Impact of Auditory Listening on Emotional States in Self-paced

Outdoor Running

Klemens Weigl

1,2 a

, Sinja Becker

1 b

, Karolin Bosch

1 c

, Nhien Thai

1 d

and Andreas Riener

1 e

Human-Computer Interaction Group, Technische Hochschule Ingolstadt (THI), Esplanade 10, Ingolstadt, Germany

Department of Psychology, Catholic University Eichst

att-Ingolstadt, Germany

Keywords:

Music, Audiobook, Emotional States, Portable Technologies, Self-paced Running, Outdoor.

Abstract:

Studies have shown that listening to music may evoke positive emotions when running. In recent years, the

growing trend of listening to audiobooks is also inevitably inﬂuencing running. However, to date, there has

been little research on the use of audiobooks in self-paced outdoor running. Therefore, our objective is to

investigate whether self-paced outdoor running with auditory stimuli such as music and audiobooks may elicit

different emotional states when compared to running with no-audio. Consequently, we adopted a repeated-

measures design with three different counter-balanced conditions: music, audiobook, and no-audio. Thereby,

thirty-two recreationally active female and male runners participated in a 10-minute running trial in each con-

dition. We assessed the impact of auditory stimuli and emotional states with the Sports Emotion Questionnaire

and the Brunel Music Rating Inventory-2. Our results of the self-report questionnaires indicate that running

with music is rated with substantially greater values when compared to the audiobook condition. Interestingly,

our ﬁndings uncover no meaningful difference among the self-rated emotional state dimensions anxiety, de-

jection, excitement, anger, and happiness across all three conditions, respectively. Hence, we conclude that

running with no-audio may evoke roughly the same emotional states as when compared to running with music

or audiobooks.

1 INTRODUCTION

With increasing digitization and the spread of new

technologies, the way we perceive and consume me-

dia is remarkably changing. Since several years,

portable devices such as MP3 players have gained

widespread popularity amongst runners around the

globe. Thereby, numerous recreationally active as

well as professional runners across age and gen-

der prefer listening to music when running outdoors.

Running has led manufacturers of music-enabled de-

vices to develop ever lighter and more ergonomic de-

vices (Karageorghis, 2016; Bigliassi et al., 2015).

Many runners prefer mobile listening to music by the

use of headphones which may create an acoustic en-

vironment that is set apart from the outside world and

dissociates themselves into a kind of cocoon (Bull,

2015). Beneﬁcial effects of music in sports have espe-

https://orcid.org/0000-0003-2674-1061

https://orcid.org/0000-0002-2418-2756

https://orcid.org/0000-0003-4520-1749

https://orcid.org/0000-0003-3160-773X

https://orcid.org/0000-0002-9174-8895

cially been found in repetitive activities like running

(Karageorghis and Priest, 2012a; Terry and Kara-

georghis, 2011). In general, musical effects are ex-

plored in the ﬁeld of exercise and sport in four areas

of research: (1) psychological, (2) psychophysical,

(3) ergogenic, and (4) psychophysiological effects of

music (Terry and Karageorghis, 2011; Karageorghis

et al., 2012; Karageorghis and Priest, 2012a; Kara-

georghis, 2016).

1.1 Audiobooks

Since the advent of mobile technology traditional

books can be provided as a digital audiobook which

is characterized by the small portable format that

creates new affordances and a multitude of possible

applications (Have and Pedersen, 2015). Conceptu-

ally, audiobooks are deﬁned as a recording of a book

that is performed by a professional narrator or the

author (Have and Pedersen, 2015). Affordances can

be advantages such as their convenience, portability,

and supplementary status to other activities, meaning

audiobooks can easily be accessed whilst engaged

Weigl, K., Becker, S., Bosch, K., Thai, N. and Riener, A.

Impact of Auditory Listening on Emotional States in Self-paced Outdoor Running.

DOI: 10.5220/0009980201890196

In Proceedings of the 8th International Conference on Sport Sciences Research and Technology Support (icSPORTS 2020), pages 189-196

ISBN: 978-989-758-481-7

189

with another activity. Moreover, it was found that

81% of the surveyed audiobook listeners enjoy

audiobooks because they are also able to do other

things in the process of listening, whereby the top

three activities while listening to audiobooks are

driving, relaxing before going to sleep and doing

housework (Audio Publishers Association, 2018).

Consumer studies indicate that the use of smart-

phones for audiobooks accounted for 73% in 2018

and are expected to continue to grow (Audio Publish-

ers Association, 2018). In recent years, audiobooks

have rapidly gained popularity (Audio Publishers

Association, 2019) and listening to audiobooks has

become more accessible and user-friendly thanks to

mobile devices and access to digital downloading

and streaming services (Rubery, 2011; Have and

Pedersen, 2012; Have and Pedersen, 2015). For

instance, the music streaming service Spotify offers a

large library of more than 50 million music tracks, in-

cluding audiobooks, and podcasts. By using Spotify’s

free version with advertisement, users can access a

huge audio catalogue without any charge, while a

pay-per-month subscription allows users to synchro-

nise playlists for ofﬂine mobile usage (Spotify USA

Inc, 2020; Kreitz and Niemela, 2010). Moreover,

listening to audiobooks is saving time by enabling

users to simultaneously perform other tasks (Have

and Pedersen, 2015). This empowerment promotes

activity and at least dual-tasking which corresponds

to our modern, fast-paced society. Additionally, user

experience is positively inﬂuenced by the enjoyment

of the ﬂexibility of listening to audiobooks wherever

the users are.

Music versus Audiobook

In a direct comparison of the two conditions treadmill

running with music versus treadmill running with an

audiobook, both, ratings of enjoyment were signiﬁ-

cantly higher and ratings of perceived exertion were

substantially lower in the music condition compared

to the audiobook condition, respectively (Miller et al.,

2010). They concluded that the situational context

was more positively altered by music, than by the di-

alog condition audiobook. In a 12-min. cycling-bout,

music reduced ratings of perceived exertion when

compared to an audiobook and a control condition

(Bigliassi et al., 2017). Furthermore, based on mea-

surements with an electroencephalogram (EEG), in

the music condition the attentional focus of the par-

ticipants was reallocated toward auditory pathways,

while inhibiting alpha resynchronization at the Cz

electrode and reducing the spectral coherence val-

ues at Cz-C4 and Cz-Fz and a reduction of the fo-

cal awareness at light-to-moderate-intensities, which

resulted in a more autonomous control of cycle move-

ments in contrast to the two other conditions, re-

spectively. In another self-paced walking study, it

was identiﬁed that music induced more dissociative

thoughts by the up-regulation of beta waves as well as

arousal and perceived enjoyment in comparison with

a podcast, and a control condition (Bigliassi et al.,

2019). This reallocation of attention on external stim-

uli in ecologically valid settings elicited more positive

affective responses in the music condition in contrast

to podcast and control.

However, although ratings of perceived exertion

were similar across music, podcast, and no-audio

sprint interval training (SIT) conditions, researcher-

selected motivational music enhanced affective re-

sponses and enjoyment to a greater extend than in the

no-audio control condition, tentatively more than in

the podcast condition (Stork et al., 2019). Moreover,

a higher peak power output could be identiﬁed in the

music condition when compared to the podcast and

no-audio condition.

A dialogic condition such as an audibook has re-

ceived scant research attention though it possesses ex-

ternal validity in this exercise domain (Karageorghis

and Priest, 2012a).

1.2 The Present Study

To date, little is known if the increasingly popular

trend of listening to an audiobook may be perceived

as positive as listening to music in the externally valid

context of self-paced outdoor running. Moreover, yet

it is unclear, whether or not emotional states in self-

paced outdoor running are affected and reported sim-

ilarly for running with music, an audiobook, or no-

audio. Consequently, in the present study, we investi-

gated the following research questions (RQ) and hy-

potheses (H):

RQ1: Music versus Audiobook. Do runners prefer

listening to music or to an audiobook in self-paced

running?

: We hypothesize that runners assign substantially

more positive ratings to the music than to the audio-

book condition in self-paced running.

RQ2: Difference between Emotional States be-

tween Running Conditions. Do runners report

different emotional states when running self-paced

dependent on auditory input?

2.1

. We assume that runners assign signiﬁcantly

more positive ratings to the positive emotional

state dimensions excitement and happiness in the

music condition compared to the audiobook and the

no-audio (baseline) condition in self-paced running.

icSPORTS 2020 - 8th International Conference on Sport Sciences Research and Technology Support

190

Thereby, we expect no meaningful difference be-

tween the audiobook and the no-audio condition.

2.2

. We hypothesize runners to assign noticeably

lower ratings to the negative emotional state dimen-

sions anxiety, dejection, and anger in the music

condition compared to the audiobook and the no-

audio (baseline) condition in self-paced running. At

the same time, we assume no considerable difference

between the audiobook and the no-audio condition.

RQ3: Associations of Emotional States and Audio

Ratings. Which associations can be found between

emotional states and audio ratings in self-paced run-

ning?

. We assume that, in self-paced running, the

self-report emotional state dimensions excitement and

happiness are positively associated with audio ratings,

whereas the self-report dimensions anxiety, dejection,

and anger are negatively associated with audio ratings

in both the music and the audiobook condition.

2 METHOD

2.1 Participants

Thirty-two recreationally active runners (11 female,

21 male) between 19 and 51 years (M = 25.34; SD

= 6.62) were recruited to participate in this study.

Recreationally active was deﬁned as regularly engag-

ing in any strenuous physical activities such as aero-

bic, running, cycling or swimming at least once per

week. According to their own statements, the study

participants spent an average of 3.7 days per week

(SD = 1.89) on strenuous exercise or training and

run on average 6.48 km per unit. Subjects were vol-

unteers and participated without compensation. In

order to engage in the study, potential participants

were required to meet the following inclusion crite-

ria: regular participation in sports exercise for recre-

ational purposes, prior experience in recreational run-

ning, and no contradictions for performing 30 min-

utes at self-paced running. Anyone with orthopedic

limitations which may have prevented them from en-

gaging in 30 minutes of running was excluded. Fur-

thermore, all participants were ﬂuent in [blanked] the

language in which the questionnaires were provided,

consumed no alcohol or drugs, and reported no diag-

nosis of a psychiatric or neurological disorder. Prior

to participating, they were required to provide writ-

ten informed consent. An institutional e-mail was dis-

tributed among members of the university and among

the members of a sports group in [blanked]. The study

information sheet was attached to this e-mail, explain-

ing the objectives and potential risks associated with

the study. The majority of the sample comprised of

students and lecturers.

2.2 Design

We adopted a repeated measures design (3 x 10-

minute of self-paced outdoor running) with three dif-

ferent conditions (independent variable): music (M),

audiobook (A), and control (C). The dependent vari-

ables (DVs) were the ﬁve dimensions of the Sports

Emotion Questionnaire and the single-factor of the

Brunel Music Rating Inventory-2 (further described

in 2.3). The study was conducted outdoors in an eco-

logically valid setting with an off-road and ﬂat run-

ning course on the campus (cf. Figure 1) with no alti-

tude difference. The distance of one lap was roughly

450 m.

Figure 1: Flat and off-road running course.

2.3 Measures and Instruments

To assess the psychological impact of auditory

stimuli (M and A) on emotional states and feelings,

we applied the Sports Emotion Questionnaire (SEQ)

(Jones et al., 2005) in each condition (M, A, and C)

and a slightly adapted version of the Brunel Music

Rating Inventory-2 (BMRI-2) (Karageorghis et al.,

2006) in both auditory conditions (M and A).

Sports Emotion Questionnaire

Emotional states and feelings of participants were

measured after each condition using the sport-speciﬁc

Sports Emotion Questionnaire (SEQ). The SEQ com-

prises of a 5-factor structure assessing the self-report

emotional state dimensions anxiety, dejection, ex-

Impact of Auditory Listening on Emotional States in Self-paced Outdoor Running

191

citement, anger, and happiness. The questionnaire

inquired participants’ currently perceived emotional

states and feelings by 22 items, each consisting of

an adjective such as ”uneasy“, ”exhilarated“, or

”cheerful“. Participants responded to the items using

a 5-level likert-scale anchored at 0 (not at all) and 4

(extremely).

Brunel Music Rating Inventory-2

To determine participants’ perceived motivational

impact of the auditory conditions M and A, the

Brunel Music Rating Inventory-2 (BMRI-2) was

applied after each auditory listening condition. This

single-factor, six-item instrument, consists of dif-

ferent auditory components: rhythm, style, melody,

tempo, instrumentation, and beat. The response

options were limited to a 7-point Likert scale ranging

from 1 (strongly disagree) to 7 (strongly agree) for

each item, for example, ”The rhythm of the music

would motivate me during exercise” (Karageorghis

et al., 2006, p. 909). Hence, the summated total

score ranges from 6 to 42, which is deemed as a

measure for the motivational quotient of music that

may stimulate or inspire physical activity which is

described as high (36–42), moderate (24–35), and

oudeterous (< 24) (Karageorghis, 2008). Reliability

has been reported between .70 and .90, Cronbach’s

Alpha coefﬁcient with .95, and references to criterion

validity have been stated (Karageorghis et al., 2006).

In the present study, we wanted to compare the

participants’ self-ratings of the motivational quality

of music with the motivational quality of audiobooks.

Henceforward, we applied a slightly modiﬁed version

of the BMRI-2 in the audiobook condition in order to

account for these speciﬁc motivational qualities (cf.

OSF link below).

Supplementary Materials: Open Science

We support the open science movement and supply

the data set, the music playlist, the modiﬁed items of

the BMRI-2, and a detailed description of the mu-

sic and audiobook selection for our study on OSF:

https://osf.io/urymp/.

2.4 Procedure

After a friendly welcome, each participant received

an introduction to the background and the main goal

of the study with all necessary information. Addition-

ally, everyone got a written summary about the study.

Then the information sheet on informed consent was

handed out, which all of them signed. Thereby, ev-

eryone was explicitly informed, that the participation

in this self-paced outdoor running study is voluntary.

In addition, everyone was instructed about the possi-

bility to stop at any time prior to the ofﬁcial end of the

study without receiving any consequences. However,

all of them completed the whole study without facing

any difﬁculties. After this introductory part, all par-

ticipants started with the initial 10-minute self-paced

running condition (C) with no-audio, which was com-

pleted only with the SEQ. In the next phase, the al-

location of all participants was randomly counterbal-

anced on either of the auditory listening conditions

(M and A). For example, the ﬁrst participant started

the ﬁrst 10-minute self-paced running condition with

music. After running, this auditory condition (M) was

rated with the BMRI-2 and the emotional states and

feelings were assessed with the SEQ. Then the par-

ticipant continued with 10-minute self-paced running

with the audiobook. After ﬁnishing (A), the BMRI-2

and the SEQ were completed. In contrast, the second

participant ﬁrst received the A condition, followed by

the M condition, and so on.

Auditory Conditions

We prepared several different devices (MP3, iPod,

Smartphone) for the M (a compilation of tracks from

the most popular Spotify running playlist with a pace

ranging from 150 to 165 beat per minute (BPM)) and

A condition to reduce organizational effort and ease

the participation of the runners. Every participant

could adjust the volume when running. Each auditory

stimuli was played throughout the respective exercise

trial via the participants’ headphones. After 10 min-

utes in operation, the audio automatically stopped,

signaling each participant to come back to the start-

ing point. There the subsequent assessments with the

questionnaires awaited them with a duration of about

5 minutes.

Upon completion of all three conditions, a sheet

with a take-home-message about the study was

handed out, accompanied with the contact and afﬁl-

iation details of the examiner, in case any questions

may arise later on. All data for this study were col-

lected anonymously. The entire participation dura-

tion lasted from approximately 45 to 60 minutes. The

study was carried out over a period of two days under

good weather conditions.

2.5 Statistical Analyses

The signiﬁcance level has been set to α = .05, if not

stated otherwise (e.g., in case of Bonferroni correc-

tion). Hence, all results with p < α are reported as sta-

tistically signiﬁcant. Since no negatively coded items

were included in the questionnaire, the total scores

icSPORTS 2020 - 8th International Conference on Sport Sciences Research and Technology Support

192

could be directly calculated for all ﬁve dimensions of

the SEQ and the single-factor of the BMRI-2. All

data were analyzed using IBM



SPSS



Statistics,

Version 25 (IBM Corp., 2017).

RQ1. Because of the rating scale response for-

mat of the BMRI-2, we focused on the single-factor

composite total score for which we could assume at

least pseudo-metric level data. Additionally, Shapiro

Wilk’s test revealed that the assumption of normality

of the BMRI-2 (DV) has been met for both conditions

M and A. Therefore, we applied a two samples depen-

dent t-test.

RQ2. Since the ﬁve emotional state dimensions of the

SEQ (DVs) are assessed with a Likert scale response

format yielding ordinal data, we applied the nonpara-

metric Friedman test for repeated measures with the

three conditions (i.e., M, A, and C) as within-subjects-

factor (IV). Moreover, normal distribution has not

been met for the negative emotional state dimensions

anxiety, dejection, and anger in either of the three con-

ditions.

RQ3. Given the ordinal data of the SEQ we applied

the nonparametric Spearman rank correlation.

Supplementary Analysis. In a supplementary and

exploratory statistical analysis, we performed a mul-

tiple linear regression analysis with the stepwise se-

lection method for the M and the A condition, respec-

tively.

3 RESULTS

We investigated the following three research ques-

tions (RQ1 to RQ3) with the accompanying two-sided

alternative hypotheses of interest (H

, H

2.1

, H

2.2

, and

). In addition, we performed supplementary statis-

tical analysis to identify any emotional states as pre-

dictors for music ratings.

RQ1: Music versus Audiobook.

In the beginning, we tested our ﬁrst hypothesis H

that recreationally active runners assign substantially

greater positive ratings to the M than to the A condi-

tion on various musical components (rhythm, style,

melody, tempo, instrumentation, and beat) in self-

paced running. Thereby, we identiﬁed a large effect

(t(30) = 4.44; p = .000; Cohen’s d = .80; Power =

99%) in favor for the M condition (M = 27.84; SD =

7.01) compared to the A condition (M = 20.81; SD =

8.59). Therefore, we could assume that the M condi-

tion is substantially higher rated than the A condition,

what yielded to an acceptance of H

RQ2: Difference between Emotional States be-

tween the Running Conditions.

To investigate our second RQ, we thematically subdi-

vided our ﬁve self-report emotional state dimensions

into positive (i.e., excitement and happiness; cf. H

2.1

)

and negative (i.e., anxiety, dejection, and anger; cf.

2.2

) emotional state assessments. For each of the ﬁve

emotional state dimensions a separate statistical anal-

ysis with the Friedman test was performed to compare

all three running conditions M, A, and C. Because

of multiple testing, we applied the Bonferroni proce-

dure to avoid alpha inﬂation and adjusted the nominal

alpha level to α

Bon f.corr.

= .01 for each condition to

maintain the overall alpha level of α = .05.

Positive Emotional States. We tested our second hy-

pothesis H

2.1

if runners assign signiﬁcantly more pos-

itive ratings to the positive self-report emotional state

dimensions excitement and happiness in the M condi-

tion compared to the A and the C condition, while we

assumed no difference between A and C. Our results

indicate no difference between all three conditions for

excitement (χ

(2, N = 32) = 1.63, p = .443) and for

happiness (χ

(2, N = 32) = 4.5, p = .105), respec-

tively. Hence, we could not conﬁrm our hypothesis

2.1

Negative Emotional States. Henceforward, we

tested our third hypothesis H

2.2

if runners assign sig-

niﬁcantly more positive ratings to the negative self-

report emotional state dimensions anxiety, dejection,

and anger in the M condition compared to the A

and the C condition, whereby we expect no differ-

ence between A and C. Based on the necessary Bon-

ferroni correction to control for the family-wise er-

ror rate, we found no difference between all three

conditions for either of the emotional state dimen-

sions anxiety (χ

(2, N = 32) = 4.67, p = .097), de-

jection (χ

(2, N = 32) = 6.07, p = .048), and anger

(χ

(2, N = 32) = 3.89, p = .143). This ﬁnding did

not yield an assumption of our hypothesis H

2.2

RQ3: Associations of Emotional States and Au-

dio Ratings. Additionally, we analysed whether the

dimensions excitement and happiness are positively

associated with audio ratings of the BMRI-2, and if

the dimensions anxiety, dejection, and anger are neg-

atively associated with audio ratings in both auditory

condition M and A (H

). We found particularly strong

and positive correlations for the positive dimensions

excitement and happiness in both conditions M and

A (cf. Table 1). Moreover, we uncovered consis-

tently negative but weak correlations for the dimen-

sions anxiety, dejection, and anger for the M and A

condition, respectively. Thereby, only the dimension

anger in the M condition revealed a sufﬁciently large

effect, whereas the others showed no signiﬁcant rela-

tionship (cf. Table 1). Therefore, the hypothesis H

can only be partly conﬁrmed, but especially for the

positive dimensions excitement and happiness.

Impact of Auditory Listening on Emotional States in Self-paced Outdoor Running

193

Table 1: Spearman Correlations of Emotional State Dimensions (SEQ) and Audio Ratings (BMRI-2).

Emotional State Dimensions

Audio Ratings Anxiety Dejection Excitement Anger Happiness

Music Condition r -.21 -.31 .55* -.45* .51*

p .250 .082 .001 .009 .003

Audiobook Condition r -.22 -.27 .64* -.28 .70

p .233 .149 .000 .132 .000

Note. N = 32. *Signiﬁcant at the Bonferroni-corrected level α = .01 (two-sided) for each condition.

Supplementary Statistical Analysis: Prediction of

BMRI-2. Based on the previous correlational ﬁnd-

ings, we were interested whether any of the ﬁve emo-

tional state dimensions may positively or negatively

predict the auditory rating of the BMRI-2 (criterion

variable). We performed multiple linear regression

analyses with stepwise selection. We identiﬁed the

emotional state dimension happiness as a stable pre-

dictor for positive BMRI-2 ratings for the M (ex-

plained variation R

= .31) and the A condition (ex-

plained variation R

= .48; cf. Table 2).

4 DISCUSSION

The primary purpose of this study was to investigate

whether self-paced running with music or an audio-

book is perceived and rated more positive than the

other condition (RQ1.) In line with prior ﬁndings

in the literature (Miller et al., 2010; Bigliassi et al.,

2017; Bigliassi et al., 2019; Stork et al., 2019) we

could conﬁrm, that in the M condition substantially

more positive audio ratings were assigned when com-

pared to the A condition. However, none of the

previous studies focused on self-paced running with

a researcher-selected audiobook with an overwhelm-

ingly rating as funny. As we expected, the potential

additional humorous induction of the funny audio-

book did not affect auditory ratings in terms of per-

ceived inspiration for self-paced running. In contrast,

the overall total score of the BMRI-2 of all runners

can be categorized as oudeterous (< 24, neither moti-

vating nor demotivating; cf. Karageorghis, 2008) for

the A condition (M = 20.81), and moderately inspir-

ing (range 24 to 35) for the M condition (M = 27.84).

Our secondary purpose of the present study was to

examine if different self-paced running conditions M,

A, and C may evolve more positive or more negative

emotional states (RQ2.) Several studies have found

smaller or larger effects of music of enhancing

the reallocation of attentional focus to distract and

dissociate from the sensation of boredom, pain, and

fatigue (Morgan and Pollock, 1977; Atkinson et al.,

2004; Edworthy and Waring, 2006; Karageorghis

and Priest, 2012b; Bigliassi et al., 2015; Bigliassi

et al., 2017; Bigliassi et al., 2019). However, our

investigation focused only on a potential perceived

and self-reported emotional state difference among

the three conditions M, A, and C, and not on the pos-

sible reallocation of attentional focus. Nevertheless,

our ﬁndings indicated no difference in self-reported

emotional state dimension in any of the three condi-

tions. Hence, we could not identify any substantial

beneﬁts in favor for any auditory condition, neither

M, nor A, and we have to acknowledge that running

with no-audio may elicit roughly the same emotional

states in self-paced outdoor running.

Our third purpose was to assess any potential as-

sociations among self-reported emotional states such

as anxiety, dejection, excitement, anger, and happi-

ness in the M and A condition and the self ratings

for the preferences of the auditory conditions (RQ3).

Of course, we did not focus on any associations in

the C (baseline) condition with no-audio, because of

no auditory ratings in this condition. As expected,

more positive auditory ratings were associated with

more positive ratings of the positive emotional state

dimensions excitement and happiness. This ﬁnding

is very plausible, also because of the negative,

though weak and not signiﬁcant correlations of the

auditory ratings (except anger in the M condition, cf.

Table 1) and the negatively deemed emotional state

dimensions anxiety, dejection, and anger (the later

only for the A condition). Hence, we may conclude

that negative emotions may have a slight but potential

impact on negative auditory ratings.

Based on the correlational ﬁndings (cf. RQ3),

we conducted additional supplementary statistical

analysis to possible identify any emotional state(s)

as predictor(s) for auditory ratings. Henceforward,

we performed a multiple linear regression analysis

with stepwise selection of any variable as potential

predictor of the criterion variable (BMRI-2). In both

auditory conditions M and A, the perceived and

self-reported emotional state happiness was iden-

tiﬁed as solid predictor (also in terms of explained

variations, cf. R

= .31 for the M and R

= .48 for

the A condition; cf. Table 2). Interestingly, though

the emotional state dimension excitement revealed

substantial positive associations with the auditory rat-

icSPORTS 2020 - 8th International Conference on Sport Sciences Research and Technology Support

194

Table 2: Multiple Linear Regression with Stepwise Selection: Prediction of Audio Ratings (BMRI-2).

Condition Variable

B SE t p 95% CI R

Music Happiness 4.84 1.33 3.64 .001* [2.13, 7.56] .31

Anxiety (excluded as predictor by stepwise selection)

Dejection (excluded as predictor by stepwise selection)

Excitement (excluded as predictor by stepwise selection)

Anger (excluded as predictor by stepwise selection)

Audiobook Happiness 5.49 1.06 5.17 .000* [3.32, 7.66] .48

Anxiety (excluded as predictor by stepwise selection)

Dejection (excluded as predictor by stepwise selection)

Excitement (excluded as predictor by stepwise selection)

Anger (excluded as predictor by stepwise selection)

Note. N = 32. Unstandardized regression coefﬁcient B and standard error (SE);

CI = conﬁdence interval;

Emotional state dimensions of SEQ.

*p < .05.

ings (cf. Table 1), in both linear regression analyses

it was excluded as predictor as well as the emotional

state dimensions anxiety, dejection, and anger.

Although several studies have been con-

ducted with researcher-selected (also referred to

as experimenter-selected) music-playlists (Atkinson

et al., 2004; Edworthy and Waring, 2006; Miller

et al., 2010; Stork et al., 2019), we cannot exclude

that personally selected songs may have elicited more

positive affective reactions. However, it was found

that in a direct comparison of experimenter-selected

versus self-selected music in sports, participants

assigned roughly the same ratings (Moss et al.,

2018). Conversely, this might prove difﬁcult if the

sample size would be increased, and it would be

critical to create a seriously standardized M condition

what could negatively inﬂuence potential positive

ﬁndings. In this case, it should be ensured to select

only music with the same BPM of the same genre

with the same loudness (db).

Another potential limitation is that the audiobook

excerpt chosen for the experiment was no classic

novel narrative, but a live recording of a reading by

the author, including background noises and audience

laughter. Some test persons stated that they experi-

enced this laughter as irritating or odd for the context

of running. This aligns with qualitative responses

from participants who stated that the narrator’s voice

appeared monotonous, as they missed the usual

driving forces of rhythm and melody. Research has

identiﬁed that responsiveness to music in exercise

stems from musical qualities such as rhythm, melody

and harmony (Karageorghis, 2016). Nevertheless,

the majority of participants post-activity stated that

the audiobook was quite pleasant and humorous.

Some participants have expressed that the story was

so capitivating that they forgot about the time when

running. What should be noted in this regard is that

the majority of the participants reported no previous

use of audiobooks during running. Listening to an

audiobook may therefore have elicited a novelty

effect that would subsequently have been reduced

with repeated exposure.

Apart from this, in future studies, a single-item

liking scale, ranging from 1 (I do not like it at all)

to 10 (I like it very much), could be considered to

identify the degree to which participants preferred

the condition (Bigliassi et al., 2017).

5 CONCLUSION

For the last decades, running has made the use of

portable technologies such as listening to music for

individual training very popular. However, the newly

emerging trend of listening to audiobooks while run-

ning has only been scantly researched. So far there

has been no study focusing on emotional states and

listening to humorous audiobooks in recreational self-

paced outdoor running. Primarily, our results revealed

a clear preference in favor for auditory self-ratings

of self-paced outdoor running with music when com-

pared to audiobooks. Interestingly, we identiﬁed no

self-reported affective differences in the emotional

state dimensions such as anxiety, dejection, excite-

ment, anger, and happiness across all three conditions

when running with music, audiobooks, and no-audio

(baseline), respectively. Hence, we assume that self-

paced outdoor running with no auditory listening may

elicit the same emotional states as evoked by self-

paced outdoor running music or audiobooks.

In future research, it would be interesting to study

the music-emotion and the audiobook-emotion link

also under high physical exertion. Additionally, in

contrast to many studies, it would be important to in-

vestigate whether outdoor running with an audiobook

Impact of Auditory Listening on Emotional States in Self-paced Outdoor Running

195

may have a positive effect in running sessions with a

longer duration than 10 to 15 minutes. Furthermore,

different genres of audiobooks such as novels from

mystery, to thriller, and biographies should also be

considered.

Finally, we conclude that even simple self-paced

running in the nature with no-audio may elicit the

same emotional states as when compared to running

with portable auditory technologies such as music and

audiobooks.

REFERENCES

Atkinson, G., Wilson, D., and Eubank, M. (2004). Effects

of music on work-rate distribution during a cycling

time trial. International Journal of Sports Medicine,

25(08):611–615.

Audio Publishers Association (2018). Consumer

Sales Survey: Another Banner Year of Robust

Growth for the Audiobook Industry. Retrieved

from https://www.audiopub.org/uploads/pdf/2018-

Consumer-Sales-Survey-Final-PR.pdf.

Audio Publishers Association (2019). Consumer Sales

Survey: New Survey Shows 50% of Americans

have listenend to an audiobook. Retrieved from

https://www.audiopub.org/uploads/pdf/Consumer-

Survey-Press-Release-2019-FINAL.pdf.

Bigliassi, M., Karageorghis, C. I., Hoy, G. K., and Layne,

G. S. (2019). The way you make me feel: psycholog-

ical and cerebral responses to music during real-life

physical activity. Psychology of Sport and Exercise,

41:211–217.

Bigliassi, M., Karageorghis, C. I., Wright, M. J., Orgs, G.,

and Nowicky, A. V. (2017). Effects of auditory stimuli

on electrical activity in the brain during cycle ergom-

etry. Physiology & behavior, 177:135–147.

Bigliassi, M., Le

on-Dom

ınguez, U., Buzzachera, C. F.,

Barreto-Silva, V., and Altimari, L. R. (2015). How

does music aid 5 km of running? The Journal of

Strength & Conditioning Research, 29(2):305–314.

Bull, M. (2015). Sound moves: iPod culture and urban

experience. Routledge.

Edworthy, J. and Waring, H. (2006). The effects of music

tempo and loudness level on treadmill exercise. Er-

gonomics, 49(15):1597–1610.

Have, I. and Pedersen, B. S. (2012). Conceptualising

the audiobook experience. SoundEffects-An Interdis-

ciplinary Journal of Sound and Sound Experience,

2(2):79–95.

Have, I. and Pedersen, B. S. (2015). Digital audiobooks:

New media, users, and experiences. Routledge.

IBM Corp. (Released 2017). IBM SPSS Statistics for Win-

dows, Version 25.0. Armonk, NY: IBM Corp.

Jones, M. V., Lane, A. M., Bray, S. R., Uphill, M., and

Catlin, J. (2005). Development and validation of the

sport emotion questionnaire. Journal of Sport and Ex-

ercise Psychology, 27(4):407–431.

Karageorghis, C. (2008). The scientiﬁc application of music

in sport and exercise, pages 109–137. Hodder Educa-

tion.

Karageorghis, C. I. (2016). Applying music in exercise and

sport. Human Kinetics.

Karageorghis, C. I. and Priest, D.-L. (2012a). Music in the

exercise domain: a review and synthesis (Part I). In-

ternational Review of Sport and Exercise Psychology,

5(1):44–66. PMID: 22577472.

Karageorghis, C. I. and Priest, D.-L. (2012b). Music in the

exercise domain: a review and synthesis (Part II). In-

ternational Review of Sport and Exercise Psychology,

5(1):67–84. PMID: 22577473.

Karageorghis, C. I., Priest, D.-L., Terry, P. C., Chatzisaran-

tis, N. L., and Lane, A. M. (2006). Redesign and initial

validation of an instrument to assess the motivational

qualities of music in exercise: The brunel music rating

inventory-2. Journal of sports sciences, 24(8):899–

909.

Karageorghis, C. I., Terry, P. C., Lane, A. M., Bishop, D. T.,

and Priest, D.-l. (2012). The bases expert statement on

use of music in exercise. Journal of sports sciences,

30(9):953–956.

Kreitz, G. and Niemela, F. (2010). Spotify–large scale, low

latency, p2p music-on-demand streaming. In 2010

IEEE Tenth International Conference on Peer-to-Peer

Computing (P2P), pages 1–10. IEEE.

Miller, T., Ann, M. S., Robertson, R. J., and Wheeler, B.

(2010). Effect of music and dialogue on perception of

exertion, enjoyment, and metabolic responses during

exercise. International Journal of Fitness, 6(2).

Morgan, W. P. and Pollock, M. L. (1977). Psychologic char-

acterization of the elite distance runner. Annals of the

New York Academy of Sciences, 301(1):382–403.

Moss, S. L., Enright, K., and Cushman, S. (2018). The in-

ﬂuence of music genre on explosive power, repetitions

to failure and mood responses during resistance exer-

cise. Psychology of Sport and Exercise, 37:128–138.

Rubery, M. (2011). Audiobooks, literature, and sound stud-

ies, volume 31. Routledge.

Spotify USA Inc (2020). Spotify Company Info. Retrieved

from https://newsroom.spotify.com/company-info/.

Stork, M. J., Karageorghis, C. I., and Ginis, K. A. M.

(2019). Let’s go: psychological, psychophysical,

and physiological effects of music during sprint in-

terval exercise. Psychology of Sport and Exercise,

45:101547.

Terry, P. C. and Karageorghis, C. I. (2011). Music in sport

and exercise.

icSPORTS 2020 - 8th International Conference on Sport Sciences Research and Technology Support

196