University of Groningen
Divided attention and driving. The effects of aging and brain injury
Withaar, Frederiec Kunna
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to
cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date:
2000
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Withaar, F. K. (2000). Divided attention and driving. The effects of aging and brain injury s.n.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the
author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately
and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the
number of authors shown on this cover page is limited to 10 maximum.
Download date: 17-06-2017
Í-t
DUmmarv
U
and the Ecologica
Chapters2 and
CHI patientsand c
with the issueof fi
In this thesis,divided attentionwas investigatedin four groupsof subjects:closed head
the most frequent <
injury (CHI) patients,young control and healthy older subjects,and older subjectswith
types of studiesart
cognitive impairments. It was studied how diffuse brain injury and normal and abnormal
processing,studie
aging affect cognitive processesinvolved in divided attentiontasks.Furthermore,it was
evaluatingprocedr
investigatedhow deficits in divided attentionrelate to performanceof instrumental
rate emergesof ab
activities of daily living (IADL), with an emphasison car driving.
particularlyoccurs
Divided attentionis defined in this thesisas the capacityto perform multiple tasks
concurrentlyand is conceivedof as the resultantof a diversity of contributing cognitive
(exceedingone mc
Furthermore,a
processes.Divided attentiondeficits can be the result of impairmentsin eachof the
traumaticallybrain
contributingbasic processes,as well as in control processesinvolved in task
(re)trainingshould
coordinationand integration(executivefunctions).
and predictionof tr
In chapter1, an overview is provided of severaltheoriesof attention.The chapter
startswith the well-known distinction betweenautomaticand controlled information
processingand describesthe two-processmodel, proposedby Shiffrin and Schneider.
operationalcapac
potentialas well.
Chapter3
This is followed by a descriptionof the model developedby Norman and Shallice.They
cognitiveimpairm
distinguisha mechanismresponsiblefor automaticprocessing,contentionscheduling,
distinguishbetwee
and a systemfor controlledprocessing,supervisoryattentionalcontrol. Evidencefrom
Predictivestudies<
neuro-imagingstudiesis presented,linking control aspectsof attentionto frontal parts of
significantlyworse
the brain. Finally, the model of Brouwer and Fasotti is described.This model describes
high prevalenceof
the interactionbetweenbasiccognitive and higher order control processesand offers
traffic accidentsan
explanationsfor deficits in complex task performance.
is found. Low to rn
Emphasisis placed on methodologicalconcernsand theoreticalnotions especially
testresultsand on-
relevantfor divided attentionstudies,such as disentanglingthe influenceofbasic
betweencognitive
cognitive and higher order control processes,resourcesand skill. Visual attention
concludeswith a d
received specialconsiderationdue to its major relevancefor perceptual-motorskills
driving, includingr
necessaryin car driving. Spotlight metaphorsare describedalong with the influential
operationalizatio
visual attentiontheoriesdevelopedby Mesulam and Posner.Two theoriesof visual
attentionrelevantfor perceptionin car driving, namely the Useful Field of View model,
2r2
Chapters4 thro
a driving simulato
Jummarv
and the Ecological Approach are also described.
Chapters 2 and 3 are review studies,describing the impact of diffuse brain injury in
CHI patients and cognitive impairments in older drivers on car driving. Chapter 2 deals
with the issue of fitness to drive in patients suffering from traumatic brain injury (TBI),
the most frequent causebeing CHI. Guidelines for assessmentare proposedand three
types of studies are reviewed: studiesof impairments of attention and information
processing,studiesof driving competence,and driver selectionstudiesapplyingand
evaluatingproceduresfor assessingfitnessto drive. From thesereviews a relicensing
rate emergesof about 5O7ofor very severeTBI patients.IJnsuccessfulrelicensing
particularly occurs in patients with a very long duration of post-traumaticamnesia
(exceeding one month) and with severeimpairments of perception and judgement.
Furthermore, a review of studiesdescribing training of driving competencein
traumatically brain-injured patients is presented.It is argued that graded proceduresfor
(re)training shouldbe developedand that assessment
should extendto training advice
and prediction oftraining success.Evaluationproceduresshouldnot only focus on
operationalcapacities,but should include measuresof executivefunctions and learning
potential as well.
Chapter3 is a review on assessment
of fitnessto drive in older driverswith
cognitive impairments. Early studieson dementiaand driving generally failed to
distinguish between safe and unsafe drivers on the basis of cognitive test performance.
Predictive studies demonstratedthat cognitively impaired subjectsas a gÍoup, perform
significantly worse than controlson both neuropsychologicaland driving measures.A
high prevalenceof cognitive impairment is found in groups of older drivers involved in
traffic accidentsand crashes.Meanwhile, a largerangein neuropsychologicaltest scores
is found. Low to moderatecorrelations have beenreported betweenneuropsychological
test results and on-road driving performance.Thus, it remains difficult to discriminate
between cognitively impaired subjectswho are fit or unfit to drive. The review
concludeswith a discussionof methodologicaldifficulties in the field of dementiaand
driving, including subjectselection,the choiceof neuropsychologicaltests,and the
operationalization of driving performance.
Chapters4 through 6 describe assessmentmethodsof driving performance,including
a driving simulatortest ride (chapter4), selectionmethods(chapter5),
2t3
Summary
neuropsychologicaltests and on-road tests (chapter 6). Chapter 4 focuses on driving
rmparnnenmwas c
simulator assessment.In an attempt to combine the merits of neuropsychological
Orientation and mr
assessment(standardization,reliability) and an on-road test (ecologically valid
thousandsubjects
measures,behavioral observations),a virtual test ride was developed in the driving
study. The mean E
simulator of the Centre for Environmental and Traffic Research (University of
subjects(4Vo)dem
Groningen). The implementation of intelligenttraffrc participants provided a convincing
medical screening
driving environment, in which interaction with other caÍs was studied and behavioral
for the full period r
observationswere scored.Results of a pilot study with eight subjects are presented.The
relicensingproced
sampleincluded five (mildly) dementedpatients, one patient with Parkinson'sdisease
consequencethat tJ
and two TBI patients. The most significant problem that arose in the pilot study was the
increasedsusceptibility for simulator-sicknessin this group.
restrictedlicenses
The test ride was subsequentlyvalidated in a group of 29 healthy older subjects.
They were experienceddrivers and their operational and tactical driving performancein
younger and score
than the others.
Of the initial 11
the driving simulator was scored as sufficient to good. The subjectsrated the
5, 80 subjectscom
resemblanceof the virtual simulator world with the actual traffic environment as
driving test. The nt
satisfactory.In particular, the intelligent interactive robot cars were perceived as "real"
80 subjectsare des
road users.Most subjectsneededtime to get used to the simulator car but they rated the
with a standardder
test ride as easyto perform. Five subjectsbecamesick and aborted the test ride. Two
domains for predic
additional subjects suffered from serious nausea.Subjects also complained about
attention,divided a
dizziness,and an uncomfortable feeling of tension. Eight subjects (28Vo)were
Driving performanr
completely free of complaints related to simulator sickness.Due to the high incidence of
Ride to Investigate
simulator sickness,even in healthy subjects,the driving simulator was not routinely
impaired driving sl
used for clinical assessments,despitethe initially high expectationsof its usefulness.
scoreson a diversit
Chapters5 and 6 describe parts of a driver evaluation study to investigate fitness to
State Examination
drive in cognitively impaired older drivers. The study consisted of three assessments:a
two subjectsscore
medical screening(for all subjects),and a supplementaryneuropsychological evaluation
cognitive decline. .
and an on-road driving test (for subjectswith cognitive impairment). The study was
performance, the cr
implemented in the official, obligatory relicensing procedure for older drivers in the
Netherlands.
stepwisemultiple r
Chapter 5 describesthe medical screeningand the selection procedure,and provides
performance (operr
reflecting overlean
an estimation of the prevalenceof cognitive impairments and an overview of relicensing
explaining one thirr
rates in older drivers. For the medical screeninga behavioral rating scale for cognitive
0.32).A logistic re
214
Summarv
impairmentswas developed.This instrument(the oPS) coveredthe domainsof
Orientationand memory,Praxis and attention,and Social functioning. Almost threethousandsubjects(n = 2992) who appliedfor licenserenewalwere includedin the
study.The mean ageof the group was74 years,ranging fiom 65 tot 94 years.only 1l I
subjects(47o) demonstratedsignsof cognitive impairmentas observedduring the
medicalscreening.Of thesepersons,45 subjects(39Eo)could renewtheir driving license
for the full period of five years.Thirty-two candidates(297o)voluntarily sroppedrhe
relicensingprocedure,and2l subjects(19Eo)failed the on-roadtest,with the
consequence
that their licensewas not renewed.An additional 10 subjects(127o)needed
restrictedlicenses.The 45 subjectswith renewedfull licenseswere significantly
youngerand scoredsignificantly betteron a diversity of neuropsychologicalmeasures
thanthe others.
Of the initial I I I subjectsselectedfrom the 2992 older drivers describedin chapter
5, 80 subjectscompletedboth the neuropsychologicalassessment
and the on-road
driving test.The neuropsychological
test resultsand driving performancedataof these
80 subjectsare describedin chapter6. The averageage ofthe subjectswas 78 years,
with a standarddeviationof sd = 5.8. The neuropsychologicalevaluationcoveredfour
domainsfor predictingdriving performance,namely perceptual-motorspeed,visual
attention,divided attentionand executivefunctions,and practicaltask performance.
Driving performancewas evaluatedby meansof an observationmethod,the TRIP (Test
Ride to InvestigatePracticalfitnessto drive). The group as a whole demonstrated
impaireddriving skills. The neuropsychologicalevaluationrevealedseverelyimpaired
scoreson a diversity of tests.Impaired scoreswere also obtainedon the Mini Mental
StateExamination(meanMMSE scorewas 23 with a standarddeviation of 4.5). Thirtytwo subjectsscoredbelow 24 onthe MMSE, reflecting mild (30%) to moderate(10Ea)
cognitivedecline. All cognitive domainswere significantly correlatedwith driving
performance,the correlationcoefficientsvarying betweenr = 0.40 and r = 0.50. In a
stepwisemultiple regressionanalysis,the domainsof visual attentionand practicaltask
performance(operationalizedas drawing from memory and copying from example,
reflectingoverlearnedskills) emergedas the strongestcombinationof predictors,
explainingone third of the sharedvariancewith driving performance(adjustedR2 =
0.32).A logistic regressionanalysisresultedin 767ocorrect testscore-based
predictions
215
Summary
in terms of passing or failing the on-road test.
patients and healtt
Chapters 7 and 8 are concerned with the nature of deficits in divided attention in
divided attention i
terms of speed and control. The study described in chapter 7 evaluated the contribution
clinical and practir
of speedand control processesto divided attention tasks in a group of 26 (very) severe
The rationaleb
CHI patients and 25 orthopedic controls. All subjectswere evaluatedthree months post-
The single task val
injrrry. Single and dual task paradignu were administeredto assessstimulus driven
of all basic cogniti
flexibility (four-choice visual and auditory reaction time task), memory driven
specific singletasl
flexibility (Trail Making test), and strategy driven flexibility (continuous tracking task
contribution of bot
and arrow identification). Secondly, the relationship between divided attention measures
coordination and ir
and subjective impairments in IADL was investigated.Reduced speedof information
statistically or exp
processing following CHI proved to be largely responsiblefor difficulties in dual task
as impairments in I
performance, but dual task measuresresulted in higher correlations with self-reported
Visual attentior
IADL than single task measures.
may be involved ir
Chapter 8 comparesthe deficits in divided attention of CHI patients and healthy
the functional fielc
older subjects.The test scoreson the divided attention tasks of the 26 subacuteCHI
In both CHI pal
patients were compared with data from 54 healthy older subjects (mean age 65 years
control processes€
with a standarddeviation of 7.6).It was hypothesizedthat aging, like CHI, would lead
divide, allocatean
to a reduction in processing speed.Impairments in control processes,emerging in dual
deficits in control'
task performance after (statistically or experimentally) controlling for single task
therefore, remains
performance, were supposedto exist for healthy older subjects,but less so for CHI
attention in these g
patients. A general reduction in the speedof information processingin healthy older
Significant con
subjects was demonstratedin most single task conditions. These effects were
driving test in olde
comparable with the reduction in speedof information processingobservedin CHI
theoretical point ol
patients. Healthy older subjectsdemonstratedsomeimpairments in control processesbut
of fitness to drive.
these were less extensive than reported in earlier studies.In fact, the results of the CHI
alarms.Therefore,
patients and the healthy older subjectswere fairly convergent.In both groups,
Despiteits expe
impairments in control processesemergedin the most complex tasks in which active
proved not to be fe
strategies to divide, allocate and shift attention were required.
sickness.On-roadt
The final chapter contains the discussion.The main theoretical themesof the thesis
are recapitulated, along with some clinical and practical implications and
methodological concerns. Three topics are considered.The first topic is concernedwith
the contribution of speedand control processesin complex task performancein CHI
216
Thus, up to date,n
Validation researc
V/ith regardto:
all older drivers, ar
Summary
patientsand healthy older subjects.The secondtopic addressesthe role of deficits in
divided attentionin driving and other IADL activities.The last topic is concernedwith
clinical and practical implications.
The rationalebehind the single/dualtask paradigmsusedin this thesisis discussed.
The single task variantswere regardedas baselinemeasuresreflectingthe effectiveness
of all basic cognitive processes,as expressedin terms of speed,neededto perform the
specific singletask. The dual task is the experimentalcondition reflectingthe total
contribution of both basic cognitive processesand of processesinvolved in task
coordinationand integration.When the influenceof basic cognitive impairmentsis
statistically or experimentally controlled, additional dual task effects can be interpreted
as impairmentsin task coordinationand integration(control processes).
Visual attentionwas given specialconsideration.It is arguedhow control processes
may be involved in a mainly perceptual task and how attention may be distributed across
the functionalfield of view.
In both CHI patientsand healthy older subjects,it was concludedthat impairmentsin
control processesemergedin the most complex tasks in which active strategiesto
divide, allocateand shift attentionwere required.It shouldbe realized,however,that the
deficits in control were mostly small and could not be demonstratedin all tasks.Speed,
therefore, remains a major factor in understandingthe nature of deficits in divided
attentionin thesegroups.
Significantcorrelationsbetweenneuropsychologicalmeasuresand an on-road
driving test in older drivers with cognitive impairmentswere established.From a
theoreticalpoint of view thesecorrelationsare promising,but for individual assessment
of fitnessto drive, predictionsof successstill lead to high numbersof missesand false
alarms. Therefore, actual observation and driving measuresremain warranted.
Despiteits experimentaland methodologicaladvantages,the driving simulator
proved not to be feasiblefor routine assessments
due to the high incidenceof simulator
sickness.On-roadtestshave not beenevaluatedyet for their psychometricproperties.
Thus, up to date,no information on validity and reliability is sufficiently available.
Validation researchin this field is thereforevery important.
With regardto relicensing,a three-stagemodel is proposed:a medical screeningfor
all older drivers,and a neuropsychologicalevaluationand an on-roadtest for drivers
211
Summary
with observedcognitive impairments. The debateof relicensing in older drivers is,
however, a controversial issue. Organizationsthat promote the rights and interests of
older individuals claim that a special relicensing procedure for older people is a form of
discrimination. In justifying relicensing procedures,licensing authorities claim that the
prevalenceof impaired fitness to drive is increasedin older people compared to middleaged subjectsdue to aging-relatedfunctional impairments and/or diseases.
The studiesdescribedin this thesis enhanceour knowledge to some degree about the
relationships between cognitive impairments and complex task performance, but also
Verdeeldeaanda
theoretischdeel e
aandachtwerd on
hersenletsel.mees
point out the limitations encounteredin this field of research.Future researchshould not
proefpersonenen
be directed in the first place at finding new neuropsychologicaltests or experimental
eerstewaag is hor
tasks being used in prediction of driving performance.A more fruitful approach would
be to develop observationinstruments that meet certain psychometric quality
functiestoornisse
characteristicsof neuropsychologicaltests and tasks. Observation instruments or
samenhangen
met
methods should combine the ecological validity of current driving tests with the
psychometric qualities of neuropsychological tests.
Activities of DailS
bij verdeeldeaand
Verdeeldeaan
meerderetaken te1
diversiteit aancog
aandachtsproble
basalecognitieveI
bij het coórdinere
In hoofdstuk 1
Het hoofdstuk beg
gecontroleerdeinf
ontwikkeld door S
model dat is ontwi
dat verantwoordel
een systeemvoor j
worden resultaten
pogen te localisen
Brouwer en Fasoí
cognitieve en hog
het uitvoeren van i
218