Moreover,older pedes-triansmightsuffer,morethanpeopleinotheragegroups,from physiological,sensoryandcognitivelimitationsthatmightaffect headmovementand headpitches. Forexample, withadvanced agetheneckcanbecomemore stiff;coordination andspeed of movementswiththearms,handsandheadisdeclining,andeye sightdecreased(Isleretal.,1997;DoriotandWang,2006),among otherfactors,restrictthedegreeoffreedomofheadmovements andaffecttheproportionofheadpitches.Studyingdrivers’the fre-quencyofside-to-sideheadturnswhileexecutingturns,Romoser andFisherfoundthat cognitiveability ratherthanphysical fac-torsasasignificantpredictorofheadturns.Itmightbethatsuch age-relatedfactors,whichwerenottestedinthisstudy,andother controlledindicatorsofvisualandcognitiveattention(suchaseye movements)wouldprovideasignificantcontributiontothe expla-nationofheadpitches.Thiscallsforastudyofcrossingbehaviour inamorecontrolledenvironment(i.e.,asimulation experiment ratherthananaturalisticfieldstudy).
Also,itwassometimesdifficulttoestimatedownward,forward, ortowardstrafficheadpitches duringcrossingduetothefixed cameraangle,crossingtrafficandthedirectionpedestrianswere walkingin.
Toobtainthepedestrians’cooperationthequestionnairewas severelylimitedbylengthandincludedonly10questions. Pedes-trians’FOFhasbeenassessedusingasinglequestion,onarather
generalnatureofFOF.Althoughdirectlyderivedfromthe defini-tionofFOF,theresponsestothisquestionrevealnothingabout theindividuals’experiencewithfalling,theirriskperceptions,risk attitudes,andperceivedbehaviourcontrol.Futureresearchcould explorehowthesemeasuresmightbelinkedtocrossingbehaviour, andwhichofthemprovideabetterexplanationandvalidationof FOFamongpedestrians.
Finally,thestudyislimitedbyitsscale.Lackofsignificanceof someofthevariablesmightbeduetotherelativelysmall num-berofobservationswithineachcellinTable1.Asitwasnotthe mainfocusofthiswork,itwasnotattemptedtodrawspecific con-clusionsoncrosswalkdesignelementsandtheireffectoncrossing behaviour.Moreover,thetwositesdifferfromeachotherinmore thanonedimension(thesignalisedintersectionisatwo-wayroad andtheunsignalisedintersectionisaone-waystreet).Observing crossingbehaviourintwositesprovideonlylimitedgroundsfor anunderstandingofcrossingbehaviourinamoregeneralcontext, asavarietyoffactorsinthecrosswalkenvironment(suchas sur-facelevel,lighting,volumeoftraffic,andsafetymeasures)canbe associatedwithpedestrians’crossingbehaviour.
5. Discussionandconclusions
The safetyof olderpedestriansin theurbanenvironmentis a pressing issue for industrialized countries with an increas-ingly ageing demographic structure. The increase in conflicts betweenpedestriansandvehicleswhichresultsfromthegrowth inmotorisedtrafficinurbanareas,callsforgreateremphasison improving thedesignof the walking environmentfor vulnera-blepedestrians.Byunderstandingthespecificneedsofvulnerable pedestrians,and theirbehaviourinthecrossingofroads, cross-walkscouldbebetterplannedanddesignedtoimprovetheirsafety. ThisstudyhasshownthatagedifferencesandFOFhave signifi-canteffectsonpedestrianbehaviouratcrosswalks.Crossingspeed islargelyexplainedbyageandgender,whereastheproportionof downwardheadpitcheswhilecrossingcanbeexplainedbygender andFOF(whichbyitselfmightbeassociatedwitholdage).
Futureresearchcouldalsoaddresstheeffectofage,gender,FOF andotherrelevantvariablesonotheraspectsofcrossingbehaviour, suchasthechoiceofcrossing location(i.e.,wherea pedestrian is more likely to cross) or their potential contributionto gap-acceptancetheory(Hamed,2001;Manuszacetal.,2005;Lassarre etal.,2007).
Futureresearchshouldincorporatefindings and understand-ingofcrosswalkbehaviourintothedesignofinterventionstotreat FOFandgazebehaviourinboththetransportandgeneralcontexts. InterventionstotreatFOFintheplanninganddesignof pedestri-ans’walkingenvironmentcouldbeevaluatedbytheireffectson walkingspeed,attentionprovidedtocrosstraffic,headpitch,and –ultimately–numberofaccidentsinjuriesandfatalitiesamong olderpedestrianswhilecrossingthestreet.
Acknowledgments
TheauthorsacknowledgethehelpprovidedbyNoaSobeland LiatZoraninthecollectionofdata.Theauthorswouldalsolike tothankthethreeanonymousreviewersfortheirinsightful com-ments,whichhelpedimprovethecontentofthispaper.
References
Alexander,B.H.,Rivara,F.P.,Wolf,M.E.,1992.Thecostandfrequencyof hospitaliza-tionforfall-relatedinjuriesinolderadults.AmericanJournalofPublicHealth 82(7),1020–1023.
CDC,2002.Deaths:LeadingCausesfor2000.NationalVitalStatisticsReport(NVSR) 50(16). Centersfor DiseaseControl and Prevention, US.Available online: http://www.cdc.gov/nchs/products/nvsr.htm.
Author's personal copy
34 E.Avinerietal./AccidentAnalysisandPrevention44 (2012) 30–34
Coffin,A.,Morrall,J.,1995.Walkingspeedsofelderlypedestriansatcrosswalks. TransportationResearchRecord1487,63–67.
Cumming,R.G.,Salkeld,G.,Thomas,M.,Szonyi,G.,2000.Prospectivestudyofthe impactoffearoffallingonactivitiesofdailyliving,SF-36scores,andnursing homeadmission.JournalofGerontologyA:BiologicalSciences55,M299–M305. DfT,2001.OlderDrivers:ALiteratureReview.DepartmentforTransport,London. Doriot,N.,Wang,X.,2006.Effectsofageandgenderonmaximumvoluntaryrange
ofmotionoftheupperbodyjoints.Ergonomics49(3),269–281.
Dunbar,G.,Holland,C.A.,Maylor,E.A.,2004.OlderPedestrians:ACriticalReviewof theLiterature.DfT,London.
Finnis,K.K.,Walton,D.,2008.Fieldobservationstodeterminetheinfluenceof populationsize,locationandindividualfactorsonpedestrianwalkingspeeds. Ergonomics51(6),827–842.
Gilchrist,I.D.,Brown,V.,Findlay,J.M.,1997.Saccadeswithouteyemovements. Nature390,130–131.
Guitton,D.,1988.Eye-headcoordinationingazecontrol.In:Peterson,B.W., Rich-mond,F.J.(Eds.),ControlofHeadMovement.OxfordUniv.Press,NewYork,pp. 196–207.
Guitton,D.,Volle,M.,1987.Gazecontrolinhumans:eye-headcoordination dur-ingorientingmovementstotargetswithinandbeyondtheoculomotorrange. JournalofNeurophysiology58,427–459.
Hamed,M.M.,2001.Analysisofpedestrians’behavioratpedestriancrossings.Safety Science38(1),63–82.
Harrell,W.,1991.Precautionarystreetcrossingbyelderlypedestrians.International JournalofAgingandHumanDevelopment32(1),65–80.
Hausdorff,J.M.,Rios,D.A.,Edelber,H.K.,1994.Gait variabilityand fallriskin community-livingolderadults:a1-yearprospectivestudy.ArchivesofPhysical MedicineandRehabilitation82(8),1050–1056.
Holland,C.,Hill,R.,2007.Theeffectofage,genderanddriverstatusonpedestrians’ intentionstocrosstheroadinriskysituations.AccidentAnalysis&Prevention 39(2),224–237.
Isler,R.B.,Parsonson,B.S.,Hansson,G.J.,1997.Agerelatedeffectsofrestrictedhead movementsontheusefulfieldofviewofdrivers.AccidentAnalysis&Prevention 29(6),793–801.
Kovalchik,S.,Camerer,C.F.,Grether,D.M.,Plott,C.R.,Allman,J.M.,2004.Agingand decisionmaking:acomparisonbetweenneurologicallyhealthyelderlyand youngindividuals.JournalofEconomicBehaviorandOrganization58(2),79–94.
Land,M.F.,1992.Predictableeye-headcoordinationduringdriving.Nature359, 318–320.
Lassarre,S.,Papadimitriou,E.,Yannis,G.,Golias,J.,2007.Measuringaccidentrisk exposureforpedestriansindifferentmicro-environments.AccidentAnalysis& Prevention39(6),1226–1238.
Legters,K.,2002.Fearoffalling.PhysicalTherapy82(3),264–272.
Manuszac,M.,Manski,C.,Das,S.,2005.Walkorwait?Anempiricalanalysisofstreet crossingdecisions.JournalofAppliedEconometrics20(4),529–548. Musselwhite,C.B.A.,2006.Prolongingsafedrivingbehaviourthroughtechnology:
attitudesofolderdrivers.In:26thInternationalCongressofAppliedPsychology ,Athens,Greece,16–21July2006.
Orimo,H.,Ito,H.,Suzuki,T.,Araki,A.,Hosoi,T.,Sawabe,M.,2006.Reviewingthe definitionof“elderly”.InternationalJournalofGeriatricsGerontology6,149– 158.
Oxley,J.,Fildes,B.,Ihsen,E.,Charlton,J.,Day,R.,1997.Differencesintrafficjudgments betweenyoungandoldadultpedestrians.AccidentsAnalysis&Prevention29, 839–847.
Parker,D.,Manstead,A.S.R.,Stradling,S.G.,Reason,J.T.,1992.Determinantsof intentiontocommitdrivingviolations.AccidentAnalysis&Prevention24(2), 117–131.
Pratt,D.W.,1981.Saccadiceyemovementsarecoordinatedwithheadmovements inwalkingchickens.TheJournalofExperimentalBiology97,217–223. Romoser,M.,Fisher,D.L.,2009.Effectsofcognitiveandphysicaldeclineonolder
drivers’side-to-sidescanningforhazardswhileexecutingturns.In: Proceed-ingsoftheFifthInternationalDrivingSymposiumonHumanFactorsinDriver Assessment,TrainingandVehicleDesign.
Salthouse,T.A.,1996.Theprocessingspeedtheoryofadultagedifferencesin cogni-tion.PsychologicalReview103,403–428.
Scheffer,A.C.,Schuurmans,M.J.,VanDijk,N.,Vanderhooft,T.,DeRooij,S.E.,2008. Fearoffalling:measurementstrategy,prevalence,riskfactorsandconsequences amongolderpersons.AgeandAgeing37,19–24.
Sparks,D.L.,1991.Theneuralcontroloforientingeyeandheadmovements.In: Humphrey,D.R.,Freund,H.-J.(Eds.),MotorControl:ConceptsandIssues.John Wiley&SonsLtd.,Chichester,pp.263–275.
Stevens,J.A.,2005.Fallsamongolderadults—riskfactorsandpreventionstrategies. JournalofSafetyResearch36,409–411.