Genealogy

Wakefulness in young and elderly subjects driving at night in a car simulator

Description
Wakefulness in young and elderly subjects driving at night in a car simulator
Categories
Published
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Accident Analysis and Prevention 41 (2009) 1001–1007 Contents lists available at ScienceDirect Accident Analysis and Prevention  journal homepage: www.elsevier.com/locate/aap Wakefulness in young and elderly subjects driving at night in a car simulator Arne Lowden a , ∗ , Anna Anund b , c , Göran Kecklund a , b , Björn Peters c , Torbjörn Åkerstedt a , b a Stress Research Institute, Stockholm University, SE-106 91 Stockholm, Sweden b Karolinska Institutet, SE-171 77 Stockholm, Sweden c Swedish National Road and Transport Research Institute, SE-581 95 Linköping, Sweden a r t i c l e i n f o  Article history: Received 12 August 2008Received in revised form 30 April 2009Accepted 4 May 2009 Keywords: AlertnessBrain activityAgeNocturnalCortisol a b s t r a c t Youngdriversareover-representedinnighttimetrafficaccidentsandseveralstudieshavesuggestedthatmany accidents are associated with elevated sleepiness levels. It has been suggested that there may bea connection between lowered wake capacity and functional sensory motor skills on the one hand andsleep deprivation at the circadian low in young drivers on the other.Performanceduringa45/mineveningandnightdriveamongyoung( n =10,agerange18–24years)andelderly ( n =10, age range 55–64 years) subjects was studied using a moving base driving simulator. EEGwas measured continuously. Every 5min, subjects were rated on the Karolinska Sleepiness Scale (KSS).Saliva cortisol was assessed before and after each drive.The results showed that sleepiness increased across each drive and was higher among young driversat night. Relative EEG power increased among older drivers for frequencies of 10–16Hz. The sigma 1frequency band (12–14Hz) proved particularly sensitive to sustained driving, and was elevated amongsubjects in the elderly group. Cortisol levels before and after the evening and night drive showed highermean levels for elderly subjects.The present study has demonstrated that young drivers were more sleepy while driving at night. Theeffects could represent a mobilization of effort and a reorganization of brain firing pattern among oldersubjects, possibly reflecting better ability and effort to resist sleepiness.© 2009 Elsevier Ltd. All rights reserved. 1. Introduction Agedifferenceindrivingperformancehasbeenobservedinsev-eral studies. The young drivers are more frequently involved intraffic accidents (Langlois et al., 1985; Lavie et al., 1987; Horne andReyner, 1995; Pack et al., 1995). In a more recent study, Åkerstedt and Kecklund (2001) f ound using accident register data correctedfor road traffic flow that young drivers (18–24 years) ran a 5–10timeshigherriskofbeinginvolvedinatrafficaccidentlateatnight(ÅkerstedtandKecklund,2001).Forwomen,thenighttimeriskwas less pronounced.The reasons why young drivers run a higher than normal riskof being involved in a driving accident late at night are not clearlyestablished, but factors related to self-confidence, risk-taking anddrug use have been suggested (Gregersen and Bjurulf, 1996).Another factor could be the elevated risk among newly licenseddrivers,wholackpracticeandexperience(Ferguson,2003).Another very likely reason is sleepiness (Cummings et al., 2001; Connor etal.,2001).Ithasbeenhypothesizedthatyoungdrivershavealower ∗ Corresponding author. Tel.: +46 8 5537 8915; fax: +46 8 5537 8900. E-mail address:  arne.lowden@stressforskning.su.se (A. Lowden). wake capacity in sleepy situations than older drivers (Åkerstedtand Kecklund, 2001). Sleepiness constitutes a significant factorsince it increases the risk of falling asleep at the wheel and low-ers the ability to maintain the functional sensory motor skillsneeded to maintain a road position (Banks et al., 2004; Lal andCraig, 2005). To maintain vigilance and acceptable performancelevels, the hypothalamic–pituitary–adrenal (HPA) axis has to beactivated. Interestingly, some of the differences in driving perfor-mance between young and elderly subjects could be related tochangesinHPAfunctionsrelatedtoaging.Dataindicatethatatolderagesaninabilitytoshutofftheallostaticsystemmightgiverisetoaprolongedstressresponse.Moreover,anaturalnocturnalriseincor-tisolexcretionseemstobepartofthenormalagingprocess(Ferrariet al., 2001).Sleepiness has been defined as an inability to maintain wake-fulness without external help (Dinges et al., 1987). Sleepiness is predominantly coupled to sleep regulation and the wake capacityof the brain, and can be operationally defined as a drive for sleep(DementandCarskadon,1982).Thedegreeofsleepinesscanbeesti- mated by the number of hours an individual has been awake, priorsleep length, but is also influenced by sleep quality, the low pointof the circadian rhythm (at late night), alcohol and drugs. For moststudies involving sleepiness the EEG (especially increases in delta 0001-4575/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.doi:10.1016/j.aap.2009.05.014  1002  A. Lowden et al. / Accident Analysis and Prevention 41 (2009) 1001–1007  and theta activity) seems to be reliable and normally very highlycorrelated with reports of sleepiness and driving decrements suchaslanedrifting(LalandCraig,2001;Banksetal.,2004).Ithasbeen suggested that the EEG may also be useful when designing driverfatigue countermeasure devices (Lal and Craig, 2005).A problem with many studies detecting sleepiness and drivingperformance is the fact that measurements are taken only duringdaytime (see for example Horne and Baulk, 2004; Otmani et al.,2005; Roge et al., 2003) without taking into account circadian fac-tors. Very few studies have investigated age effects in connectionwith night driving despite observed differential age–accident out-comes. We therefore decided to perform a study with good facevalidity by including night drive conditions close to the circadianlow and choosing a non-vigilant environment with rural drivingand very little traffic in order to induce additional sleepiness. Itseemed relevant to include EEG based methodology paired withself-evaluation of sleepiness and effort, driving performance andsome indicator of the activity of the HPA axis (saliva cortisol). Theaim of this study was to investigate the effects of an evening andnight drive on wakefulness and driving performance among youngandelderlysubjectsinanadvancedmovingbasedrivingsimulator. 2. Methods  2.1. Subjects and design Atotalof20subjectstookpart.Tensubjectswereyoungdrivers(aged18–24years)and10wereelderlydrivers(aged55–64years).Half the subjects in each age group were female. Subjects wererecruited by advertisement in the local newspaper. For inclusion,the subjects were required to fill in a background questionnairein which they stated they were healthy, had driving experience of morethan5000kminthepastyear,nopriorjetlag,nopresentsleepdisturbances, no shift work during the past month and no plannedorperformedextensivephysicalactivity.Morningness/eveningnessscores(TorsvallandÅkerstedt,1980)didnotdifferbetweengroups. One young male subject had to be excluded from the EEG analysesdue to technical artefacts.The study was approved by the Regional Ethical Committee atLinköping University, Sweden.  2.2. Moving base driving simulator  A dynamic, Hi-Fi moving base driving simulator was used inthe study. It comprised a cut-off vehicle cab (Volvo 850, VolvoPersonbilar Sweden AB, Gothenburg, Sweden) simulating accel-eration in three dimensions, trough roll, pitch and linear lateralmotion. The visual system displayed a scenario on a 120 ◦ -widescreen 2.5m in front of the driver. The sound system generatednoise and infrasound resembling the internal environment in amodern passenger car. The vibration system simulated the sensa-tionsthedriverexperiencesfromcontactbetweentheroadsurfaceand the vehicle. The driving scenario was a rural two-lane roadwith lanes 3.75m wide, a right lane border, and a randomizedsmooth curvature. The road repeated itself every 18km. The ambi-ent light conditions corresponded to daylight—cloudy but clearwith good visibility. The signal speed limit was 110kmh − 1 andthere was no oncoming traffic or cars to follow or overtake. Driv-ing performance was evaluated by lane drifting. Other measuressuch as brake reactions, incidents (one wheel across a line) orcrashes(allwheelscrossingthelineorgoingofftheroad)occurredvery seldom and could not be used. The simulator has been exten-sively validated and documented in technical reports (Aurell et al.,2000).  2.3. Experimental protocol Arepeatedmeasurementdesignwasused.Eachparticipantper-formed two drives, one in evening conditions and an additionaldriveatnight.ThedriversvisitedtheVTI(SwedishRoadandTrans-port Research Institute) twice. The first visit included a trainingsession and the second contained both evening and night driveconditions. Both drives were conducted on the same day in orderto reduce intra-individual differences by using similar electrodeplacement and similar baseline signal characteristics to enablecomparisons between evening and night drives. The procedure of performing dual drives in 1 day is common for shift workers goingback and fourth from work and forms a naturalistic scenario.Trainingwascarriedoutforatleast3daysbeforetherealexper-iment. During the first visit, subjects were informed about theexperimental procedure. They were also instructed and trained inthe use of the Karolinska Sleepiness Scale.The subjects reported their sleep habits and quality in theKarolinska Sleep Diary (Åkerstedt et al., 1994) three nights before the test. Also coffee and alcohol consumption was reported. Sub- jects were asked to abstain from coffee after 11:00h on theexperimental day. They were asked to rise no later than 08:00hin the morning. Two subjects were run in pairs during the eveningandnight.Subjectonearrivedat17:00handsubjecttwoat18:00h.When each subject arrived, the electrodes were applied immedi-ately. The evening condition was an early drive between 18:00hand 19:30h (subject one) or between 19:30h and 21:00h (subjecttwo).Bothwereofferedamealat00:30handthefirstsubjectthendrove in the night condition between 02:30h and 04:00h and thesecond subject between 04:00h and 05:30h. The meals consistedofareadymadeselectedmealofeitherfishormeat(600–700kcal).Between conditions, subjects watched television or read, super-vised by the test leaders.After 45min of driving the driver was informed that a drowsi-ness warning system (DWS) was being activated, after whichsubjects continued driving for another 45min (see report VTI17 9,Anund et al., 2004). The present study only reports on the first45min of undisturbed driving. Saliva samples were drawn beforeandafterthedriveandsomequestionsondrivingperformanceandstress were asked after each condition.  2.4. Electrophysiological measurements—EEG, EOG and EMG To measure EEG, the VITAport system was used. This includedsixteen channels for physiological measurements (Vitaport-2 digi-tal recorder, TEMEC Instruments B.V., Kerkrade, The Netherlands).Electroencephalogram (EEG), Electrooculogram (EOG) and Elec-tromyogram (EMG) results were recorded. The input data werestored on a Flash card (Viking, USA) and downloaded at the endof each condition to a PC hard drive. The sampling frequency was256Hz for EEG and 512Hz for EOG/EMG. The EEG was measuredthrough three bipolar derivations positioned at Fz-A1, Cz-A2 andOz-Pz. The EMG electrodes were placed under the jaw. The elec-trodes in use were copper-plated and the EOG and EMG electrodeswere disposable Ag/AgCls.The 264Hz digital EEG signal from the solid state, battery-powered unit was subjected to an FFT analysis by re-sampling thesignal at the frequency of 64Hz with a 0.35–70Hz band pass filterand a window of 4s. In the next step, all 4-s bins of the EEG con-taining arousals or artefacts were removed. Arousals and artefactswere visually defined according to the Sleep Disorders AssociationCriteria (1992). If an entire 20-s epoch was disturbed it wasexcluded from analysis. The power in frequency bands (1–35Hz)was integrated over time to yield mean values for every 4s in thedelta(1–4Hz),theta(4–8Hz),alpha(8–12Hz)andbeta(16–32Hz)frequency bands respectively and was further transformed to yield   A. Lowden et al. / Accident Analysis and Prevention 41 (2009) 1001–1007   1003 mean power for 5-min bins. To standardize for individual differ-ences relative power was calculated using the means of the first5min of driving divided by the standard deviation over that period(Baulketal.,2001).Deltapowerisnotreportedsinceitwasfoundto be too contaminated with artefacts from blinks and smaller move-ments, and sleep was very scarce in the present study.  2.5. Karolinska Sleepiness Scale (KSS) Subjective ratings of sleepiness were obtained using a modifiedversion of the Karolinska Sleepiness Scale (KSS; Reyner and Horne,1997).Thisconsistedofa9-pointscalewithverbalanchorsforeverystep (1=extremely alert, 2=very alert, 3=alert, 4=rather alert,5=neither alert nor sleepy, 6=some signs of sleepiness, 7=sleepy,but no effort to remain awake, 8=sleepy, some effort to keep alert,9=very sleepy, great effort to keep alert, fighting sleep). KSS hasbeen validated against EEG measurements of sleep (Åkerstedt etal., 1991). During each driving session, the subjects gave a verbalrating of sleepiness through the intercom every 5min while driv-ing. A printout of the scale was placed on the steering wheel as areminder to the subject.  2.6. Saliva cortisol The participants were instructed to sample saliva before andafterdriving(Salivette ® ,Sarstedt;RommelsdorfGermany).Salivarycortisol was determined through RIA (Orion Diagnostica, Finland).Thelowerlimitofsensitivitywas20nmol/Linplasmaand1nmol/L in saliva and the average inter- and intra-assay coefficient of varia-tion never exceeded 10%.  2.7. Driving performance During each drive, measurements were obtained for speed, lat-eralpositionandsteeringwheelangle.Thesamplingfrequencywas33.33Hz.Max,min,meanandstandarddeviationswerecalculatedfor 5-min intervals. The lateral position was measured using theperpendicular distance between the right side of the right frontwheelandtheleftsideoftheright-handlaneborder.Inthepresentstudy,thestandarddeviationofthelateralpositionwasselectedasthis seems to be a measurement sensitive to sleepiness in continu-ousperformance(Arnedtetal.,2001).Incidentsoccurredwhenthe driver touched or crossed the lane boundaries.A questionnaire was filled in after each drive. The questionsconcerned self-evaluation of performance and symptoms duringthe drive. Items included “performance”, “effort”, “realistic drive”,“sickness feelings”, “boredom”, “worry” and “stress”. Ratings weremade on a 7-point scale (1=very good/no symptoms, 7=verybad/many symptoms).  2.8. Statistical analysis The EEG data from the first 45min of each drive were trans-formed to 5-min epochs by calculating the mean level of eachmeasurement. Thus, minutes 1–5 formed the first measurement,minutes 6–10 the second, etc. EEG data gave the spectral power invarious frequency bands. In order to control for individual differ-ences, the first 5min of the early drive was used as a reference orbaseline value. The mean difference from the baseline level wascalculated and used in all further analysis. Driving performanceusing the standard deviation for the lateral position was calcu-lated into 5-min means. Minutes when subjects had an incidentoraccidentwereexcludedfromthecalculations.Meanvalueswereanalyzed mainly with a two-way repeated measurement analysisof variance (ANOVA) with an additional grouping factor. The twomain factors were Condition (evening–night drive) and Time (ninemean values representing every 5min of the first 45-min drive).A third factor, Age Group (young–elderly), were used as a group-ing factor and included in all models. Main effects and interactionswere calculated and presented in tables. This procedure includeda Huyhn–Feldt correction for unequal variances. For saliva mea-surements taken before and after each drive, the factor Time in theANOVAs included four mean values. To avoid skewness, data weretransformed before testing using the best fit, the inverted squareroot.For questions on health and screening the mean differencesbetweengroupswerecalculatedandtestedusingtwo-tailed t  -tests.Data from the sleep diary 3 days prior to driving were analyzedusingaone-wayANOVAwithdayasthemainfactorandAgeGroupas a grouping factor. All results in figures are presented as meanand standard error. The significance level was set to 5% in all anal-yses. 3. Results Table 1 summarizes the ANOVA results for variables measuredduring each 45min drive, the relative EEG power frequency bandsranging from 4 to 32Hz, self-evaluated sleepiness levels (KSS),saliva cortisol and driver performance expressed as the standarddeviation for lateral position. Mean values for corresponding vari-ables divided into nine 5-min intervals across 45min of driving inboththeeveningandnightconditionareplottedinFig.1,including the standard error of means.Sincesubjectswereseated,didnottalkandconcentratedonthetask, relatively few artefacts were observed in EEG measurementswhile driving. Less than 10% of the 4-s epochs were excluded dueto artefacts.Significant main effects of condition were obtained for EEGpower in the frequency ranges 8–10 and 14–32Hz ranges. More-over, total power yielded a similar effect representing a relativeincrease in power during the night drive. Across both the eveningand night drive, a low incidence of falling asleep was observed andnoincreaseinthetaactivitythatcouldbeassociatedwithsleepwasthus found. Only one subject had an incidence of falling asleep andthis occurred during the early drive.EEG power activity increased across the nine 5-min bins duringthe 45-min drive and a main effect of time became significant inthe alpha and sigma 1 (8–14Hz) bands but not in any other bands.Thiseffectwasalsoreflectedinasignificantincreaseintotalpower(4–32Hz) across time. A main effect of age was exclusively foundfor higher frequency bands above 12Hz (12–32Hz), power beingincreased for the elderly group during both conditions.Some significant interactions indicate that the age groupsreacted differently during the experiment. Fig. 1 offers some clar- ity regarding the direction of the interactions. Age differences aresmall during the evening drive in all frequency bands, but are veryapparenttowardstheendofthenightdrive;thedifferencesbecomesignificant for alpha 2 trough sigma 2 ranges (10–16Hz). In thesigma 1 range (12–14Hz), the age effect already seemed to appearduring the evening, as an age and time interaction was found.A self-evaluation of sleepiness using the Karolinska SleepinessScalewasperformedevery5minwhiledriving.Theresultsarepre-sentedinTable1andinFig.2.Thelevelofsleepinessappearedtobe veryhighlyelevatedduringthenightdriveandthetimeofdayeffectwasstrong.Theriseinsleepinessduringeachdrivealsobecamesig-nificant,indicatingthatthedrivingtaskwashighlysleep-inducing.Thesignificantinteractionsconfirmthatelderlyandyoungsubjectsreacteddifferentlyintheirperceptionofsleepinessduringthenightdrive.Theyounggroupreachedconsistentlyhigherlevelsofsleepi-nessatnightthantheelderlygroupandtheeffectwaspronouncedtowards the end of the drive.  1004  A. Lowden et al. / Accident Analysis and Prevention 41 (2009) 1001–1007   Table 1 Results from a three-way ANOVA using outcome values of EEG frequency bands, sleepiness (KSS), saliva cortisol and driving performance (lateral position) obtained duringan early evening and late night 45-min drive in a simulator,  n =10 ( n =9 in EEG measures).Condition Time Age C × A T × A C × T C × T × ATheta (4–8Hz) a – – – – – – –Alpha 1 (8–10Hz) a 7.86 * 5.76 * – – – – –Alpha 2 (10–12Hz) a – 6.26 ** – 7.01 * – – –Sigma 1 (12–14Hz) a – 3.37 * 10.28 ** 6.12 * 3.07 * – –Sigma 2 (14–16Hz) a 5.18 * – 8.56 ** 7.18 * – – –Beta 1 (16–24Hz) a 10.47 ** – 8.89 ** – – – –Beta 2 (24–32Hz) a 6.84 * – 5.01 * – – – –Total (4–32Hz) a 7.73 * 5.18 * – 4.97 * – – –KSS 77.38 *** 53.13 *** – 4.91 * 3.19 * – 2.78 * Cortisol (1/sqrt) – 6.12 * 15.1 ** – 5.05 * 11.2 ** –Lateral position (SD) 6.26 * 11.57 *** – – – 3.83 * ––=n.s. ANOVAs using the factors of condition (C, evening/night drive), age (A, young/elderly) and time (T, 9 times during each 45min drive). a Data based on 4-s, artefact-free, 5-min periods, relative to first 5min/standard deviation for that period. *  p <0.05. **  p <0.01. ***  p <0.001 Only one age difference was found in self-evaluations of symp-toms during the drive. The young group felt more boredom bothduringboththeeveningandnightthedrivingtask( F  =6.5,  p =0.020,mean young=5.3 ± 0.3, elderly=3.6 ± 0.4).A significant effect of age was found for levels of cortisolobtained before and after each drive (see Table 1 and Fig. 2). In all Fig. 1.  Standardized means and standard errors of 5-min bins (  x -axis) during anearly evening (left in figure) and late night (right) 45-min drive in a simulator. Firstpanel, theta 4-8Hz of spectral frequency bands relative first 5-min drive dividedby the SD of that period (  y -axis). Second panel, alpha 8–12Hz bands. Third panel,sigma 12–16Hz bands. Fourth panel, beta 16–32Hz bands. Fifth panel, KarolinskaSleepiness Scale (KSS) of ratings given every 5min during the drive. Sixth panel,driving performance (SD lateral position). Young,  n =10 (open circles) and elderlysubjects,  n =10 (filled circles,  n =9 in EEG measures). measurements, mean levels were higher within the elderly group,corresponding to the significant main age effect. The significanttime effect formed a U-shaped pattern, levels being somewhatelevated before the evening drive and after the night drive. Thesignificant interaction reflects the increase in cortisol within theelderly group after night driving.Driving performance, see Table 1 and Fig. 1, as represented by the lateral position (SD), showed that the conditions differed, thenight condition being worse. Performance within each conditionalso became worse, particularly at night, since the interaction of condition and time became significant. However, no age effects ondriving performance were observed.Since half the group had an earlier start of the drive of 1.5h, thevariables in Table 1 was also tested against a new grouping factor(early or late start of the drive). No significant effects were derivedfrom the factor start of the drive, either for the group factor or inthe interaction with time across the drive. 4. Discussion Theaimofthestudywastoinvestigatesleepinessdevelopmentanddrivingperformanceinyoungandelderlydriverswhendrivingin the evening and at night. In summary the results showed, as Fig. 2.  Salivary cortisol measures (nmol on  x -axis) taken before and after an earlyeveningandlatenightdriveinacarsimulator.Young, n =10(opencircles)andelderlysubjects,  n =10 (filled circles).   A. Lowden et al. / Accident Analysis and Prevention 41 (2009) 1001–1007   1005 expected, a clear effect of condition (time of day) and increase insleepiness across the duration of drive. Age differences were foundfor EEG measurements, elderly subjects showing increased powerin frequency bands above 12Hz particularly in the sigma 1 band(12–14Hz). The difference could indicate a particular sensitivity toageinsustaineddriving.PossiblytheincreaseintheEEGsigmabandcould represent a protection against severe sleepiness, along withthe observed elevated cortisol values within the elderly group. Itseems likely that young drivers could more rapidly develop severesleepiness during prolonged driving, also affecting incidents andaccidents.Earlier works on nighttime driving have reported increases insubjective evaluations of sleepiness and in an EEG increase of alpha(theta)poweractivity(CailleandBassano,1977;TorsvallandAkerstedt,1987;KecklundandAkerstedt,1993).TheseEEGchangesatnighthavealsobeenassociatedwithworsedrivingperformance(Wylie et al., 1996; Gillberg et al., 1996). In the present study there was an abundant increase in the frequencies associated with out-rightsleepandfallingasleep(5–7Hz).Theincidenceofsleepwithineach 45-min drive was also low and only one subject fell asleep(during the evening drive). This likely explains why the effects ontheta power became insignificant. An increase in alpha activityhas also been reported in many studies of vigilance performance(review by Oken et al., 2006) and also in sustained monotonous driving(HorneandBaulk,2004;Otmanietal.,2005),andtheresults of the present study confirms the notion that an increase in alphapower is associated with increases in sleepiness. Thus, the signifi-cant differences between the evening and night drive in the alphabands(8–12Hz)couldapplytothedifferentdegreesofwakefulnessof the brain, the night drive inducing more sleepiness in the wakeEEG.Onestudy,whichcomparedthreeagegroupsinanexperimentalcar simulator study at night, found a positive correlation ( r  =0.41)betweenspectralpowerinthealphaband–usedtomeasuresleepi-ness–andincidentsofrunningofftheroad(Campagneetal.,2004).Although all groups made errors, only the middle-aged (40–50years) and the young (20–30 years) groups, but not the elderlygroup (60–70 years), showed significant correlations. For the lastgroup a significant correlation between errors and EEG sleepinesswere found for periods of theta activity. The authors explain theresults by suggesting a slower impact of low vigilance on drivingperformance by age. Another reason for the difference could be adiminishing ability to produce alpha activity by age but this rela-tionship is not clearly demonstrated for the total EEG alpha band(8–12Hz).Another study performed at rest stops on highways in Francefound that young drivers (<30 years) showed better reaction timeperformanceafter2–4hofdrivingbutlargerdecrementsafter8hof drivingthanolderdrivers,indicatinggreatervulnerabilitytofatiguein young people (Philip et al., 1999). In a follow-up study, sleep deprived young (20–25 years) and elderly (52–63 years) subjectswere tested on a reaction time test (Philip et al., 2004). Although both groups recorded similar performance impairment, the younggroup showed a 70% decrement, compared to only 15% for theelderly group. An additional, age-related, difference was the ini-tially longer (worse) reaction times before sleep deprivation in theelderlygroup.Thestudystressedthesignificanceofself-perceptionofperformanceandtheriskamongyoungdriversofoverestimatingperformance during sleep loss.The increase in sleepiness during the night drive was expected,but the lack of effects for lateral position is problematic, indicatingthat this measurement did not develop sufficiently strong sensi-tivity within the rather short period of 45min of driving. On theotherhand,therewasasignificantinteractionofconditionandtimethat still underscores the relevance of comparing this particularperformance measurement with sleepiness obtained in the study.The present study shows that the elderly group reported lesssleepiness at night. This could partly be explained by higher lev-els of brain activity in the 12–32Hz power spectrum among theelderlygroupcomparedtotheyounggroup.Thisincreaseofhigherfrequency spectral power (>12Hz) could be associated with wake-fulness and activation of the neocortex through activation of thecorticothalamicsystemresultinginadesynchronizationoftheEEG(Oken et al., 2006). The higher levels of cortisol among older sub-  jects could form part of this brain activation. There seem to existan association between EEG sleep stages, sleep propensity and thehypothalamo–pituitary–adrenocortical system although the circa-dian patterns seem to be rather independent (Steiger, 2002). Sleep depth in normal subjects have been associated with low cortisollevels (Gronfier et al., 1999) and high cortisol groups associated withshortsleep.Elderlyhavebeenfoundtohavepersistenthighermean nocturnal levels of cortisol (Van Cauter et al., 1996). Simi- lar findings are found for patient groups where hypercortisolismis associated with shallow sleep, and it is evident that cortisolpromotes vigilance and impairs sleep (Steiger, 2002; Chang andOpp, 2001). But knowledge of the mechanisms of firing intensitiesamong cortical neurons is still limited in the living brain (Steriade,2001).It is also possible that brain activity during wakefulness willshow age-related changes. Increases in faster frequency bands(beta)havebeenshowntocorrelatetoaging(Duffyetal.,1984)but the expected reduction of slow frequency activity in elderly werenot observed in the present study.Thepresentstudywasperformedduringlatenighthourswherea reduction of the circadian drive for sleep is present for elderly(Dijk and Duffy, 1999). The possibility that age differences were related to background factors cannot be ruled out. From the back-ground questionnaires it was recognized that the young group hadgreatersleepneedanddelayedtheirrisingtimesonfreedayscom-paredtotheelderlygroup.Ontheotherhand,intermsofthesleepdata, the groups did not differ and subjects appeared to follow theinstruction to go to bed at around midnight for 3 days precedingthe experiment. But given the lack of more precise measurements,suchasactigraphydata,wecannotcompletelyruleoutthepossibil-ityofthegroupshavingdifferentsleep/wakehistoriesandcircadianphases when entering the study. But on the other hand, the indexofdiurnaltype(TorsvallandÅkerstedt,1980)didnotshowanydif- ferences between age groups. It cannot be ruled out that a phasedifference between groups could explain some of the differencesincortisolandsubjectivesleepiness(KSS)betweenagegroupsandsince we know that modulation of sleepiness is also partly relatedto the suprachiasmatic nucleus-circadian regulating system (Dijkand Czeisler, 1995).The order for presentation of the early and late drive was notbalanced. The results of the second drive could have been influ-enced by the first drive. In the present design it was regarded as abenefit to have the same EEG hookup of the first and second drivesincetheanalysisalsocontainsaFFTanalysisbeingmuchsensitiveto differential interferences of background noise for each hookup.The drives were also separated by at least 6h with the possibilityto rest and have a meal inbetween. The subjects sleep history wasinthissetupsimilarfortheearlyandlatedrivewhichwouldnotbegrantedbylettingsubjectsmakeadditionalvisitstothelaboratory.Finally the behaviour of dual drives is normal among groups goingback and fourth from a night shift and could thus add face validityto the results.The data very clearly demonstrate that sleepiness levelsincreased during the course of the 45-min drive both during theevening and night drive. The design of the study that restrictedanalysis to 45min thus seemed to be relevant. On the other hand,sleepiness was not maximized within the elderly group since KSSmeanratingsreachedonlyslightlyabove7onthescale.Severelev-
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks