Evaluation of EEG Responses to Sedative and Stimulative Music Using the Muse 2
Jenna B.

1
Duncan ,

Rachael L.

1Department

2
Kovaly ,

Abigail G.

2
Metcalf ,

of Biology, Slippery Rock University;

Introduction
Studies using electroencephalography (EEG) to examine responses to musical
stimulation have been conducted since 1959 (Fachner & Stegemann, 2013). While
EEG is a widely used method for investigating mental activity by measuring
electrical signals in the brain, this method is cumbersome and costly. Thus, research
and clinical applications using EEG are limited. With technological advancements,
portable and commercially used EEG devices, such as the MUSE 2 headset,
provide opportunities to investigate brain waves in front-temporal regions (What it
measures, 2019), which are known to be affected by music (Fachner & Stegemann,
2013; Kucikiene & Praninskiene, 2018). The use of a portable EEG device to
examine brain function during musical activity may inform music therapy practice by
confirming biological principles underlying current therapeutic approaches,
developing biological markers predictive of treatment outcomes, and providing livetime biofeedback during therapy sessions.
The purpose of the present study is to evaluate EEG responses to sedative (slow)
and stimulative (fast) music using the MUSE 2 EEG headset, comparing such
activity with self-reported mood ratings for each category of music. The self-reported
mood rating component of the study replicates work done by Boyle (1982) while
adding EEG measures provides novel and objective mood-related data. Overall, the
study is intended to be the first of many that examine the relationship between music
and brain activity using a commercially available EEG recording device.

Method
Participants
Eighteen volunteers from the SRU graduate Physician Assistant and Occupational
Therapy programs participated in this study, without incentive. The average
participant age was 24.3 years. A majority of participants identified as women (n =
16, 89%) and White (n = 16, 89%) and a smaller number of participants identified as
male (n = 2, 11%) and biracial (n = 2, 11%). Participants' length of training in music
varied from no experience (n = 7, 39%), 1 – 4 years experience (n = 2, 11%), to 5 or
more years experience of training in music lessons or ensembles (n = 9, 50%).

Materials
Participants' brain wave activity was recorded using the MUSE 2 (Model MU03; InterAxon Inc.) headset. In addition to 3 reference electrodes, this headset
records brain activity from the AF7, AF8, TP9, and TP10 positions (See Figure
1). The headset is positioned across the forehead and secured behind the
ears. TOZO-10 wireless Bluetooth earbuds were utilized to play musical selections
from a linked iPad.

Music Characteristic
Stimulative
Stimulative
Stimulative
Sedative
Sedative
Sedative

Composer
Gould
Bernstein
Sousa
Copeland
Albert
Brahms

Nicole D.

2Department

2
Hahna ,

Vern H.

2
Miller ,

and Amber M.

1
Eade

of Music, Slippery Rock University

Composition Title
“Ride Out”
“Prelude, Fugue, and Riff”
“Stars and Stripes Forever”
“Concerto for Clarinet and Orchestra”
“Feelings”
“Bb Piano Concerto”

Table 1: Stimulative and Sedative Excerpts

Procedure

Adjective Pair
Sedative 𝑥
Happy or Sad
2.94
Restless or Calm
4.41
Joyous or Gloomy
3.35
Whimsical or Serious
3.46
Vigorous or Quiet
4.48
Majestic or Soothing
3.74
Playful or Dignified
3.93
Exhilarated or Dreamy
4.35

Stimulative 𝑥
1.65
2.02
1.63
1.91
1.57
1.98
1.67
1.54

Difference
1.29
2.39
1.72
1.55
2.91
1.76
2.26
2.81

t
5.71
9.83
8.04
4.58
20.1
7.59
7.28
15.36

p
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001

Table 2: Analysis of the combined mean mood response to sedative and stimulative music for all
18 participants.

Before beginning the study, participants were screened for COVID-19 symptoms. If
no symptoms were identified, the informed consent was read and signed.
Participants were randomly assigned into one of two groups (A or B), which
consisted of playlists that alternated the presentation of stimulative and sedative
music in a counterbalanced fashion. They were then presented the following videos
to watch while interacting with the researcher via zoom:
• Measuring of participant head circumference
• Circumference then recorded by the researcher
• A four-part video series on placement of the MUSE 2 headband and TOZO-10
Bluetooth earbuds
• Connectivity was tested by having the participant relax facial muscles and
close their eyes. If problems arose, the participant was asked to adjust the
location of the headband and connectivity was assessed again.

The present data illustrates significant differences between self-reported mood
responses of participants to sedative and stimulative music, replicating the results
found by Boyle (1982). These findings were not fully mirrored in the brain activity
recorded by the Muse 2 headband, as a significant difference between responses
while listening to sedative and stimulative music was found at only one frequency
point on a single sensor. Even so, no official conclusions or correlations can be
made at this time, as the present data represent only preliminary results.

Once the headband and earbuds were successfully connected, a C Major scale was
played through the Bluetooth earbuds to confirm the participant could hear the music
within a 60-80dB range. Before playing each subsequent excerpt of music, the
participant was asked to sit as still as possible with their eyes closed for the duration
of the stimulus. Using the MindMonitor application in conjunction with the MUSE
2 headband, a recording of brain activity was completed while each excerpt
played. Following each excerpt, participants were asked to rate the selection on
a scale of one to five for each of seven adjective pairs: happy/sad, restless/calm,
joyous/gloomy,
whimsical/serious,
vigorous/quiet,
majestic/soothing,
playful/dignified, and exhilarated/dreamy (Boyle, 1982). This same procedure was
repeated for each of the remaining musical excerpts.

Due to COVID-19, the onset of this study was delayed from June 2020 to February
2021, greatly decreasing the availability of participants and research facilities while
also leading to substantial procedural modifications. In adherence with COVID-19
restrictions, the following were implemented: additional cleaning procedures, use of
air ionizers, participant screening for COVID-19 related symptoms, isolation of
participants and researchers into separate rooms, and ensuring research spaces
were unoccupied for 30 minutes between participants. These additional procedures
altered the efficiency of this study, adding length to the study setup, each participant
session, and time between participants. As such, data from only a portion of
the planned participants for this study have been collected to date. Participants are
continuing to be recruited and tested into Summer 2021.

Results
Analysis of EEG recordings using the MATLAB EEGLAB toolbox revealed a single
significant (p < 0.05) difference between stimulative and sedative responses in the
gamma frequency range for the AF8 sensor only (Figure 2). There were no other
significant findings for any of the 4 data sensors. A 2-tailed t-test of combined
participants mean mood ratings revealed significant (p < 0.001) differences between
sedative and stimulative music for all eight adjective pairs (Table 2).

Discussion

Additional limitations of this study include:
• Muse 2 headset recording from only 4 brain locations, giving limited EEG data.
• Potential loss of data from sensors that fail to connect properly due to participant
head size/shape
• Potential for needed earbud adjustment mid-study due to size differences in
participants ear canals

References
Boyle, J. D. (1982). College students’ verbal descriptions of
excerpts of stimulative and sedative music. Research Symposium on the
Psychology and Acoustics of Music, 105-117

Seven musical selections were utilized in this study, 3 stimulative, 3 sedative, and 1
neutral. The stimulative and sedative excerpts were selected in accordance with
previous literature (Boyle, 1982; see Table 1), while the neutral piece was composed
specifically for this study and was played as the first excerpt for each participant.

Fachner, J., & Stegemann, T. (2013). Electroencephalography and music therapy:
On the same wavelength? Music and Medicine, 00 (0), 16. doi:10.1177/1943862113495062
Kučikienė, D., & Praninskienė, R. (2018). The impacts of music on the
bioelectrical oscillations of the brain. ACTA Medica Lituanica, 25(2), 101106. doi:10.6001/actamedica.v25i2.3763
What it measures. (2019). Retrieved from https://choosemuse.com/what-itmeasures/

Acknowledgements
•
Figure 1: Left: Image of TOZO-10 earbuds and Muse 2 headband. Right: Relative channel
positions for AF7, AF8, TP9, and TP10 sensors

Figure 2: Frequency Spectrum Graph for AF8 sensor. Significant finding represented by the gray bar
at ~90 Hz (p < 0.05)

•

This study was approved by the SRU Institutional Review Board (IRB Protocol
2021-002-08-B)
Funding for this project was obtained through a College of Health, Environment,
and Science Research Grant