evo samo za vktr-a, hoces cijeli dokument u pdf-u??
EFFECTS on EYE-FIXATION-RELATED POTENTIAL
of a DRIVER’S CONVERSATION and VISUAL SEARCH
Yoshihisa TERADA, Toru NAKADA, Koji MORIKAWA
Panasonic Corporation
3-4 Hikaridai, Seika-cho, Soraku-gun, Kyoto 619-0237, Japan
TEL: +81-774-98-2410 FAX: +81-774-98-2533
E-mail: {terada.yoshihisa, nakada.toru, morikawa.koji}@jp.panasonic.com
Yong Wook JEON, Tatsuru DAIMON
Faculty of Science and Technology, Keio University
3-14-1 Hiyoshi, Kouhoku-ku, Yokohama-shi, Kanagawa 223-8522, Japan
Tel: +81-45-566-1441 Fax: +81-45-566-1617
E-mail: {jyw0673, daimon}@ae.keio.ac.jp
ABSTRACT
This paper describes the effects of a driver’s conversation and visual search on
eye-fixation-related potential (EFRP). Experiments are conducted with a driving simulator.
Electroencephalogram (EEG) is performed when the participants were assigned a driving task
and a cognitive task concurrently. Cognitive tasks include conversation, visual search, and
manipulation of a mobile phone. EFRP is calculated by averaging the EEG waves whose
offsets are determined by saccadic eye movements. The amplitude of the EFRP signal is used
for estimating the level of a driver’s distraction. The results show that the amplitude of
lambda response in EFRP decreases when considerable attention is required to execute a
cognitive task. Simulation results using the obtained ERRP data are also discussed.
INTRODUCTION
Drivers have to pay attention to many things in addition to driving a car, such as conversation
with passengers and manipulation of the car navigation system, audio player, and air
conditioner. For supporting safe driving, the physical and psychological conditions of drivers
should be considered. In particular, the arousal level and the distraction level affect safe
driving. The arousal level is related to the driver’s level of drowsiness, and the distracted state
is the case when the driver does not completely concentrate on driving in spite of a high
arousal level. Conventionally, the blink frequency and alpha waves of the
1
electroencephalogram (EEG) have been used for assessing the driver arousal level [1] [2].
However, few studies have been reported on assessing the driver’s distraction levels.
Event-related brain potential (ERP), which is a type of EEG, has been proposed as a tool for
assessing attention [3]. Ebe et al. [4] applied the P300 component of the ERP to assess the
level of driving concentration. In the study, P300 components are elicited by visual stimuli
such as brake light of the front car, LED light at the car pillar, or projected stimuli at the
windshield and obtained. Although visual stimuli are necessary to obtain P300 responses, it is
difficult to apply these in a practical situation.
Eye-fixation-related potential (EFRP) has been proposed as an index that does not require
visual stimuli. EFRP is a type of EEG potential that is associated with saccadic eye
movements instead of visual stimuli. Yagi [5] has shown that EFRP reflects the level of visual
attention. Miki [6] has reported that EFRP can be used as an index of the distraction level for
a usability evaluation of manipulation in a car. They evaluated the difference in distraction
levels using the EFRP amplitude between the manual operation and the voice operation of a
car navigation system.
However, the manner of assessing the level of driving concentration and distraction caused by
the driver’s thought process such as conversation or memory recall, which are cognitive tasks
that require concentration for recalling memory, remains unclear. The aim of our study is to
evaluate a driver’s distracted state, which is a state of low driving attention caused by
thinking.
Our previous study has shown the relationship between EFRP and the distracted state caused
by an n-back task [7]. The n-back task needs short-term memory, and the drivers had to
memorize, recall, and respond to the number presented sequentially while using a driving
simulator. The lambda response of EFRP increased when the driver could completely
concentrate on driving. In contrast, the lambda responses of EFRP decreased when the driver
found it difficult to concentrate completely on driving while carrying out an n-back task.
Although the n-back task is a thinking task, there are many distractions besides a short-term
memory task, such as long-term explicit memory tasks that consist of an episodic memory
recall task and a semantic memory recall task, visual search task for navigation, and the
manipulation of appliances. In this study, we have focused on these distractions and reported
the experimental results of EFRP during various tasks.
METHODS
To examine the effect of the distraction levels on EFRP, EEGs and electrooculograms (EOGs)
are performed under various mental workload conditions.