Analysis of Electroencephalogram and Pulse Waves during Music

Listening

Wenbiao Wang, Yuyu Hu, Mayumi Oyama-Higa, Takashi Suzuki

Department of Systems Innovation, Osaka University, Toyonaka, Osaka 560-8531, Japan

Tiejun Miao

CCI Corporation, Shinagawa-ku, Tokyo 141-0001, Japan

Chaos Technology Research Laboratory, Ōtsu, Shiga 520-2134, Japan

Junji Kojima

Rakuwakai Otowa Hospital, Yamashina-ku, Kyoto 607-8062, Japan

Keywords: Scalp Electroencephalogram, Finger Plethysmogram, Largest Lyapunov Exponent, Parasympathetic Nerve.

Abstract: A tentative study is performed on the psychological effects of music based on the analysis of physiological

indexes. We measured simultaneously the scalp electroencephalograms (EEG) and fingertip pulse waves for

six healthy subjects before, during and after listening to music. The values of their low frequency (LF) and

high frequency (HF) components of heart rate variability are obtained. By the method from chaos analysis,

we calculated the largest Lyapunov exponents (LLE) of both scalp EEG and finger plethysmogram.

Comparing the data of the resting condition and the music-listening condition, we observed significant

tendencies over all subjects. We found that both values of LF and HF decreased, and so did the mean value

taken over 14 EEG channels which were used for computation of LLE, whereas the distribution of the

values tends to become average. Furthermore, it is notable that this averaging degree of LLE of scalp EEG

generally accords with the decreasing rate of HF which reflects the activity of parasympathetic nerves.

1 INTRODUCTION

It is long believed that music has a certain impact on

human mental performance. In recent years

numerous studies have been carried out on exploring

its psychological effect and therapeutic applications

(Unkefer, 1990; McCraty et al., 1998; Schneck and

Berger, 2006). However, just as what Schneck and

Berger mentioned in their monograph, the existing

music literature offers a bewildering array of

unconnected ideas, thoughts, and theories.

Our attention is drawn to changes of certain

physiological indexes that are caused by the

behaviour of listening to music. A preceding work

(Miao et al., 2011) has found that music yields a

decrease in both the largest Lyapunov exponents

(LLE) obtained from finger plethysmograms and

those from the occipital and right cerebral areas of

scalp electroencephalograms (EEG), which gave

good agreement with the theoretical prediction

obtained by a mathematical model they proposed.

The choice of the index LLE is justified, since it

not only characterizes the exponential diverging rate

of trajectories in chaotic systems according to its

definition, but can serve as a significant indicator of

“mental immunity” as discovered by recent studies

(Imanishi and Oyama-Higa, 2006; Oyama-Higa and

Miao, 2006; Oyama-Higa, Miao and Mizuno-

Matsumoto, 2006; Hu et al, 2011). Specifically,

mental health can be kept only if LLE fluctuate

normally over time; continuously low or high values

indicate low adaptability to external environment or

excessive nervousness, respectively. This study still

made use of this index.

In addition, since spectral analysis on heart rate

variability is able to evaluate the activity of the

autonomic nervous system, we took into account the

high frequency (HF, 0.15-0.40 Hz) component,

Miao T., Oyama-Higa M., Kojima J., Suzuki T., Wang W. and Hu Y. (2012).

Analysis of Electroencephalogram and Pulse Waves during Music Listening.

In Proceedings of the Sixth International Symposium on e-Health Services and Technologies and the Third International Conference on Green IT

Solutions, pages 31-35

DOI: 10.5220/0004473700310035

 SciTePress

regarded as an index of parasympathetic nerve

activity, and the low frequency (LF, 0.04-0.15 Hz)

component, influenced by both sympathetic and

parasympathetic nerves. A properly defined ratio

“autonomic nerve balance (ANB)” was also

considered.

To sum up, this tentative study is aimed at

exhibiting and explaining the music effect in terms

of changes in LF, HF and LLE of both finger

plethysmogram and scalp EEG.

2 EXPERIMENT AND METHOD

2.1 Experiment Procedure

The subjects are healthy students from Osaka

University in Japan. They include 5 males and 1

female (labelled with alphabets A to F), whose

average age is 24.63 with a standard deviation of

2.45. Informed consent was obtained from all

subjects. The place of the experiment is an

examination room of Rakuwakai Otowa Hospital in

Kyoto, Japan. The measuring instruments are a

photoplethysmography sensor (Mini PGL, Model

MPULSE-01) and a multi-channel EEG recorder

(Neurofax EEG-1200, developed by Nihon Kohden

Corporation) with 14 active electrodes.

We chose two famous Japanese songs for the

subjects to listen to: Jidai (Time), a 1975 song by

Miyuki Nakajima, and Kawa no nagare no yō ni

(Like the Flow of the River), the last single recorded

by decreased prominent enka singer Hibari Misora.

Both are highly recognized songs, with well-crafted

poetic lyrics and melodic gentle music.

The subjects were asked to lie down on a bed

and keep their eyes closed during the whole process.

When the instruments were ready and the subjects

were relaxed, the first five-minute measurement

began. Then the music was played through their

headphones, while their pulse waves and EEG were

taken for another five minutes. After listening to

music, their resting condition was measured for the

last five minutes.

2.2 Analysis Method

The method for estimating LLE taken from the

subjects is the same with the recent work (Miao et

al., 2011). The improved Rosenstein algorithm (Liu

et al., 2005) was employed to reconstruct the phase

space. The false nearest neighbour method gave the

embedding dimensions d = 4 for time series of

finger plethysmograms and d = 8 for that of scalp

EEG. The first minimum of average mutual

information (Fraser and Swinney, 1986) was applied

to determine the time delay. We found the time

delay being 50ms for both plethysmogram and EEG

in the experiments.

LF and HF were obtained by the analysis

software “Lyspect” (Oyama-Higa et al., 2012),

developed by Chaos Technology Research Lab. The

results are displayed in a panel (Figure 1), where the

line graph at the bottom shows changes of LF (in

red) and HF (in blue) over time. The autonomic

nerve balance (ANB), shown in the right-side

semicircular graph, is defined as a normalized value:

(1)

Thus, ANB < 5 indicates predominance of

parasympathetic nerve while ANB > 5 indicates

sympathetic predominance.

Figure 1: Lyspect analysis results.

3 ANALYSIS AND RESULT

3.1 LLE of Scalp EEG

Figure 2 shows the mean values of LLE over the 14

scalp EEG channels, obtained from the subjects

during the three five-minute conditions: before,

during and after listening to music.

The specific changes at each channel are

displayed by topographical two-dimensional maps

(Figure 3), in which deeply coloured area indicates a

high value and vice versa.

EHST/ICGREEN 2012

Figure 2: LLE of EEG: mean values taken over the 14

channels.

Figure 3: LLE of EEG: distribution at each of the 14

channels.

From Figure 2, we found in all subjects but

Subject E a “decreasing then increasing” tendency in

the mean of LLE taken over the EEG channels.

Besides, what deserve more attention are the maps

of Figure 3. We observed that, in all subject except

Subject E, the distribution of LLE at EEG channels

became average after they listened to music.

3.2 LLE of Finger Plethysmogram

Figure 4 exhibits LLE of finger plethysmogram, also

taken from the conditions prior to, during and

posterior to music listening, respectively.

Figure 4: LLE of plethysmogram.

There exist both increasing (Subjects C, D, E and

F) and decreasing (Subjects A and B) tendencies in

the subjects. Moreover, the average LLE of

plethysmogram over the 6 subjects increased after

exposure to music. Therefore, in terms of changes in

LLE of finger plethysmogram, this result does not

correspond with our preceding study (Miao et al.,

2011).

3.3 HF and LF

Changes in HF and LF throughout the three

conditions are illustrated by Figure 5.

Figure 5: HF and LF

With the exception of Subject E, the main

tendency is obvious: both values of their LF and HF

Analysis of Electroencephalogram and Pulse Waves during Music Listening

decreased under the influence of music.

Moreover, together with Figure 3, a notable

phenomenon is observed: the decreasing rate of HF

from “Before” condition to “Music” condition is

consistent, in general, with the degree of averaging

in LLE of EEG under “Music” condition. Since the

latter is represented by the standard deviation, a

comparison is given in the following tables.

Table 1: Decreasing rate of HF and standard deviation of

LLE of 14 EEG channels under “Music” condition.

Subjects

Decreasing rate of HF from

"Before" to "Music"

Rank

A 39.77

B 58.25

C 31.72

D 7.31 5

E -50.38 6

F 13.46 4

Subjects

Standard deviation of LLE of 14

EEG channels during "Music"

Inverse

rank

A 27.31 2

B 17.22 1

C 28.62 3

D 34.55 4

E 42.17 6

F 36.39 5

3.4 ANB

Finally, we took into account changes of ANB, as

shown in Figure 6.

Figure 6: ANB

It can be observed that ANB of each subject was

maintained near the balance value 5 throughout the

three conditions, disregarding that for almost all of

them the LF and HF experienced a decrease and then

an increase.

4 CONCLUSIONS AND REMARK

This study observed changes in LLE of both scalp

EEG and finger plethysmogram, as well as LF and

HF components of heart rate variability. Several

significant tendencies were found out in all subjects

excluding one exception.

As a complement to our results, we discovered

later that the same tendencies can still be observed in

Subject E, who always served as an exception as

stated above, if he listened to his favourite songs. In

fact, according to an interview of the 6 subjects after

the experiment on their feelings towards the two

songs, Subject E harboured an antipathy, while the

rest 5 subjects felt indifferent to or showed some

appreciation for them.

We remark that not many studies in this field

exist so far evolving the joint analysis of LLE of

both EEG and plethysmogram. Furthermore, the

consideration of LF and HF is what distinguishes

this study from the previous work (Miao et al., 2011).

However, there still exist several limitations in

this study. The changes in LLE of plethysmogram

and ANB were not instrumental in the explanation,

and analysis was not performed in terms of the LLE

at each specific EEG channel. The underlying

relationship between the music effect and the

listener’s feeling towards the music is unclear. What

is more, to test with only 6 subjects hardly sufficed

to convince ourselves of the generality of the result.

To overcome these limitations is a subject for future

work.

ACKNOWLEDGEMENTS

The authors would like to express their sincere

appreciation to Nihon Kohden Corporation for

generously providing us with the measuring

instrument.

REFERENCES

Fraser, A.M. and Swinney, H.L. (1986). Independent

Coordinates for Strange Attractors from Mutual

Information. Phys Rev Lett., 33, 1134–1140.

Hu, Y., Wang, W., Suzuki, T. and Oyama-Higa, M. (2011).

Characteristic Extraction of Mental Disease Patients

by Nonlinear Analysis of Plethysmograms.

International Symposium on Computational Models

for Life Sciences, AIP Conference Proceedings 1371,

92-101.

EHST/ICGREEN 2012

Imanishi, A. and Oyama-Higa, M. (2006). The Relation

Between Observers’ Psychophysiological Conditions

and Human Errors During Monitoring Task. 2006

IEEE Conference on Systems, Man, and Cybernetics,

2035–2039.

Liu, H.F., Dai, Z.H., Li, W.F., Gong, X. and Yu, Z.H.

(2005). Noise Robust Estimates of the Largest

Lyapunov Exponent. Physics Letters A., 341, 119–127.

McCraty, R., Barrios-Choplin B., Atkinson M. and

Tomasino D. (1998). The Effects of Different Types of

Music on Mood, Tension, and Mental Clarity.

Alternative Therapies In Health And Medicine, 4(1),

75-84.

Miao, T., Oyama-Higa M., Sato S., Kojima J., Lin J. and

Sato, R. (2011). Plethysmogram and EEG: Effects of

Music and Voice Sound. International Symposium on

Computational Models for Life Sciences, AIP

Conference Proceedings 1371, 282-289.

Oyama-Higa, M. and Miao, T. (2006). Discovery and

Application of New Index for Cognitive Psychology.

2006 IEEE Conference on Systems, Man, and

Cybernetics, 2040–2044.

Oyama-Higa, M., Miao, T., Kaizu, S. and Kojima, J.

(2012). Mental Health Self-check System Using

"Lyspect". In preparation.

Oyama-Higa, M., Miao, T. and Mizuno-Matsumoto, Y.

(2006). Analysis of Dementia in Aged Subjects

Through Chaos Analysis of Fingertip Pulse Waves.

2006 IEEE Conference on Systems, Man, and

Cybernetics, 2863–2867.

Schneck, D.J. and Berger, D.S. (2006). The Music Effect:

Music Physiology and Clinical Applications (1st ed).

London: Jessica Kingsley Publishers.

Unkefer, R. F. (1990). Music Therapy in the Treatment of

Adults With Mental Disorders: Theoretical Bases and

Clinical Interventions. New York: Schirmer Books.

Analysis of Electroencephalogram and Pulse Waves during Music Listening