Lessons Learned Regarding Simulator Sickness in Older Adult Drivers

(This is a sample cover image for this issue. The actual cover is not yet available at this time.) This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright Author's personal copy Accident Analysis and Prevention 53 (2013) 127–132 Contents lists available at SciVerse ScienceDirect Accident Analysis and Prevention journal homepage: www.elsevier.com/locate/aap The use of adaptation to reduce simulator sickness in driving assessment and research Joshua E. Domeyer a , Nicholas D. Cassavaugh a,b,∗ , Richard W. Backs a,b a b Department of Psychology, Central Michigan University, Mt. Pleasant, MI 48859, United States CMU Center for Driving Evaluation, Education and Research, Mt. Pleasant, MI 48859, United States a r t i c l e i n f o Article history: Received 4 March 2011 Received in revised form 27 November 2012 Accepted 31 December 2012 Keywords: Simulator sickness Motion sickness Driving Simulation Older adults a b s t r a c t The technical advancement of driving simulators has decreased their cost and increased both their accuracy and fidelity. This makes them a useful tool for examining driving behavior in risky or unique situations. With the approaching increase of older licensed drivers due to aging of the baby boomers, driving simulators will be important for conducting driving research and evaluations for older adults. With these simulator technologies, some people may experience significant effects of a unique form of motion sickness, known as simulator sickness. These effects may be more pronounced in older adults. The present study examined the feasibility of an intervention to attenuate symptoms of simulator sickness in drivers participating in a study of a driving evaluation protocol. Prior to beginning the experiment, the experimental groups did not differ in subjective simulator sickness scores as indicated by Revised Simulator Sickness Questionnaire scores (all p > 0.5). Participants who experienced a two-day delay between an initial acclimation to the driving simulator and the driving session experienced fewer simulator sickness symptoms as indicated by RSSQ total severity scores than participants who did not receive a two-day delay (F(1,88) = 4.54, p = .036, partial 2 = .049). These findings have implications for improving client well-being and potentially increasing acceptance of driving simulation for driving evaluations and for driving safety research. © 2013 Elsevier Ltd. All rights reserved. 1. Introduction According to statistics from the Federal Highway Administration, from 1997 to 2010 there was a 28 percent increase in the number of licensed drivers over the age of 65 years old. The increase in older licensed drivers is of concern because these drivers may eventually suffer from cognitive declines due to aging (Craik and Salthouse, 2007) that may affect their driving performance. While older drivers are not necessarily more likely than younger drivers to be involved in a crash, they are more likely to be killed or injured in such a crash. It is important to help older drivers avoid crashes while helping them maintain the ability to drive safely. This will necessitate research on and evaluation of older drivers in a way that is safe and informative. One approach to this is to use simulators because it is possible to standardize and quantify evaluation procedures while in a safe environment. Unfortunately, some people, notably older adults, experience simulator sickness in simulators. In fact, the authors have noted considerable participant attrition rates in simulator-based experiments. Thus, making ∗ Corresponding author at: Department of Psychology, Central Michigan University, Mt. Pleasant, MI 48858, United States. Tel.: +1 989 774 2882. E-mail addresses: Cassa1nd@cmich.edu, xunil2@gmail.com (N.D. Cassavaugh). 0001-4575/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.aap.2012.12.039 driving simulators more accessible to older adults might be viewed as advantageous for driving safety and accident prevention. Simulator sickness has been described as a unique form of motion sickness (Rizzo et al., 2003). Many studies and anecdotal reports indicate that older adults suffer greater simulator sickness symptoms than other groups (Stanney et al., 2002; Freund and Green, 2006; Mullen et al., 2010). Indeed, in a study examining lefthand turning ability older adult attrition was 40 percent whereas younger adult attrition was 14 percent (Edwards et al., 2004). These data and the authors’ own observations suggest a need to attenuate simulator sickness in order to make simulation available to more drivers. It is therefore important to explore the phenomenon of simulator sickness and attempt to understand interventions that have utility for reducing it. 1.1. Simulator sickness Simulator sickness is a phenomenon that is affected by simulator features and participant characteristics. It produces symptoms that are similar to, but typically less severe than, those of motion sickness such as nausea, ocular discomfort, and disorientation (Kennedy et al., 1993). In this regard, the sensory conflict theory of motion sickness provides an important framework for understanding simulator Author's personal copy 128 J.E. Domeyer et al. / Accident Analysis and Prevention 53 (2013) 127–132 sickness (Reason and Brand, 1975). The theory states that the symptoms of motion sickness are a result of conflicting visual and vestibular cues. In motion sickness, such as that experienced by some while attempting to read in a moving vehicle, vestibular motion cues are coupled with an absence of visual motion cues. In contrast, in simulator sickness visual motion cues are coupled with an absence of vestibular motion cues. Not every individual experiences simulator sickness to the same extent, even in identical situations. Reason’s (1978) neural mismatch model of sensory conflict theory states that susceptibility is a product of an individual’s overall experience with motion sickness. This would suggest that older adults would be less susceptible to motion sickness than younger adults because they have more exposure to situations that would produce motion sickness. In contrast, several studies have found that older participants are more susceptible to simulator sickness than other age groups (Edwards et al., 2004; Park et al., 2004; Brooks et al., 2010). This may be due to a lack of experience with simulated environments on the part of older adults. If this is indeed the case, then providing some experience with the simulator prior to prolonged exposure in research or clinical settings may improve tolerance. Providing longterm exposure to a clinical simulator is not a practical solution, given the required costs and the need for the client to return for many sessions. One approach to reduce symptoms of simulator sickness is the use of adaptation or time delay. Results from several studies have shown decreased simulator sickness symptoms with repeated exposure within and between days (Gower et al., 1987; Hu and Stern, 1999; Hill and Howarth, 2000; Howarth and Hodder, 2008; Teasdale et al., 2009). This reduction in simulator sickness symptoms due to time delay between simulator sessions has been shown to persist up to a month or longer (Hu and Stern, 1999). Howarth and Hodder (2008) found that simulator sickness symptoms decreased over 10 days of simulator exposure with a session on each day. Teasdale et al. (2009) found that older adults’ (ages 65–84 years old) simulator sickness symptoms as measured by a subjective questionnaire decreased over subsequent simulator sessions. They found that older adults adapted to simulation over several sessions. After the fifth session the older adults did not differ from the initial baseline condition on simulator sickness scores. These results indicate that adaptation could be used to attenuate simulator sickness symptoms. Therefore, the goal of the present research was to determine whether a short-term exposure session which preceded the clinical exposure session by two days would improve tolerance. 1.2. Present study For the present purposes, we will use the term acclimation for the name of the first brief exposure to the simulator that was used to allow the participants to adjust to the simulator. We will use the term adaptation to mean an initial, brief simulator acclimation session followed two days afterward by the full simulator test protocol. We should also note that Howarth and Hodder (2008) have used the term habituation to indicate adaptation with a time delay. To avoid confusion with the many definitions of that term, we will be using “adaptation”. The present study was conducted in the context of a larger project to validate a clinical driving evaluation protocol. Investigation of the adaptation’s effectiveness was a secondary aim of that larger project and placed several constraints upon the design. A complete description of the validation study is beyond the scope of this paper, but descriptions of the cognitive tasks and simulator task sessions can be found in Tuttle et al. (2009) and Backs et al. (2011). Fig. 1. Order of tasks and RSSQ administrations by session and group. We examined the effects of an adaptation on self-reported simulator sickness scores while in a high fidelity driving simulator. Participants were given a simulator sickness acclimation before beginning a cognitive evaluation session and again before beginning the driving session in the simulator. The sessions were administered on separate days. The order of these sessions was counterbalanced between participants. Thus, some participants participated in an acclimation session two days before a driving session (i.e., the adaptation; the Lag group) and others participated in the driving session immediately after the acclimation (the immediate group). Fig. 1 shows the order of the two groups. Revised Simulator Sickness Questionnaire (RSSQ) scores were obtained before and after the acclimation during the cognitive session and before and after the driving session which also contained an acclimation. Thus, we obtained four scores for each individual. 1.3. Hypotheses As the main goal was to determine whether the adaptation would reduce simulator sickness symptoms, the primary hypothesis relates to the effect of the adaptation. Prior work noted above also suggested that there would be an age effect. We investigated both hypotheses. 1.3.1. Adaptation The main prediction involved post-driving session RSSQ scores. We predicted that the Lag group would report lower post-driving RSSQ total severity scores than the immediate group. Furthermore, we predicted a similar significant effect of group on RSSQ nausea scores. Based on unpublished data from our lab, we made no predictions of the effect of group (lag, immediate) on disorientation, ocular discomfort or strain/confusion RSSQ scores. 1.3.2. Age Because of prior experience with participants of varying ages in simulated driving in our laboratory (Cassavaugh et al., 2009; Domeyer, 2009), we predicted a significant effect of age on total severity and nausea with older adults reporting higher scores (more sickness). Again, based on findings from Domeyer (2009) we did not predict a similar effect on the other subscale scores (disorientation, ocular discomfort, and strain/confusion). 2. Materials and methods 2.1. Participants Total participants were 120 (40 young; 40 middle; 40 old) individuals recruited for a driving assessment protocol validation. They were either students from Central Michigan University recruited through the psychology subject pool or members of the community recruited through flyers placed at surrounding organizations. Students were given course credit for their participation. Members of the community were paid $28 per hour for their participation. Participants were assigned to groups by order of recruitment. Author's personal copy J.E. Domeyer et al. / Accident Analysis and Prevention 53 (2013) 127–132 The young group consisted of 24 females and 16 males ranging from 18 to 28 years of age (M = 21, SD = 2 years). The middle-aged group consisted of 25 females and 15 males ranging from 30 to 58 years of age (M = 46, SD = 9 years). The older group consisted of 24 females, 13 males, and 3 for whom sex was not recorded ranging from 60 to 90 years of age (M = 70, SD = 8 years). 2.2. Research design Participants performed a series of driving acclimation scenarios in both the cognitive and driving evaluation sessions. Participants reported levels of simulator sickness symptoms before and after simulation on each day (Fig. 1). 2.3. Apparatus Data were collected in the AAA Michigan Driving Simulator. The simulator is a DriveSafety DS-600c driving simulator (DriveSafety Corporation) which provides a 180-degree forward field of view. Side-view and rear-view mirrors are implemented as LCD displays. The simulator is equipped with a 1.5 degree-of-freedom motion base, which provides motion cues in pitch with minimal forward/reverse translation and pitch washout after initial change in velocity. The cab of the simulator is the forward passenger cabin of a compact car. 2.4. Measures 2.4.1. Revised Simulator Sickness Questionnaire Simulator sickness scores were recorded using the RSSQ (Kim et al., 2004). The RSSQ is a modified version of Kennedy et al.’s (1993) Simulator Sickness Questionnaire (SSQ) and extends the scope of the SSQ by: (a) changing the scale from an interval scale between 1 and 4 to an interval scale between 1 and 10, (b) adding 8 additional items (drowsiness, visual flashbacks, stomach awareness, confusion, vomiting, pallor, difficulty equilibrating, muscle stiffness for strain), and (c) adding an additional subscale (strain/confusion). The resulting questionnaire provides a total severity score as well as subscale scores measuring nausea, disorientation, ocular discomfort and strain/confusion. Both the SSQ and RSSQ drew on the Motion Sickness Questionnaire (Kellog et al., 1965) for the initial list of symptoms. Kim et al. (2004) found that the correlation between the RSSQ and SSQ scales was high for total severity and the common subscales (r > .70 for total severity and the common subscales). The nausea, ocular discomfort, disorientation, strain/confusion, and total severity scales are weighted in the RSSQ. Conclusions by Kim et al. (2004) suggest that the RSSQ should be treated as a partial replication and extension of the SSQ. Readers are referred to the Kim et al. (2004) paper for more details pertaining to the RSSQ. 2.5. Procedure Data collection took place at Central Michigan University’s Center for Driving Evaluation, Education and Research. The validation consisted of two sessions conducted on different days. The sessions were separated by one day. That is, if a participant performed the first session on Monday, then that participant performed the second session on Wednesday of the same week. One session consisted primarily of computer-based cognitive testing (the “cognitive session”). The other session consisted primarily of simulated driving in a number of scenarios designed to test driving performance (the “driving session”). The order of the sessions was counterbalanced. Testing on either day was immediately preceded by an acclimation to the driving simulator. See Fig. 1 for clarification on the order of blocks in the study. 129 Table 1 Description of the driving segments in the third acclimation scenario in which participants were in full control of the simulator. Speed limit Distance (km) Intersection 56.3 km/h (35 mph) 72.4 km/h (45 mph) 88.5 km/h (55 mph) 72.4 km/h (45 mph) 0.6 1.0 1.0 1.2 Four-way with four-way stop signs Four-way with two-way stop signs Four-way with traffic light T intersection with stop sign The cognitive session required approximately 1.5 h. Participants first completed a 10 min acclimation in the driving simulator. The RSSQ was administered before and after the acclimation in the cognitive session. They then completed neuropsychological tasks and the cognitive battery (Tuttle et al., 2009). When the tests were completed, participants drove an approximately 10-min acclimation in the driving simulator. Then, participants completed 11 different scenarios in the driving simulator to measure driving skills such as situation awareness, divided attention, and planning. The RSSQ was administered before the acclimation and after completing the tasks in the driving simulator. 2.5.1. Acclimation procedure The acclimation that was administered at the start of each session consisted of four separate drives. In each drive, the roadway was a two-lane rural roadway with hard shoulder. Participants were permitted to repeat each scenario until they were comfortable with it. This was a subjective evaluation on the part of the participant. Most participants requested only a single run in each acclimation scenario. First, participants sat in the simulator while custom software controlled speed and steering. We refer to this scenario as the “autodrive” scenario. Experience has shown the authors that most, if not all, participants tend to over-control the simulator upon first exposure, producing significant motion cues and the potential for immediate, severe simulator sickness followed quickly by exiting the experiment or clinical session. This allowed participants to acclimate to the motion associated with starting and stopping in the driving simulator. The target speed was 96.6 km/h (60 mph) and the distance was 1.8 km. In the second scenario, participants controlled the steering while the software controlled speed. The environment, target speed, and distance were the same as “autodrive”. We refer to this scenario as the “cruise control” scenario. In the third and fourth drives, participants were in full control of the simulator. In the third, participants were asked to drive a long section of straight road with intersections, stop signs, traffic lights and speed limit changes. We asked them to obey traffic laws, adjust their speed accordingly and stop at stop signs and traffic lights. Details are provided in Table 1. The fourth and final acclimation scenario consisted of a single four-way intersection. Participants started 22 meters from the intersection, approached the intersection, stopped, and turned either right or left as instructed. After the turn, the simulation ended. Participants practiced turning multiple times to become familiar with the simulator’s handling and turning characteristics. In most cases two turns in each direction was sufficient for participants to learn how to execute the turns properly. 2.5.2. Analysis procedure The total severity and subscale scores were computed with the RSSQ using the method outlined by Kim et al. (2004). The individual symptoms were first weighted by values that were determined by subject matter experts during the creation of the RSSQ. The weightings represent the expert ratings of the importance of the contributions of a particular symptom to simulator sickness. The Author's personal copy 130 J.E. Domeyer et al. / Accident Analysis and Prevention 53 (2013) 127–132 scales were then adjusted to have equal variance in the method outlined by Kennedy et al. (1993) for the SSQ. This results in separate total severity, nausea, disorientation, ocular discomfort and strain/confusion scores. For the primary hypotheses which related to the effects of age and group we used an ANCOVA to account for any baseline differences. Post-driving RSSQ scores (D4 and D2) were analyzed with a 3 (age: young: 18–29 years, middle: 30–59 years, old: 60 + years) × 2 (group: lag vs. immediate) ANCOVA with baseline score (C1 and D1) as a covariate. The respective covariates were significant for all of the scales (p < .05). We report adjusted means and SEs for ANCOVA analyses. For analysis of baseline RSSQ scores (C1 and D1 from Fig. 1), we used a 3 (age: young: 18–29 years, middle: 30–59 years, old: 60 + years) × 2 (group: lag vs. immediate) ANOVA. Analyses were performed for each RSSQ scale (total severity, nausea, disorientation, ocular discomfort, and strain/confusion). Twenty-five people exited the experiment due to discomfort and were excluded from the analysis due to incomplete data sets. These individuals were 5 young, 8 middle-aged, and 12 older adults. Thus, the final sample included 35 young (males = 16, females = 19), 32 middle-aged (males = 13, females = 19) and 28 older adults (males = 10, females = 18). Table 3 Baseline RSSQ total severity score means (SDs) in each group by age. Group Age Immediate Lag Overall Young (N = 35) Middle (N = 32) Old (N = 28) 5.03 (7.63) 3.46 (5.25) 4.31 (6.60) 1.87 (2.40) 1.25 (1.97) 1.52 (2.15) 0.59 (.940) 1.28 (1.80) 0.93 (1.45) 3. Results in total severity (F(2,89) = 5.40, p = .006, partial 2 = .108) showing that younger adults (M = 4.31, SD = 6.60) had higher baseline RSSQ scores than older adults (M = 0.93, SD = 1.45)or middle-aged adults (M = 1.52, SD = 2.15).Baseline analyses for total severity are presented in Table 3. The effect of age was not significant for the nausea subscale in the baseline RSSQ scores (F(2,89) = 2.58, p = .081, partial 2 = .055). There was a significant effect of age on baseline RSSQ scores for ocular discomfort (F(2,89) = 3.42, p = .037, partial 2 = .072) and strain/confusion (F(2,89) = 3.44, p = .036, partial 2 = .072) scales. Younger adults indicated more symptoms than older adults. This result is interesting because the effect of age that existed prior to simulation was eliminated by taking into account baseline scores. Further exploratory analysis revealed that the effect might persist after exposure to the simulator but was not significant due to large variance. More research is needed to determine whether this apparent effect is genuine. 3.1. Age 3.2. Adaptation We predicted that there would be a significant effect of age on the total severity and nausea scales of the RSSQ with older adults indicating more symptoms than younger adults. Participants’ baseline RSSQ scores (C1 and D1) were used as a covariate. Contrary to our prediction, there was no effect of age on the total severity scale after exposure to the simulator when taking into account the baseline scores (F(2,88) = 0.49, p = .659, partial 2 = .009). Additionally, there was no effect of age on nausea when taking into account baseline scores (F(2,88) = 0.18, p = .834, partial 2 = .004). In other words, the simulated drive did not affect each age group differently. None of the other subscales attained significance for an effect of age after simulation while accounting for baseline scores. Table 2 shows the ANCOVA results for all of the subscales. After answering the primary hypothesis, we wanted to know if there was an effect of age in the baseline RSSQ scores (C1 and D1). The effect of age was significant for the baseline analyses We predicted that there would be significant post-driving differences in RSSQ scores between the experimental groups in the total severity and nausea subscales of the RSSQ after taking into account their baseline states. Participants’ baseline RSSQ scores (C1 and D1) were used as a covariate. Consistent with our prediction, the effect of group was significant in total severity (F(1,88) = 4.54, p = .036, partial 2 = .049) with higher adjusted scores for the immediate group (M = 12.97, SE = 1.87) than the lag group (M = 7.37, SE = 1.84). Significantly lower scores in the lag group suggest that the adaptation had an effect in reducing overall symptoms. Contrary to our prediction the effect of group on the nausea subscale did not attain significance (F(1,88) = 2.22, p = .139, partial 2 = .025). ANCOVA results and effect means are presented in Table 4.There was also a significant effect of group for post-simulation scores on disorientation (F(1,88) = 4.66, p = .033, partial 2 = .050), ocular discomfort (F(1,88) = 6.02, p = .016, Table 2 Overall RSSQ total severity and subscale ANCOVA results. (Sub)scale Variable F p Partial 2 Nausea Group (df = 1,88) Age (df = 2,88) Group × age (df = 2,88) 2.22 0.18 0.77 .139 .834 .463 .025 .004 .017 Disorientation Groupa (df = 1,88) Age (df = 2,88) Group × age (df = 2,88) 4.66 2.71 2.06 .033 .072 .132 .050 .058 .045 Ocular discomfort Groupa (df = 1,88) Age (df = 2,88) Group × age (df = 2,88) 6.02 0.06 1.50 .016 .940 .227 .064 .001 .033 Strain/confusion Groupa (df = 1,88) Age (df = 2,88) Group × age (df = 2,88) 4.15 0.31 1.45 .044 .727 .240 .045 .007 .032 Total Severity Groupa (df = 1,88) Age (df = 2,88) Group × age (df = 2,88) 4.54 0.49 0.41 .036 .612 .659 .049 .011 .009 a p < .05. Author's personal copy J.E. Domeyer et al. / Accident Analysis and Prevention 53 (2013) 127–132 131 Table 4 Adaptation RSSQ subscale score ANCOVA-adjusted means and standard errors. Group Nausea Disorientationa Ocular discomforta Strain/confusiona Total severitya Lag (D4) Immediate (D2) 7.38 (1.74) 11.10 (1.76) 9.41 (2.21) 16.25 (2.24) 7.84 (2.02) 14.93 (2.05) 6.17 (1.32) 10.02 (1.34) 7.37 (1.84) 12.97 (1.87) a p < .05. partial 2 = .064) and strain/confusion (F(1,88) = 4.15, p = .044, partial 2 = .045) further supporting the claim that the lag group reported fewer symptoms than the immediate group. 4. Discussion The present findings are consistent with other work (Howarth and Hodder, 2008; Teasdale et al., 2009) and further demonstrate that a single brief exposure to the simulator followed by a day with no exposure (i.e., adaptation) produces declines in report of simulator sickness symptoms. Consistent with our hypothesis we found that total severity scores decreased with an adaptation to the simulator. Unfortunately, we did not find the predicted decrease in nausea scores with the adaptation. Adaptation as an intervention is more amenable to clinical schedules than a schedule including multiple exposures per day or multiple single exposures over several days. The results reported here support the notion that simulator sickness is at least partially associated with the participant’s level of experience with the simulated environment. We did not find that older adults were different from middleaged or young adults in post-simulation simulator sickness ratings. Interestingly, older adults indicated fewer symptoms in their baseline (pre-driving) simulator sickness scores than younger adults. This may warrant further investigation to determine if the difference in baseline scores is real or an artifact. The differences in baseline simulator sickness ratings potentially indicate differences in how young and older adults rate their overall wellness which may be associated with health variability. That is, older adults experience more variability in health and may therefore rate their baseline state more favorably than young adults. Further research is needed to determine whether age and group (lag or immediate) interact. These findings suggest an avenue by which simulated environments may be made accessible to a population which currently may not be taking advantage of them. Indeed, given the rise in consumer-grade 3-D entertainment such as TVs and computer displays this type of situation (referred to as cybersickness in this context) may become more prevalent (O’brien and Baime, 2010). Research findings such as ours suggest that gradual exposure and experience with these types of devices may alleviate symptoms of cybersickness over time. Similarly, simulator sickness as a phenomenon may effectively disappear over time as experience with 3-D virtual environments continues to increase among the general population. Because this study was conducted as part of a larger project with other goals, the adaptation was confounded with the number of actual acclimations the participant received. That is, those participants in the lag group experienced the acclimation sequence twice prior to performing the driving session. This meant that they received an acclimation on day one, no session on day two, and an acclimation and driving session on day three. Future research should account for this confound by eliminating the pre-driving adaptation session for the lag group. Future research should also determine the benefits of reducing simulator sickness among clients. Fortunately, the presence of simulator sickness has not been shown to affect performance in driving simulators (Mullen et al., 2010). However, reducing simulator sickness could result in increased well-being or even acceptance of simulation as a valid assessment tool. It is because of these reasons that we feel further exploration of simulator sickness is warranted. However, there may be other effects associated with a simulated environment, separate from simulator sickness, that researchers should be concerned with. For instance, Muth and colleagues (Muth et al., 2006; Muth, 2009) have identified effects associated with uncoupled motion that affect performance on some cognitive tasks, that cannot be attributed to motion sickness, and that persist for a number of hours after exposure. Such effects may be insidious in that they are not immediately noticeable because they are not necessarily accompanied by simulator sickness symptoms. A driver suffering from such effects after participating in research or assessment in a simulator could pose a danger to him/herself or others on the road. Adaptations such as the one proposed here could help alleviate those problems, but future research is warranted to determine if this is case. The number of older drivers is expected to continue to increase in the coming years and the demand for driving assessments is likely to increase. When a driver’s safety behind the wheel is in question to begin with (as might be the case if a driver was referred for an evaluation), the wisdom of performing an on-road driving assessment is questionable. Even closed-circuit tracks may not be the wisest choice in that situation. A shortened adaptation schedule may open the door to increased use of simulators in driving assessment and in post-assessment remediation of driving performance deficits. This might in turn help reduce the numbers of accidentinvolved senior drivers, thus reducing injuries, fatalities, and costs associated with them. In addition, simulators are useful for training novice drivers in basic vehicle control and hazard perception. Many of them may also be susceptible to simulator sickness and would benefit from an adaptation protocol that would make the simulator accessible to them. In each case (older, experienced drivers and younger, novice drivers), a reduction in accident involvement could be expected as a result. Acknowledgments We wish to thank Stephanie Tuttle and Davis Conley for their assistance with data collection. We also wish to thank Dr. Linda Boyle and 2 anonymous reviewers for their helpful comments. An earlier version of these data were presented at the Association for the Advancement of Automotive Medicine 54th Annual Conference. This research was funded by a Central Michigan University Vision 2010 award to Richard W. Backs. References Backs, R.W., Tuttle, S., Conley, D., Cassavaugh, N.D., 2011. Attention factors compared to other predictors of simulated driving performance across age groups. In: Proceedings of the Sixth International Driving Symposium on Human Factors in Assessment, Training, and Vehicle Design, Iowa City, IA. Brooks, J.O., Goodenough, R.R., Crisler, M.C., Klein, N.D., Alley, R.L., Koon, B.L., Logan, W.C., Ogle, J.H., Tyrrell, R.A., Willis, R.F., 2010. Simulator sickness during driving simulation studies. Accident Analysis and Prevention 42, 788–796. Cassavaugh, N.D., Domeyer, J., Backs, R.W., 2009. The effect of age on decision making during unprotected turns across traffic. In: Fifth International Symposium on Human Factors in Driver Assessment, Training and Vehicle Design, Big Sky, MT, pp. 97–103. Craik, F.I.M., Salthouse, T.A., 2007. The Handbook of Aging and Cognition, third ed. Lawrence Earlbaum Associates, Mahwah, NJ, p. 672. Author's personal copy 132 J.E. Domeyer et al. / Accident Analysis and Prevention 53 (2013) 127–132 Domeyer, J., 2009. The effect of cognitive priming on simulator discomfort. Unpublished Undergraduate Honors Project, Central Michigan University. Edwards, C.J., Creaser, J.I., Caird, J.K., Lamsdale, A.M., Chisholm, S.L., 2004. Older and younger driver performance at complex intersections: implications for using perception-response time and driving simulation. In: Proceedings of the Second International Symposium on Human Factors in Driver Assessment, Training, and Vehicle Design, Park City, UT. Freund, B., Green, T.R., 2006. Simulator sickness amongst older drivers with and without dementia. Advances in Transportation Studies (Special Issue), 71–74. Gower, D.W., Lilienthal, M.G., Kennedy, R.S., Fowlkes, J.E., Baltzley, D.R., 1987. Simulator Sickness in the AH-64 Apache Combat Mission Simulator. USAARL, Fort Rucker, AL. Hill, K.J., Howarth, P.A., 2000. Habituation to the side effects of immersion in a virtual envirionment. Displays 21, 25–30. Howarth, P.A., Hodder, S.G., 2008. Characteristics of habituation to motion in a virtual environment. Displays 29, 117–123. Hu, S., Stern, R.M., 1999. The retention of adaptation to motion sickness eliciting stimulation. Aviation, Space, and Environmental Medicine 70 (8), 766–768. Kellog, R.S., Kennedy, R.S., Graybiel, A., 1965. Motion sickness symptomatology of labyrinthine defective and normal subjects during zero gravity maneuvers. Aerospace Medicine 4, 315–318. Kennedy, R.S., Lane, N.E., Berbaum, K.S., Lilienthal, M.G., 1993. Simulator sickness questionnaire: an enhanced method for quantifying simulator sickness. The International Journal of Aviation Psychology 3, 203–220. Kim, D.H., Parker, D.E., Park, M.Y., 2004. A New Procedure for Measuring Simulator Sickness – The RSSQ. University of Washington Human Interface Technology Laboratory, Seattle, WA, 1–14. Mullen, N.W., Weaver, B., Riendeau, J.A., Morrison, L.E., Bedard, M., 2010. Driving performance and susceptibility to simulator sickness: are they related? American Journal of Occupational Therapy 64, 288–295. Muth, E.R., 2009. The challenge of uncoupled motion: duration of cognitive and physiological after effects. Human Factors 51 (5), 752–761. Muth, E.R., Walker, A.D., Fiorello, M., 2006. Effects of uncoupled motion on performance. Human Factors 48 (3), 600–607. O’brien, M., Baime, J., 2010. Cybersickness: A Virtual Bummer – Science Nation. National Science Foundation, Arlington, VA. Park, G.D., Rosenthal, T.J., Allen, R.W., Cook, M.L., 2004. Simulator sickness results obtained during a novice driver training study. In: Proceedings of the Human Factors and Ergonomics Society 48th Annual Meeting, pp. 2652–2656. Reason, J.T., Brand, J.J., 1975. Motion Sickness. Academic Press, London. Reason, J.T., 1978. Motion Sickness Adaptation: A Neural Mismatch Model. Journal of the Royal Society of Medicine 51, 819–829. Rizzo, M., Sheffield, R.A., Stierman, L., Dawson, J., 2003. Demographic and driving performance factors in simulator adaptation syndrome. In: Proceedings of the Second International Driving Symposium on Human Factors in Driver Assessment, Training and Vehicle Design, pp. 201–208. Stanney, K.M., Kingdon, K.S., Kennedy, R.S., 2002. Dropouts and aftereffects: examining general accesibility to virtual environment technology. Proceedings of the Human Factors and Ergonomics Society 46th Annual Meeting, 2114–2118. Teasdale, N., Lavalliere, M., Tremblay, M., Laurendeau, D., Simoneau, M., 2009. Multiple exposition to a driving simulator reduces simulator symptoms for elderly drivers. In: Proceedings of the Ffifth International Driving Symposium on Human Factors in Driver Assessment, Traioning, and Vehicle Design, Big Sky, MT. Tuttle, S., Cassavaugh, N.D., Backs, R.W., 2009. Attention function and structure of older and younger adult drivers. In: 5th International Driving Symposium on Human Factors in Driver Assessment, Training and Vehicle Design, Big Sky, MT.