Reliability of a virtual reality driving simulator for individuals with neurological disorders

Abstract

BACKGROUND:

Virtual reality (VR) offers an innovative method to assess the impact of neurocognitive deficits on daily activities like driving.

OBJECTIVE:

This study sought to evaluate the reliability of a VR driving simulator (VRDS) in assessing driving behaviors.

METHODS:

Participants included 91 individuals with a neurological condition (e.g., acquired brain injury, multiple sclerosis) and 59 participants with no history of neurological disorder. Internal consistency and split half reliability were examined from data for all participants for driving speed, lane positioning, and steering on a rural segment of the virtual reality driving simulator route. Test-retest reliability was examined for data from the 22 drivers who completed the same route sections across two time periods.

RESULTS:

Internal consistency and split half reliability were excellent (alphas $>$ 0.90) for speed and lane positioning on a basic rural route. Test-retest reliability was variable with adequate to good reliability of lane positioning and steering but poor reliability of speed.

CONCLUSIONS:

Findings suggest that VRDS is a generally reliable instrument for measuring speed, lane positioning, and steering in a virtual basic rural driving environment. Reliability of repeated testing is not consistent, likely due to practice effects, highlighting the importance of using a comparison group in VRDS studies.

Keywords

Virtual reality driving simulator reliability

1. Introduction

Virtual reality (VR) technology offers new opportunities for the development of innovative neuropsychological assessment and clinical tools that can explore cognition in ecologically valid environments [1, 2, 3, 4, 5, 6, 7, 8, 9]. One application of VR is in the use of driving simulation for examining driving capacity in individuals with neurological disorders. An emerging body of research suggests that virtual reality driving simulation (VRDS) is a promising approach for examining the cognitive demands of driving in individuals with neurological impairments [10, 11, 12, 13, 14].

VRDS offers unique advantages for evaluation of driving performance, which can be important for functional independence following neurological compromise. Specifically, VRDS 1) allows for safe assessment of challenging driving scenarios, which are often difficult to capture with traditional behind the wheel assessments; 2) utilizes realistic scenarios that involve complex (i.e., dual-task) behaviors, which allows for nuanced assessment of performance variables not available in current clinical approaches for evaluation of driving; and 3) offers a systematic and objective approach to measuring driving performance in contrast to traditional neuropsychological testing or behind the wheel assessments, which are often subjective and difficult to control [10, 15, 16].

Despite its potential utility and growing research base, clinical use of VRDS remains limited. This may be partially due to the paucity of research regarding the feasibility, validity, and reliability of VRDS as a tool for informing clinical decisions. Discriminant validity, or the ability of VRDS to discriminate between healthy drivers and those with neurological disorders, is the most commonly investigated metric within the current literature [17, 18, 19, 20]. Discriminant validity has also been investigated when comparing performance between healthy drivers and those with visual deficits [21, 22], varying levels of prior driving experience [23], sleep deprivation [24], and among aging populations [17, 25, 26]. Research has also pointed to the external validity of VRDS [27] while studies examining other validity metrics (e.g. internal validity) remain limited.

In addition to validity, it is important to evaluate consistency and reliability of this clinical instrument. VRDS can allow clinicians to examine specific abilities that contribute to the complexity of driving that are captured with several novel variables (e.g. average speed, variability in speed, change in lane position). Establishing internal consistency of VRDS can ensure clinicians that these variables will consistently and equally assess driving performance. VRDS is also clinically beneficial in that it allows for standardized repeated assessment. Establishing test-retest reliability of VRDS would allow for a standardized comparison of baseline and follow-up assessment of the same individual, thereby allowing the use of VRDS for repeated assessment of driving skills including monitoring of change in performance (particularly relevant in studying impaired performance in progressive disorders) and assessment of driver retraining.

The Applied Neuro-technologies VRDS has been used in several research studies to evaluate driving skills in a variety of neurological disorders including multiple sclerosis, acquired brain injuries, attention deficit/hyperactivity disorder, autism spectrum disorder, and post-traumatic stress disorder, as well as typically developing individuals across a wide age range [11, 12, 13, 14, 28, 29]. The current study sought to evaluate the reliability of that VRDS in assessing basic driving behaviors of neurologically compromised and healthy participants.

2. Methods

2.1 Participants

Data for this study were pulled from three separate existing studies that examined simulated driving abilities in individuals with neurological conditions (concussion, acquired brain injury, or multiple sclerosis) compared with neurologically healthy individuals [29, 30]. Specifically, these data were pooled to create a larger sample size given predefined criteria and consistent driving methodologies across the three studies [29]. All three studies included a training session, baseline drive, and drive with secondary tasks on a VRDS, along with completion of neuropsychological tests. Participant demographic information is shown in Table 1. Participants included 46 individuals diagnosed with a mild to severe traumatic brain injury, 2 individuals diagnosed with a brain tumor, 1 individual who sustained a stroke, 42 individuals diagnosed with multiple sclerosis, and 59 individuals with no history of neurological disorder. Diagnoses were corroborated by medical reports and/or physicians who referred individuals to the study. All studies were approved by the Drexel University Institutional Review Board.

Table 1
Participant demographics

	Full sample	$n$	Non-neurological/healthy participants	$n$	Participants with neurological disorders	$n$
Age	18–60 years $M=$ 38.80 SD $=$ 13.54	150	21–60 years $M=$ 40.37 SD $=$ 11.56	59	18–60 years $M=$ 37.78 SD $=$ 14.66	91
Sex	92 females (61%)	150	36 females (61%)	59	56 females (61%)	91
Race	80 white (65%) 41 black (33%) 3 other (2%)	124	26 white (44%) 31 black (53%) 2 other (2%)	59	54 white (83%) 10 black (15%) 1 other (1%)	65
Education	12–22 years $M=$ 15.41 SD $=$ 2.17	113	12–22 years $M=$ 15.60 SD $=$ 2.25	58	12–19 years $M=$ 15.20 SD $=$ 2.08	55

2.2 Procedure

Data were collected as part of several studies examining driving performance in individuals with neurological disorders compared with neurologically healthy controls, all administered in the span of 7 years. Across studies, participants completed several driving tasks on the same VRDS. The simulator displayed a pre-programmed driving environment across three monitors to help provide a deep level of immersion and a believable driving experience. Driver input was provided via a modified steering column and foot pedals adapted from an actual vehicle, and digital sensors tracked the motion inputs. The simulator simultaneously tracked and stored multiple events, including vehicle position, speed, lane deviation, target object distances, raw steering input, steering angle, accelerator position, brake pedal position, turn signal position, and auxiliary data such as key position, gearshift position, and car stereo operation. The simulator’s rendering engine utilized medium resolution geometry and high-resolution textures to provide high quality visuals.

Across studies, all participants completed the following tasks on the Applied Neuro-technologies VRDS:

1.
Training: Participants completed two training sessions: one for practicing basic speed and lane management and another for practicing making turns and stops-which included feedback and coaching regarding driving at posted speed limits, using turn signals, and making complete stops at stop signs.
2.
Basic rural route: Participants completed a basic 8-mile route consisting of a 2-lane highway with few cars or other distractions. The speed limit was 40 mph throughout the route and participants were not required to change lanes or navigate environmental challenges.
3.
Highway: Participants drove on a divided highway with frequent curves and oncoming traffic in the opposite lane. The speed limit increased to 55 mph.
4.
Residential/commercial zone: Participants exited the highway and drove through a residential area where the speed limit reduced to 25 smph. Participants were required to make several turns through a neighborhood that included houses, cars, and pedestrians. They then entered a commercial zone, which included oncoming traffic in the opposite lane, commercial buildings (e.g. restaurants), and parked cars. The speed limit increased from 25 to 35 mph.

For 22 participants with no history of neurological impairment, data to examine test-retest reliability were available from one study in which participants completed a second session 2 to 8 weeks after the first session. During the second session, participants again completed the basic rural drive, highway, and residential/commercial zone. There were several confounding factors limiting the available data for use. First, the data included participants who had sustained concussions as well as neurologically healthy participants. Given that a change in driving performance in patients who were recovering from concussion was expected and therefore likely to produce poor data for examining test-retest reliability, only data from neurologically healthy participants ( $n=$ 22) were included. In addition, the second session included several additional challenges and distractions, meaning participants were likely to demonstrate different driving behaviors. However, the basic rural drive, which occurred at the beginning of the session prior to introduction to new challenges and distractions, was identical to the drive completed during the first session. Therefore, only data from that drive was included.
2.3 Method of analyses

Driving behaviors of interest included speed, deviation from the center of the driver’s lane, and steering wheel movement. Parameters utilized in this study were derived from previous similar driving studies conducted by our group in addition to existing literature indicating specific parameter for differentiating driving performance.

Table 2
Internal consistency for baseline rural drive

Participants	$N$	Cronbach’s alpha
		Mean speed	Variability in speed	Variability in lane position	Variability in steering
Full sample	150	0.98	0.94	0.92	0.95
Non-neurological	59	0.98	0.89	0.89	0.94
Neurological	91	0.99	0.96	0.92	0.96

Note. Demographic information was not available for all participants. Sample sizes for different types of demographic information are displayed.

Table 3

Split-half reliability

Participants	$N$	Spearman brown corrected pearson’s $r$
		Mean speed	Variability in speed	Variability in lane position	Variability in steering
Full sample	150	0.99	0.96	0.93	0.99
Non-neurological	59	0.99	0.92	0.91	0.97
Neurological	91	0.99	0.97	0.93	0.99

Note. The neurological group included participants with history of traumatic brain injury, brain tumor, stroke, or multiple sclerosis.

These behaviors were recorded by the VRDS 16 times per second. These data were then summarized for each quarter mile section of the driving route to create the following variables: mean speed, mean variability in speed (standard deviation), mean variability in lane position (standard deviation of distance from center lane position), and mean variability in steering (standard deviation of steering wheel position).

Internal consistency and split half reliability were calculated from data for all participants on the basic rural drive, using averaged data from quarter mile segments. In other words, driving behavior (speed, variability in speed, variability in lane position, and variability in steering) during each quarter mile section was treated as a separate data point numbered in order of completion (e.g., quarter mile 1, quarter mile 2, etc.). Internal consistency was measured by calculating Cronbach’s alpha. Split-half reliability was attained by placing every other quarter mile segment into one of two group. Because split-half reliability was based on half the trials, a Spearman-Brown correction was applied.

Test-retest reliability was calculated using Pearson’s $r$ for data from the 22 participants who completed the driving route during a second session. Data for these analyses included all quarter mile segments from the basic rural drive, as well as sections from the highway and residential/commercial zones that were administered in exactly the same way (i.e. without additional challenges or task demands) during both sessions. Data from sections of the route that differed between session 1 and session 2 were omitted. With respect to strength of reliability, coefficients $\geqslant$ 0.90 were considered excellent, 0.80–0.89 was considered good, and 0.70–0.79 was considered acceptable.

Table 4

Test-retest reliability for whole rural route baseline drive in non-neurological sample ( $N=$ 22)

Driving variable	Mean 1 (SD)	Mean 2 (SD)	$r$
Mean speed	40.23 (2.40)	41.39 (2.36)	0.28
Variability speed	1.47 (0.55)	1.42 (0.53)	0.65 ${}^{*}$
Variability lane	1.15 (0.30)	1.28 (0.38)	0.83 ${}^{*}$
Variability steer	1.04 (0.38)	1.00 (0.22)	0.74 ${}^{*}$

Note. ${}^{*}p<$ 0.001.

Table 5

Test-retest reliability for select portions of highway, residential, and commercial drives in non-neurological sample ( $N=$ 22)

Driving variable	Mean 1 (SD)	Mean 2 (SD)	$r$
Mean speed	43.86 (15.00)	45.80 (15.72)	0.96 ${}^{*}$
Variability speed	3.80 (3.88)	3.45 (4.49)	0.93 ${}^{*}$
Variability lane	2.65 (1.79)	2.27 (1.33)	0.57 ${}^{*}$
Variability steer	7.00 (7.17)	6.80 (6.77)	0.98 ${}^{*}$

Note. ${}^{*}p<$ 0.001.

3. Results

Internal consistency, split-half reliability, and test-retest reliability for the full sample and each neurological group are displayed in Tables 2 through 4. Internal consistency and split-half reliability of driving variables during the basic rural route were excellent with coefficients between 0.8 and 0.9. All driving variables assessed, including mean speed, variability in speed, variability in lane positioning, and variability in steering, had excellent consistency and reliability. Findings held across neurological conditions, as well as with patients with no history of neurological impairment.

Test-retest reliability was more variable. For the baseline drive on the basic rural route, reliability of variability in lane positioning and steering were adequate to good. Although test-retest reliability of variability in speed was statistically significant, it was less than adequate. Reliability of mean speed was not statistically significant. On select portions of the highway, residential, and commercial drives (i.e., portions that did not vary from session 1 to session 2), test-retest reliability was excellent for mean speed, variability in speed, and variability in steering. Test-retest reliability for variability in lane positioning was statistically significant but less than adequate.

4. Discussion

The current study sought to evaluate the reliability of the Applied-Neurotechnologies Virtual Reality driving simulator in assessing driving performance in individuals diagnosed with multiple sclerosis or acquired brain injuries, and individuals with no history of neurological impairment. Findings suggest that VRDS is a reliable tool for measuring driving behaviors. Internal consistency and split-half reliability were strong across a variety of driving variables and within both healthy and neurologically-impaired drivers. This lends additional confidence to findings from studies that have used this instrument [12, 29]. In addition, it suggests that this VRDS may be used as a reliable clinical and research tool for measuring driving behaviors consistently across various sections of the virtual route.

Despite good internal reliability, test-retest reliability in a sub-sample of 22 healthy participants was variable. Findings are likely due, at least in part, to participants habituating to specific sections of the routes over time. This was particularly true during the basic rural route, which is a much lengthier and basic, less complex driving segment than the highway, commercial, and residential routes. Interestingly, driving performance did not necessarily improve over time. For instance, mean speed increased and became more discrepant from the speed limit from session 1 to session 2. This may be due to participants adjusting to observer effects and therefore becoming less concerned about adhering to the speed limit. It may also suggest that mean speed is a less stable measure in driving simulation. Indeed, previous research has found mean speed to be faster when individuals are driving an instrumented car compared with driving a simulator [31]. Regarding lane positioning, changes between the first and second session varied, as participants demonstrated increased variability during the second session on the easier, basic rural route, and decreased variability during the second session on the more difficult highway and residential/commercial routes. This pattern suggests that driver engagement may have adjusted to the level of demand of the driving environment, leading to better performance during more engaging and difficult driving tasks. Taken together, test-retest results suggest that drivers’ behavior in the VRDS changes over time, which may be due to habituation, level of demand, or other factors. Inadequate test-retest reliability highlights the importance of using comparison groups to account for changes in driving behavior on the VRDS over time. Given that there is no clear standard of what constitutes “good” driving, samples of interest (e.g., individuals with acquired brain injuries) are typically compared to neurologically healthy or typically developing individuals. This study suggests that although driving across similar segments within a single session is quite reliable, session to session driving varies substantially even in neurologically healthy individuals. Although a variety of factors are likely contributing to these differences, practice effects are almost certainly impacting performance. To reduce practice effects bias, it is recommended that samples with neurological conditions continue to be compared to a control group who have completed the same number of driving sessions under similar conditions and driving environments.

Research has consistently pointed to the importance of utilizing ecologically valid tools to capture deficits in activities of daily living experienced by individuals with neurological disorders [32, 33, 34]. Research utilizing virtual reality for assessment of driving behaviors in individuals with neurological and neurocognitive impairments is growing. Virtual reality has also been used as a clinical tool in assessing fitness to drive and in training new or rehabilitating drivers [35, 36, 37, 38]. Establishing reliability and validity of the current study’s VRDS, which has previously been utilized in multiple studies improves confidence in interpretation of research and clinical findings stemming from this instrument. This study takes the first step by establishing internal consistency of rural segments of the VRDS route across several driving behaviors in a neurologically diverse sample.

Interpretation of findings is limited by a small sample size. In addition, reliability could not be tested for all driving environments and routes implemented by this VRDS because they were too varied to reasonably allow for consistent driving behaviors. Future research should aim to assess test-retest reliability in a larger and more diverse sample. Practice effects may be reduced by testing smaller sections of the route over repeated time frames and by allowing more time for habituation during the first session. Furthermore, research should investigate the reliability and utility of VRDS for measuring more complex driving behaviors, such as completion of turns, stops, and lane changes to further inform driver retraining and/or tailored interventions geared at complex activities of daily living, such as driving. Research is also needed to assess the validity of VRDS in predicting on-the-road performance. Specifically, though the VRDS is an important step towards measuring driving performance in an ecologically valid manner (i.e., utilizes realistic scenarios that involve complex behaviors and resemble daily driving), it still has limitations as it is conducted within a controlled environment that does not exactly mirror daily driving. In fact, incorporation of VRDS elements in an on-road performance test remains an important area of future research. Toward this end, this study provides the first step in establishing reliability of VRDS for clinical and research utility, which paves the way for future studies involving the VRDS and its potential predictive value over real-life driving-related behaviors and functional outcomes that may influence individuals with neurological disorders as well as healthy controls.

Footnotes

Conflict of interest

The authors do not have any conflicts of interest to report.

References

Brooks

McNeil

Rose

Greenwood

Attree

Leadbetter

. Route learning in a case of amnesia: A preliminary investigation into the efficacy of training in a virtual environment. Neuropsychological Rehabilitation. 1999; 9(1): 63-76.

Cromby

Newman

Tasker

. Successful transfer to the real world of skills practiced in a virtual environment by student with severe learning disabilities. In: Sharkey

, editor. Proceedings of the 1st European Conference on Disability, Virtual Reality and Associated Technologies Reading, UK: University of Reading, 1996.

Grewe

Lahr

Kohsik

Dyck

Markowitsch

Bien

, et al. Real-life memory and spatial navigation in patients with focal epilepsy: ecological validity of a virtual reality supermarket task. Epilepsy Behav. 2014; 31: 57-66.

Jovanovski

Zakzanis

Ruttan

Campbell

Erb

Nussbaum

. Ecologically valid assessment of executive dysfunction using a novel virtual reality task in patients with acquired brain injury. Appl Neuropsychol Adult. 2012; 19(3): 207-20.

Paglia

La Cascia

Rizzo

Riva

La Barbera

. Assessment of executive functions in patients with obsessive compulsive disorder by NeuroVR. Stud Health Technol Inform. 2012; 181: 98-102.

McGeorge

Phillips

Crawford

Garden

Della Sala

Milne

Callander

. Using virtual environments in the assessment of executive dysfunction. Presence: Teleoperators and Virtual Environments. 2001; 10(4): 375-83.

Mendozzi

Motta,

Barbieri,

Alpini,

Pugnetti

. The application of virtual reality to document coping deficits after a stroke: report of a case. CyberPsychology and Behavior. 1998; 1(1): 79-91.

Shallice

Burgess

. Higher order cognitive impairments and frontal lobe lesions in man. In: HSE

Levin

Benton

, editor. Frontal Lobe Function and Dysfunction Oxford University Press; 1991; pp. 125-38.

Tarnanas

Schlee

Tsolaki

Muri

Mosimann

Nef

. Ecological validity of virtual reality daily living activities screening for early dementia: Longitudinal study. JMIR Serious Games. 2013; 1(1): e1.

10.

Cox

Davis

Singh

Barbour

Nidiffer

Trudel

, et al. Driving rehabilitation for military personnel recovering from traumatic brain injury using virtual reality driving simulation: a feasibility study. Mil Med. 2010; 175(6): 411-6.

11.

Guinosso

Johnson

Schultheis

Graefe

Bishai

. Neurocognitive correlates of young drivers’ performance in a driving simulator. J Adolesc Health. 2016; 58(4): 467-73.

12.

Patrick

Hurewitz

McCurdy

Agate

Daly

Tarazi

, et al. Driving comparisons between young adults with autism spectrum disorder and typical development. J Dev Behav Pediatr. 2018; 39(6): 451-60.

13.

Schultheis

Simone

Roseman

Nead

Rebimbas

Mourant

. Stopping behavior in a VR driving simulator: A new clinical measure for the assessment of driving. Conf Proc IEEE Eng Med Biol Soc. 2006; 1: 4921-4.

14.

Schultheis

Rebimbas

Mourant

Millis

. Examining the usability of a virtual reality driving simulator. Assist Technol. 2007; 19(1): 1-8; quiz 9-10.

15.

Schultheis

Weisser

Ang

Elovic

Nead

Sestito

, et al. Examining the relationship between cognition and driving performance in multiple sclerosis. Arch Phys Med Rehabil. 2010; 91(3): 465-73.

16.

Schultheis

Weisser,

Manning,

Blasco,

Ang,

. Driving behaviors among community-dwelling persons with multiple sclerosis. Arch Phys Med Rehabil. 2009; 90(6): 975-81.

17.

Gianutsos

Campbell

Beattie

Mandriota

. The driving advisement system: a computer-augmented quasi-simulation of the cognitive prerequisites for resumption of driving after brain injury. Assist Technol. 1992; 4(2): 70-86.

18.

Lundqvist

. Neuropsychological aspects of driving characteristics. Brain Inj. 2001; 15(11): 981-94.

19.

Szlyk

Myers

Zhang

Wetzel

Shapiro

. Development and assessment of a neuropsychological battery to aid in predicting driving performance. J Rehabil Res Dev. 2002; 39(4): 483-96.

20.

Zesiewicz

Cimino

Malek

Gardner

Leaverton

Dunne

, et al. Driving safety in Parkinson’s disease. Neurology. 2002; 59(11): 1787-8.

21.

Netz

Wolbers,

. Hemianopsia and Driving. Europa Medicophysica. 2001; 37(4): 275-8.

22.

Szlyk

Taglia

Paliga

Edward

Wilensky

. Driving performance in patients with mild to moderate glaucomatous clinical vision changes. J Rehabil Res Dev. 2002; 39(4): 467-82.

23.

Blaauw

. Driving experience and task demands in simulator and instrumented car: A validation study. Human Factors. 1982; 24(4): 474-86.

24.

Roge

Pebayle

El Hannachi

Muzet

. Effect of sleep deprivation and driving duration on the useful visual field in younger and older subjects during simulator driving. Vision Res. 2003; 43(13): 1465-72.

25.

Brouwer

Waterink

, Wolffelaar

Rothengatter

. Divided attention in experienced young and older drivers: lane tracking and visual analysis in a dynamic driving simulator. Hum Factors. 1991; 33(5): 573-82.

26.

Lee

Drake,

Cameron,

. Identification of appropriate assessment criteria to measure older adults’ driving performance in simulated driving. Australian Occupational Therapy Journal. 2002; 49(3): 138-45.

27.

Hoffman

McDowd

. Simulator driving performance predicts accident reports five years later. Psychol Aging. 2010; 25(3): 741-5.

28.

Neyens

Boyle

Schultheis

. The effects of driver distraction for individuals with traumatic brain injuries. Hum Factors. 2015; 57(8): 1472-88.

29.

Vickers

Schultheis

Manning

. Driving after brain injury: Does dual-task modality matter? NeuroRehabilitation. 2018; 42(2): 213-22.

30.

Raphail

Vickers

Leist

Schultheis

. Clinical Utility of the Multiple Sclerosis Functional Composite over the Expanded Disability Status Scale for Screening Driving Capacity in Multiple Sclerosis. American Journal of Physical Medicine and Rehabilitation, 2019; in press.

31.

Godley

Triggs

Fildes

. Driving simulator validation for speed research. Accid Anal Prev. 2002; 34(5): 589-600.

32.

Classen

Shechtman

Stephens

Davis

Justiss

Bendixen

, et al. The impact of roadway intersection design on driving performance of young and senior adults. Traffic Inj Prev. 2007; 8(1): s69-77.

33.

Kraft

Amick

Barth

French

Lew

. A review of driving simulator parameters relevant to the operation enduring freedom/operation iraqi freedom veteran population. Am J Phys Med Rehabil. 2010; 89(4): 336-44.

34.

Wolf

Chuh

Floyd

McInnis

Williams

. Effectiveness of occupation-based interventions to improve areas of occupation and social participation after stroke: an evidence-based review. Am J Occup Ther. 2015; 69(1): 6901180060p1-11.

35.

Kubler

Kasneci

Rosenstiel

Heister

Aehling

Nagel

, et al. Driving with glaucoma: Task performance and gaze movements. Optom Vis Sci. 2015; 92(11): 1037-46.

36.

Spreij

Ten Brink

Visser-Meily

JMA

Nijboer

TCW

. Simulated driving: The added value of dynamic testing in the assessment of visuo-spatial neglect after stroke. J Neuropsychol. 2018.

37.

Cox

Brown

Ross

Moncrief

Schmitt

Gaffney

, et al. Can youth with autism spectrum disorder use virtual reality driving simulation training to evaluate and improve driving performance? An Exploratory Study. J Autism Dev Disord. 2017; 47(8): 2544-55.

38.

White

Moussavi

. Neurocognitive treatment for a patient with alzheimer’s disease using a virtual reality navigational environment. J Exp Neurosci. 2016; 10: 129-35.

Reliability of a virtual reality driving simulator for individuals with neurological disorders

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1. Introduction

2. Methods

2.1 Participants

Table 1 Participant demographics

Table 2 Internal consistency for baseline rural drive

4. Discussion

Footnotes

Conflict of interest

References

Table 1
Participant demographics

Table 2
Internal consistency for baseline rural drive