Using Virtual Reality Techniques to Study Cognitive Processes in Car

Driving Activity

Emmanuelle M

etrier

, Paul Richard

, Vincent Boucher

and Christophe Boujon

Laboratoire de Psychologie des Pays de la Loire (LPPL), Universit

e d’Angers, Angers, France

Laboratoire de Recherche en Ing

enerie des Syst

emes (LARIS), Universit

e d’Angers, Angers, France

Centre d’

Etudes et d’Expertise sur les Risques, l’Environnement, la Mobilit

e et l’Am

enagement, Les Ponts-de-C

e, France

Keywords:

Car Driving, Eye-movements, Attention, Boundary Extension, Virtual Reality.

Abstract:

A central question in cognitive sciences is how behaviors adapted to the situations encountered are produced.

This question can be addressed in different ways and often requires the researcher to choose between highly

controlled and standardized laboratory situations (commonly referred to as artiﬁcial settings) and studies un-

dertaken in natural settings which may be more realistic, but cannot be controlled as required by a rigorous

scientiﬁc approach. Using car driving as an example, our study will show how virtual reality (VR) offers a

compromise between these two alternatives. Indeed, VR can simulate controlled immersive environments that

offer different levels of realism. Moreover, VR makes it possible to implement different devices. For instance,

VR enables researchers to analyze oculomotor behavior, which is fundamental in the ﬁeld of car driving and is

considered an indicator of attentional deployment. The work presented in this paper is based on a car driving

simulator currently under development and aimed at studying the cognitive processes involved in car driving

such as attentional processes and anticipatory mechanisms.

1 INTRODUCTION

A central question in cognitive sciences is how be-

haviors adapted to the situations encountered are pro-

duced. This question is not as trivial as it seems

because producing relevant behavioral reactions re-

quires the mobilization of complex cognitive pro-

cesses. Car driving thus lies at the heart of numer-

ous researchers concerns and is perceived, in partic-

ular, as an  integrated multi-task behavior engaging

several processes and requiring different interrelated

skills that rely on interconnected visual, motor, and

cognitive brain systems  (Graydon et al., 2004). One

of the difﬁculties commonly encountered is the need

to choose between highly controlled laboratory situa-

tions and studies undertaken in natural settings. While

the former confers a more or less artiﬁcial character

on the situations, the latter, which is more realistic,

does not enable researchers to control all the parame-

ters of the situations that are expected to be standard-

ized.

The situations created in laboratories are gener-

ally deﬁned by their artiﬁcial nature. Indeed, the sit-

uations presented to participants sometimes greatly

differ from the situations encountered in natural set-

tings. To address this problem, some researchers have

focused on presenting driving situations that are as

ecological as possible; i.e., situations presenting nat-

ural scenes depicting either static driving scenes (i.e.,

photographs, e.g., (Galpin et al., 2009) or dynamic

video clips of driving (Crundall et al., 2003)). In ad-

dition to presenting stimuli closer to natural driving

conditions, this choice is largely driven by the need

to present strictly identical situations for all partic-

ipants. Indeed, without this condition, participants

performances cannot be compared and the validity of

the research studies is likely to be prejudiced.

As engaging as it might seem, this mode of work-

ing has its limitations because the researcher depends

on the characteristics of the environment and events

occurring during the capture of movies or still images.

A second limitation is that the subject cannot act on

the environment (even if the researcher is careful to

present situations as close as possible to natural driv-

ing by asking, for instance, the subject to follow the

road by turning a steering wheel), while car driving

mobilizes both perceptual and action-related knowl-

edge (Bl

attler et al., 2012). This highlights the neces-

sity for devices that allow subjects to interact with the

environment in order to be closer to natural driving

138

MÃl’nÃl’trier E., Richard P., Boucher V. and Boujon C.

Using Virtual Reality Techniques to Study Cognitive Processes in Car Dr iving Activity.

DOI: 10.5220/0006230901380145

In Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2017), pages 138-145

ISBN: 978-989-758-229-5

conditions. An alternative approach to highly con-

trolled experimental situations involves undertaking

studies in natural settings. This approach has been

used by Underwood et al. (2003), who observed the

behavior of car drivers in natural driving situations.

While studies in natural settings present high eco-

logical validity, their greatest shortcoming is that they

do not allow researchers to control the environment in

which the subject is driving (i.e., the different events

that occur despite participants being presented with a

similar path, or the impossibility of predicting trafﬁc

conditions, as in the case of Underwood et al. (2003)),

a necessary condition for acceptable validity. As a re-

sult, the situations differ from participant to partici-

pant and are more difﬁcult to compare.

The development of virtual environments (VEs)

has made possible a compromise between natural and

experimentally controlled situations. Indeed, Virtual

Reality (VR) simulations allow researchers to control

the complexity of the presented environment (e.g., in

terms of perceptive richness, with, for instance, the

addition or withdrawal of visual details such as the

presence of trafﬁc signs) or occurring events (e.g.,

presence or absence of distractors such as vehicles or

pedestrians, or events such as car accidents) (Rizzo,

2002). Moreover, VR makes it possible to implement

different devices (Bian et al., 2013; Bian et al., 2015;

Zhang et al., 2015; Lei et al., 2016). For instance,

VR enables researchers to analyze oculomotor behav-

ior which is fundamental in the ﬁeld of car driving

and is considered an indicator of attentional deploy-

ment. Finaly, the relevance of such an approach de-

pends on the presentation of situations that are highly

controlled and strictly identical across participants, as

well as the fact that the approach accords the encoun-

tered situations a more ecological nature by offering,

in particular, the possibility of interacting in real time

with the environment in which the subject may be im-

mersed.

The objective of this paper is to present a car driv-

ing simulator currently under development and aimed

at studying the cognitive processes involved in car

driving such as attentional processes and anticipatory

mechanisms. In the next Section, we describe some

existing driving simulators that have been developed

so far and present some related works about visual

attention and eye-movements in driving, load theory

of attention, cognitive control and anticipatory mech-

anisms. Section 3 presents the VEs we have devel-

oped and our driving simulations. In Section 4, we

focus on the measure and analysis of collected eye-

movements. In Section 5, we present our approach

to investigate cognitive processes such as anticipatory

mechanisms in driving simulation. Finally, we will

present our conclusions and suggest avenues for fur-

ther work.

2 RELATED WORK

2.1 Driving Simulators

Nowadays, two types of driving simulators are avail-

able : (1) static simulators, with which the user ex-

periences no acceleration or forces and (2) dynamic

simulators, which are based on motion platforms.

The most advanced static simulators incorporate a

fully instrumented vehicle placed in front of a semi-

cylindrical or multi-screen display. Among these sys-

tems, we can cite the SIM2 simulator from the French

Institute for Sciences and Technologies of Transport,

Development and Networks (IFSTTAR). This simu-

lator incorporates a Citro

en Xantia placed in front of

a three-sided screen (Auberlet et al., 2010). There is

also the Fraunhofer IAO simulator, consisting of a Re-

nault Scenic placed in front of a three-sided screen,

with a 180-degree ﬁeld of view. Note that this simula-

tor offers two screens behind the vehicle (Marberger,

2008). The simulator developed at the University of

Leeds consists of a Rover 216 and a curved screen

with a 120-degree ﬁeld of view (Blana and Golias,

2002).

Most dynamic simulators integrate acceleration

feedback via the use of a motion platform. We can

cite for example the SHERPA simulator from PSA

Peugeot-Citro

en, which was duplicated at the LAMIH

Laboratory at the University of Valenciennes. This

simulator, consisting of a Peugeot 206 mounted on

a Rexroth Hydraudyne platform, uses a three-sided

screen with a 180-degree ﬁeld of view (Younsi et al.,

2009). In the same category, the VERA simulator of

the Technology Environment Safety Transport Lab-

oratory of the University of Naples consists of one

half of a C2 Citro

en, mounted on a six-stage plat-

form of freedom. This vehicle is placed in front of

a screen made up of three faces and has also a 180-

degree ﬁeld of view. LCD screens in the central

and exterior mirrors provide rearward vision (Torri-

eri et al., 2008). The CARRS-Q (Center for Accident

and Road Safety) Advanced Simulator at the Univer-

sity of Queensland consists of a vehicle placed on a

Rexroth platform. The simulator is equipped with

a three-sided screen with a 18-degree ﬁeld of view.

LCD screens are placed in both the center and exterior

mirrors to provide rearward vision (Haines, 2011).

Dynamic simulators also include the KMUDS-4 from

Kookmin University in Korea. This simulator con-

sists of a vehicle, a three-sided frontal screen with a

Using Virtual Reality Techniques to Study Cognitive Processes in Car Driving Activity

139

140-degree ﬁeld of view. The system also has a rear

screen with a 50-degree ﬁeld of view. The restitution

of movements is carried out by four motors located

under the vehicle (Lee et al., 2007). The Ford VIRT-

TEX simulator uses a vehicle in a dome containing

a cylindrical screen with a 180-degree ﬁeld of view

and a rear cylindrical screen with a 120-degree ﬁeld

of view (Artz et al., 2001). The system is mounted on

a platform with six degrees of freedom. In order to re-

produce large-amplitude displacements in one or two

dimensions, the dynamic simulators can be placed on

rails. One-dimensional systems include the MARS

simulator of the IFAS (Institut f

ur Fahrzeugtechnik

und Antriebssystemtechnik, Germany). The rail used

allows a lateral displacement of 1.5 m (Breidenbach

and Tomaske, 2004). There is also the simulator III

of the National Institute for Road and Transport Re-

search in Sweden. The rail used allows a displace-

ment of 7.5 m. Moreover, the cab can be oriented by

90 degrees, making it possible to reproduce longitu-

dinal and lateral movements (Nordmark et al., 2004).

The Daimler simulator consists of a dome contain-

ing a vehicle and a 360 degrees cylindrical screen,

mounted on a platform with six degrees of freedom.

The system uses a rail with an amplitude of move-

ment of 12 m (Zeeb, 2010). The Cards 2 simulator

developed by Renault, also falls into this category. It

is composed of a cabin equipped with a dashboard of

a M

egane, integrating a manual gearbox with ﬁve ra-

tios. Three screens with a horizontal viewing angle of

150 degrees and a vertical 40-degree ﬁeld of view are

used. Two LCD screens are placed in the exterior mir-

rors, and the image of the central mirror is projected

onto the top of the center screen. The simulator is

mounted on a platform with six degrees of freedom,

allowing a displacement of ± 20 cm and a rotation

of ± 20 degrees (Reymond and Kemeny, 2000). Re-

nault has also developed the Ultimate Dynamic Sim-

ulator which consists of a cabin equipped with a La-

guna dashboard. The steering wheel is equipped with

an active feedback system, while the pedals and shift

lever are equipped with a passive feedback system.

The images are projected onto a cylindrical screen

having a horizontal 150-degree ﬁeld of view and a

vertical 40-degree ﬁeld of view. The NADS-1 sim-

ulator from the University of Iowa has 19.5 m lon-

gitudinal and lateral displacements (Greenberg et al.,

2006). Note that the cabin incorporates a 360 degrees

cylindrical screen.

Some dynamic simulators are based on a speciﬁc

architecture for the simulation of movements and ac-

celerations. For example, the CyberMotion simula-

tor from the Max Planck Institute, uses a robot with

six degrees of freedom, on which the driver’s position

and screen are positioned (Grabe et al., 2010). In the

same category, there is the Desdemona simulator of

the TNO (Mayrhofer et al., 2009).

2.2 Visual Attention and

Eye-movements

Vision is the major sense used during the driving

activity (Sivak, 1996). The Human Visual System

(HVS) is able to resolve ﬁne details in a scene, as

well as in a moving sequence, at close and far dis-

tances (Snowden et al., 2006). Over the last two

decades, HVS which lies at the heart of transportation

studies has focused on analyzing visual attention and

gaze tracking (Jacob, 1991; Lemonnier et al., 2014;

Lemonnier et al., 2015). In this context, the trans-

port ﬁeld has beneﬁted from recent progress in eye-

tracker devices and from the emergence of computa-

tional models of visual attention and a new research

area, i.e., VR.

Although eye tracking technology is susceptible

to data quality issues (Ahlstrom et al., 2012), eye

movements have been widely studied in the context of

driver behavior, attention management, and to assess

task related visual demand (Wierwille., 1993; Mori

et al., 2013; Sodhi et al., 2002; Aoki et al., 2010).

While many eye movement measures are available

today (Holmqvist et al., 2011), some are easier to

calculate than others. The most common eye move-

ment events are ﬁxation and saccade. Fixation corre-

sponds to the period of time the eye remains nearly

still and a saccade is the rapid motion of the eye from

one ﬁxation to another. Analyzing ﬁxation events is

especially interesting in driving studies because it is

widely accepted that individuals are blind during a

saccade. To identify the information that draws the

drivers attention outside the vehicle (e.g. trafﬁc signs,

signals, pavement markings), the prime measures are

those that determine whether the driver is looking at

a particular location or signal referred to as Area of

Interest (AOI). Several characteristics of eye move-

ments such as eye blinks, pupil dilation and saccadic

peak velocity reﬂect variations in attention, emotion

and mental workload over time.

2.3 Load Theory of Attention and

Cognitive Control

The ability to focus attention on the elements of en-

vironment to be treated in priority and ignore dis-

tractions is essential to the production of adapted be-

haviors. However, this blindness to distractors might

be damaging in some situations, as in the case of

car driving where it is vital to detect the presence

HUCAPP 2017 - International Conference on Human Computer Interaction Theory and Applications

140

of other vehicles and trafﬁc signs (Lavie, 2010, for

a review). The ability to detect distractors during

the execution of a task requiring attention can be ex-

plained by Lavies load theory of attention and cog-

nitive control (Lavie et al., 2004; Lavie and Fock-

ert, 2005; Lavie, 2010). This theory tends to resolve

the debate relative to the locus of attentional selec-

tion (early, late) by introducing the idea that focused

attention in the face of distractions depends both on

the amount and type of information load involved by

the task. The load theory posits that focused atten-

tion is enhanced under conditions of high perceptual

load and degraded in situations involving high load on

cognitive control processes (e.g., working memory).

Indeed, tasks involving high perceptual load consume

the overall processing capacity, leaving no more re-

sources available for the perception of irrelevant dis-

tractors. Such a prediction is in agreement with the

view of early attentional selection, according to which

the individuals perceive only the objects to which they

pay attention because of limited perceptual process-

ing capacity. Conversely, in the case of tasks involv-

ing a low perceptual load, the processing capacity is

not totally used, allowing the processing of distractive

information. This second prediction is in agreement

with the theory of late selection that postulates that

perception refers to an unlimited capacity automatic

process. In this case, the efﬁciency of attentional se-

lection, which refers to “the extent to which distrac-

tors that have been perceived can be prevented from

gaining control over behavior” (Lavie, 2010, p. 143),

depends on the load on cognitive control processes,

a high working memory load being for instance the

origin of increased interference of distractors.

2.4 Anticipatory Mechanisms

Anticipatory mechanisms involved in the perception

of the surrounding world is at the heart of the produc-

tion of adaptive behavior (Bl

attler et al., 2012), for

instance in the case of car driving. Thus, the abil-

ity to realize predictions concerns both the likely fu-

ture trajectory of a moving object or the likely evo-

lution of the dynamic environment in which the ob-

server is embedded, a phenomenon known as repre-

sentational momentum (Freyd and Finke, 1984) and

the prediction of areas of the environment just out-

side the observers visual ﬁeld. This phenomenon is

known as boundary extension (Intraub and Richard-

son, 1989) and is described as the tendency to overes-

timate the spatial expanse of a previously perceived

scene, the visual system inferring information that

could be present just beyond the boundaries of a view.

As a result, the observers memory includes extended

boundaries in comparison to the boundaries of the

original scene. While this phenomenon is generally

described as a memory error, it presents an important

adaptive value by facilitating the interactions with the

environment through, for instance, tasks such as nav-

igation (Hale et al., 2016).

Several studies have shown that car driving ex-

pertise modulates boundary extension (M

etrier and

Didierjean, 2013), with experts processing the scenes

belonging to their expertise ﬁeld more effectively than

novices (Reingold et al., 2001; Mourant and Rock-

well, 1972). However, one of the limitations of these

studies is that they relied on the presentation of road-

scene photographs passively grasped by observers,

whereas car driving implies perceptivo-motor repre-

sentations and abilities (Bl

attler et al., 2012).

3 VIRTUAL ENVIRONMENTS

AND DRIVING SIMULATION

3.1 Virtual Environments

Two complementary VEs have been developed for the

study of cognitive processes (attentional and predic-

tion) and the perceptual mechanisms involved when

one is driving a vehicle. The ﬁrst is an urban en-

vironment (Fig. 1) composed of buildings arranged

along linear axes and intersections with trafﬁc signs

and trafﬁc lights. Restaurants and caf

es were mod-

eled and placed in the streets. Each intersection can be

set according to each experiments needs. The trafﬁc

lights therefore behave realistically and are fully con-

trolled using ﬁnite state machines. As shown in Fig-

ure 1, below, a pavement marker is also present (white

and yellow lines on the ground). Vehicles and charac-

ters with realistic autonomous behaviors can also be

placed in the environment. Artiﬁcial lights for illumi-

nating the streets and the buildings are also present.

To increase the realism of the simulation, vegetation

was integrated (bushes and trees). These elements can

be animated in real time according to weather condi-

tions (air stream).

Lastly, the VE offers a more or less rich sound-

scape. Depending on the trafﬁc, different audio clips

can be played. By default, the noise generated by the

car engine and the braking is present in the simulation

even though, naturally, it can be cut off. It is worth

noting that sound information increases the realism

of the simulation but is also perceived as an impor-

tant driving aid. The Figure 2, provides a view of the

countryside VE during the driving simulation. This

VE comprises open ﬁelds and has minimal vegetation.

During the simulation, however, the driver may pass

through small villages comprising farms and some

Using Virtual Reality Techniques to Study Cognitive Processes in Car Driving Activity

141

Figure 1: A user’s view of the virtual city during the driving

simulation.

Figure 2: A user’s view of the countryside VE during the

driving simulation.

houses. The driver might also pass through a forest

with many trees.

3.2 Driving Simulation

Two driving approaches are proposed: (1) free driv-

ing in which the driver operates the vehicle and has

full dynamic control of simulation parameters (steer-

ing, speed, etc.) and (2) constrained driving in which

the driver has only partial control of the vehicles be-

havior. In the latter case, the vehicle advances on rails

(Fig. 3) and its speed is kept constant. The driver con-

trols when the vehicle starts and when it stops. This

second approach was implemented in order to ensure

that all subjects had the same visual stimuli. With re-

gard to our experiment, we chose to represent neither

the inside (virtual camera positioned in front of the

steering wheel) nor the hood of the vehicle, especially

when the user had two wheels in front of him/her (a

physical one and the virtual one) during the driving

simulation.

Our experimental platform is based on a simple

conﬁguration (Fig. 4) and provides haptic feedback

using the Logitech G27 steering wheel. This de-

vice has the following characteristics: (1) a power-

ful dual-motor force feedback mechanism with heli-

cal gearing, (2) a lever for six-speed engagement of

Figure 3: Example of rails used to constraint the car path.

reverse gear with pressure, (3) the ability to change

speed/tachometer (4) a leather coated wheel with a di-

ameter measuring 28 cm and (5) an accelerator pedal,

steel brake and clutch.

Figure 4: Experimental platform for driving simulation and

eye-tracking.

4 MEASURE AND ANALYSIS OF

EYE-MOVEMENTS

Twenty four (24) participants took part in a driving

simulation experiment based on the driving simula-

tor previously described. The VEs were projected

on a screen 55 cm high and 107 cm wide. Partici-

pants were positionned approximately 170 cm from

the screen, which was placed at a horizontal angle of

view of 35

◦

and a vertical angle of view of 18

◦

. A

RED eye-tracking system from SMI, running at 60

Hz, was used (see Figure 4). The experiment lasted

approximately 30 minutes (ten minutes of free train-

ing, ten minutes in the rural environment and ten min-

utes in the urban environment). Eye movements were

ﬁrst analyzed using the gaze standard deviation. We

HUCAPP 2017 - International Conference on Human Computer Interaction Theory and Applications

142

posited that the standard deviation would reﬂect men-

tal workload and would depend on the environment

(rural/urban) (Burnham et al., 2014). The horizontal

standard deviation for the rural environment was 3.4

◦

and the vertical deviation was 3.5

◦

. These weak val-

ues can be explained through the monotonous envi-

ronment along the route and the drivers concentration

on his/her trajectory alone, i.e., concentration on the

vanishing point. Sample of eye-movement paths and

hot spots during the exploration of the virtual city are

given in Figures 5 and 6 respectively.

Figure 5: Eye-movement paths during the exploration of the

virtual city.

For the urban environment, the horizontal stan-

dard deviation was 7.0

◦

and the vertical deviation

was 3.0

◦

. The vertical/horizontal difference can be

explained through the perspective projection of the

world; while the vertical dimension represents the

perspective, the horizontal dimension is related to a

ﬁxed distance. Gaze exploration (looking at cars, traf-

ﬁc lights, etc.) leads to greater eye movement in a hor-

izontal direction. Scene complexity also increased the

standard deviation. After these ﬁndings, we turned

our interest to the utility/distracter ratio (on the road)

and the AOI approach, which requires us to detect the

road in every frame and compare this with the eye

position. To this end, a robust image processing al-

gorithm should be implemented and applied to each

frame of the acquisition movies. These studies are

currently under development will be presented in a

future paper.

5 ANTICIPATORY MECANISMS

IN DRIVING SIMULATION

In addition to the creation of situations closer to those

encountered in natural settings, VR enables to imple-

ment experimental tasks aimed at studying speciﬁc

Figure 6: Hot spots during the exploration of the virtual city.

cognitive processes. One of the goals we pursued here

is the study of the anticipatory mechanisms involved

in the perception of the surrounding world. In this

context, the use of VR will enable us to study this

question by proposing a driving task to novice and

expert car drivers, during which they will move in the

virtual city either as drivers (i.e., involved in the driv-

ing action condition) or as passengers (i.e., less in-

volved condition).

To study anticipatory mechanisms in driving, a

boundary extension task will be incorporated at dif-

ferent times: during driving, the scene will be frozen

for a very brief duration (250 ms) before being re-

placed by a black screen (1 s). Immediately after-

wards, the frozen scene will reappear, but the distance

at which it is presented will have been modiﬁed. The

task of the subject will be to indicate whether the sec-

ond scene is closer-up, further away, or at the same

distance as the original. In order to present compa-

rable situations from one participant to another (i.e.,

strictly identical scenes), we will use the rail that has

been implemented in order to standardize subjects

paths and viewpoints. This task will enable us to bet-

ter understand the simultaneous effects of expertise

and involvement in the action on boundary extension

and will also enable the immersion of the participants.

This appears to be a fundamental question because, in

real life, human beings are embedded in environments

in which they act toward goals.

6 CONCLUSION AND FUTURE

WORK

We have presented a car driving simulator aimed

at studying the cognitive processes involved in car

driving such as attentional processes and anticipatory

Using Virtual Reality Techniques to Study Cognitive Processes in Car Driving Activity

143

mechanisms. To this aim, speciﬁc experimental tasks

have been developped. The proposed experimental

platform is equipped with an eye-tracking system to

collect speciﬁc eye movement data. Despite the gap

between the real and the virtual world, VR is an inter-

esting compromise between highly controlled experi-

mental situations and studies undertaken in a natural

setting. Indeed, VR makes it possible to associate a

controlled study of cognitive processes with situations

closer to everyday life by enabling an interaction with

the environment. In future, we plan to add more sen-

sors to get additional data such as physiological data

(EMG, ECG, and GSR) for a more in-depth analysis

of users behavior and performance.

REFERENCES

Ahlstrom, C., Victor, T., Wege, C., and Steinmetz, E.

(2012). Processing of eye/head tracking data in large-

scale naturalistic driving data sets. IEEE Transactions

on Intelligent Transportation Systems, 13(2):553–564.

Aoki, H., Nakayasu, H., Kondo, N., and Miyoshi, T. (2010).

Cognitive study on drivers behavior by vehicle trajec-

tory and eye movement in virtual environment. In Pro-

ceedings of the 3

International Conference on Ap-

plied Human Factors and Ergonomics, pages 6768–

6777.

Artz, B., Cathey, L., Grant, P., Houston, D., and Greenberg,

J. (2001). The design and construction of the visual

subsystem for virttex, the driving simulator at the ford

research laboratories. In Proceedings of Driving Sim-

ulation Conference, pages 69–79.

Auberlet, J., Pacaux, M., Anceaux, F., Plainchault, P., and

Rosey, F. (2010). The impact of perceptual treat-

ments on lateral control: A study using ﬁxed-base and

motion-base driving simulators. Accident Analysis &

Prevention, 42(1):166–173.

Bian, D., Wade, J., Swanson, A., Warren, Z., and Sarkar,

N. (2015). Physiology-based affect recognition during

driving in virtual environment for autism intervention.

In Proceedings of the 2Nd International Conference

on Physiological Computing Systems, PhyCS 2015,

pages 137–145, Portugal. SCITEPRESS - Science and

Technology Publications, Lda.

Bian, D., Wade, J. W., Zhang, L., Bekele, E., Swanson,

A., Crittendon, J. A., Sarkar, M., Warren, Z., and

Sarkar, N. (2013). A novel virtual reality driving en-

vironment for autism intervention. In Proceedings

of the 7th International Conference on Universal Ac-

cess in Human-Computer Interaction: User and Con-

text Diversity - Volume 2, UAHCI’13, pages 474–483,

Berlin, Heidelberg. Springer-Verlag.

Blana, E. and Golias, J. (2002). Differences between vehi-

cle lateral displacement on the road and in a ﬁxed-base

simulator. Human Factors, 44:303–313.

attler, C., Ferrari, V., Didierjean, A., and Marm

eche, E.

(2012). Role of expertise and action in motion ex-

trapolation from real road scenes. Visual Cognition,

20(9):988 – 1011.

Breidenbach, C. and Tomaske, W. (2004). The use of

driving simulators in simulating the on-road drive of

wheeled vehicles. In Proceedings of Driving Simula-

tion Conference, pages 333–341.

Burnham, B. R., Sabia, M., and Langan, C. (2014). Com-

ponents of working memory and visual selective at-

tention. Journal of Experimental Psychology: Human

Perception and Performance, 40(1):391–403.

Crundall, D., Chapman, P., and Underwood, G. (2003). Eye

movements and hazard perception in police pursuit

and emergency response driving. Journal of Exper-

imental Psychology: Applied, 9:163–174.

Freyd, J. J. and Finke, R. A. (1984). Representational

momentum. Journal of Experimental Psychology:

Learning, Memory and Cognition, 10:126 – 132.

Galpin, A., Underwood, G., and Crundall, D. (2009).

Change blindness in driving scenes. Transportation

Research (Part F), 12:179–185.

Grabe, V., Pretto, P., Giordano, P. R., and Bultho, H. H.

(2010). Infuence of display type on drivers perfor-

mance in a motion based driving simulator. In Pro-

ceedings of Driving Simulation Conference, pages 81–

88.

Graydon, F. X., Young, R., Benton, M. D., Genik, R., Posse,

S., Hsieh, L., and Green, C. (2004). Visual event de-

tection during simulated driving: Identifying the neu-

ral correlates with functional neuroimaging. Trans-

portation Research (Part F), 7:271–286.

Greenberg, J., Curry, R., Blommer, M., Kozak, K., Artz, B.,

Cathey, L., and Kao, B. (2006). The validity of last-

second braking and steering judgments in advanced

driving simulators. In Proceedings of Driving Simula-

tion Conference, pages 143–153.

Haines, A. (2011). The carrs-q advanced driving simulator.

In Road Safety Forum 2011: Directions in Road Safety

Research.

Hale, R. G., Brown, J., and McDunn, B. A. (2016). Increas-

ing task demand by obstructing object recognition in-

creases boundary extension. Psychonomic Bulletin &

Review, 23:1497–1503.

Holmqvist, K., Nystrom, M., Andersson, R., Dewhurst, R.,

Jarodzka, H., and van de Weijer, J. (2011). Eye Track-

ing: A comprehensive guide to methods and measures.

Oxford University Press, Oxford, United-Kingdom.

Intraub, H. and Richardson, M. (1989). Wide-angle memo-

ries of close-up scenes. Journal of Experimental Psy-

chology: Learning, Memory and Cognition, 15:179–

196.

Jacob, R. (1991). The use of eye movements in human-

computer interaction techniques: what you look at is

what you get. ACM Transactions on Information Sys-

tems (TOIS), 9(2):152–169.

Lavie, N. (2010). Attention, distraction and cognitive con-

trol under load. Journal of Eye Movement Research,

19(3):143–148.

Lavie, N. and Fockert, J. D. (2005). The role of working

memory in attentional capture. Psychonomic Bulletin

& Review, 12:669674.

HUCAPP 2017 - International Conference on Human Computer Interaction Theory and Applications

144

Lavie, N., Hirst, A., Fockert, J. D., and Viding, E. (2004).

Load theory of selective attention and cognitive con-

trol. Journal of Experimental Psychology: General,

133:339354.

Lee, W. S., Sung, D. H., Lee, J. Y., Kim, Y. S., and

Cho, J. H. (2007). Driving simulation for evaluation

of driver assistance systems and driving management

systems. In Proceedings of Driving Simulation Con-

ference North America, pages 153 – 161.

Lei, M., Meng, G., Zhang, W., Wade, J., and Sarkar, N.

(2016). Brain gamma oscillations of healthy peo-

ple during simulated driving. In Proceedings, Part

II, of the 9th International Conference on Intelligent

Robotics and Applications - Volume 9835, ICIRA

2016, pages 453–458, New York, NY, USA. Springer-

Verlag New York, Inc.

Lemonnier, S., Br

emond, R., and Baccino, T. (2014). Dis-

criminating cognitive processes with eye movements

in a decision-making driving task. Journal of Eye

Movement Research, 7(4):1–14.

Lemonnier, S., Br

emond, R., and Baccino, T. (2015). Gaze

behavior when approaching an intersection: Dwell

time distribution and comparison with a quantitative

prediction. Transportation Research (Part F): Trac

Psychology and Behaviour, 35:60–74.

Marberger, C. (2008). Absolute and perceived validity of

the fraunhofer iao immersive driving simulator. In

Proceedings of Driving Simulation Conference, pages

195 – 203.

Mayrhofer, M., Schlusselberger, J. S. R., Wentink, M., and

Bles, W. (2009). Optimal motion control of next gen-

eration simulator motion platform desdemona for car

driving application. In Proceedings of Driving Simu-

lation Conference, pages 173–199.

etrier, E. and Didierjean, A. (2013). Can car driving ex-

pertise modulate boundary extension? L’ann

ee Psy-

chologique :Topics in Cognitive Psychology, 113:523

– 545.

Mori, M., Nakayasu, H., and Miyoshi, T. (2013). Synchro-

nized analyses of eye movement of driver and vehicle

trajectories using driving simulator and eye tracking

system. Transaction of the Japan Society of Mechani-

cal Engineering (Series C), 79(803):2408–2423.

Mourant, R. R. and Rockwell, T. H. (1972). Strategies of

visual search by novice and experienced drivers. Hu-

man Factors, 14:325–335.

Nordmark, S., Jansson, H., Palmkvist, G., and Sehammar,

H. (2004). The new vti driving simulator, multipur-

pose moving base with high performance linear mo-

tion. In Proceedings of Driving Simulation Confer-

ence, pages 45–55.

Reingold, E., Charness, N., Schultetus, R., and Stampe, D.

(2001). Perceptual automaticity in expert chess play-

ers: Parallel encoding of chess relations. Psychonomic

Bulletin and Review, 8:504 – 510.

Reymond, G. and Kemeny, A. (2000). Motion cueing in the

renault driving simulator. Vehicle System Dynamics,

34:249–259.

Rizzo, M. (2002). An abstract virtual environment tool to

assess decision-making in impaired drivers. In Pro-

ceedings of the Second International Driving Sympo-

sium on Human Factors in Driver Assessment, Train-

ing and Vehicle Design. J. D. Lee, M. Rizzo & D. V.

McGehee (Eds.).

Sivak, M. (1996). The information that drivers use: Is it

indeed 90% visual? Perception, 25(9):1081 – 1089.

Snowden, R. J., Thompson, P., and Troscianko, T. (2006).

Basic vision: An introduction to visual perception.

Oxford University Press, Oxford, United-Kingdom.

Sodhi, M., Reimer, B., and Llamazares, I. (2002). Glance

analysis of driver eye movements to evaluate distrac-

tion. Behavior Research Methods, Instruments, &

Computers, 34(4):529–538.

Torrieri, V., D’Apuzzo, M., Galante, F., Pernetti, M.,

Ganieux, D., Troale, A., and Gall

ee, G. (2008). Ef-

fects of road pavement proﬁlles on driver behaviour

in the test driving simulator. In Proceedings of Driv-

ing Simulation Conference, pages 147–159.

Underwood, G., P, . C., Brocklehurst, N., Underwood, J.,

and Crundall, D. (2003). Visual attention while driv-

ing: Sequences of eye ﬁxations made by experienced

and novice drivers. Ergonomics, 46:629 – 646.

Wierwille., W. W. (1993). An initial model of visual sam-

pling of in-car displays and controls. Vision in vehi-

cles, 4:271–282.

Younsi, K., Floris, J., Rajaonah, B., Simon, P., Loslever, P.,

and Popieul, J.-C. (2009). Study of driver’s behavior

during overtaking situations. In Proceedings of Inter-

national Conference on Road Safety and Simulation,

Paris, France.

Zeeb, E. (2010). Daimler’s new full-scale high-dynamic

driving simulator - a technical overview. In Proceed-

ings of Driving Simulation Conference, pages 157–

165.

Zhang, L., Wade, J., Swanson, A., Weitlauf, A., Warren,

Z., and Sarkar, N. (2015). Cognitive state measure-

ment from eye gaze analysis in an intelligent vir-

tual reality driving system for autism intervention.

In Proceedings of the 2015 International Conference

on Affective Computing and Intelligent Interaction

(ACII), ACII’15, pages 532–538, Washington, DC,

USA. IEEE Computer Society.

Using Virtual Reality Techniques to Study Cognitive Processes in Car Driving Activity

145