See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/51603296

Pleasure to Play, Arousal to Stay: The Effect of Player Emotions on Digital Game
Preferences and Playing Time

Article  in  Cyberpsychology, Behavior, and Social Networking · August 2011

DOI: 10.1089/cyber.2010.0040 · Source: PubMed

CITATIONS READS

61 955

4 authors, including:

Karolien Poels Wijnand A Ijsselsteijn

University of Antwerp Eindhoven University of Technology
124 PUBLICATIONS   4,522 CITATIONS    296 PUBLICATIONS   10,471 CITATIONS   

SEE PROFILE SEE PROFILE

Yvonne De Kort
Eindhoven University of Technology
165 PUBLICATIONS   4,782 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Human Factors and User Experiences of Stereoscopic 3D displays View project

Telepresence, Immersion & Embodiment in Mediated Environments View project

All content following this page was uploaded by Yvonne De Kort on 06 June 2014.

The user has requested enhancement of the downloaded file.

CYBERPSYCHOLOGY, BEHAVIOR, AND SOCIAL NETWORKING
Volume 15, Number 1, 2012 ORIGINAL ARTICLES
ª Mary Ann Liebert, Inc.
DOI: 10.1089/cyber.2010.0040

Pleasure to Play, Arousal to Stay: The Effect

of Player Emotions on Digital Game Preferences
and Playing Time

Karolien Poels, Ph.D.,1 Wouter van den Hoogen, Ph.D.,2 Wijnand Ijsselsteijn, Ph.D.,2 and Yvonne de Kort, Ph.D. 2

Abstract

This study investigated how player emotions during game-play, measured through self-report and physiological
recordings, predict playing time and game preferences. We distinguished between short-term (immediately after
game-play) and long-term (after 3 weeks) playing time and game preferences. While pleasure was most pre-
dictive for short-term playing time and game preferences, arousal, particularly for game preferences, was most
predictive on the longer term. This result was found through both self-report and physiological emotion mea-
sures. This study initiates theorizing about digital gaming as a hedonic consumer product and sketches future
research endeavors of this topic.

Introduction game experience,7,15,16 it seems particularly relevant to study

how emotional experiences relate to consumption related

P laying digital games is mostly an intentional activity

people engage to enjoy themselves.1 Malone2 described
digital games as ‘‘a new kind of intrinsically motivating ac-
variables like play duration and game preferences. We focus
on the emotional dimensions of pleasure, arousal, and dom-
inance (referred to as PAD-dimensions).17 The pleasure di-
tivity.’’ Intrinsic motivation is situated within the game-play mension refers to the pleasantness or enjoyment of a certain
itself (e.g., being immersed in a fantasy world), or on the level experience. The arousal dimension indicates the level of
of achievement in the game (e.g., feelings of mastery).1 Con- physical and mental activation associated with the experi-
sequences from outside the game (e.g., monetary rewards), ence. Dominance concerns feelings of control and influence
resulting from extrinsic motivations, are less important for the over others and surroundings. These types of emotions make
majority of players.1 Intrinsic motivation can be linked to up a fundamental part of the player’s emotional experiences
‘‘hedonic consumption’’.3,4 In contrast to goal-directed utili- during game-play.18–20
tarian consumption, hedonic consumption involves experi- The relation between player emotions and consumption of
encing the product, for its own sake, because the product itself digital games has only received scant attention. Kempf13
is enjoyable.5 Digital games are typically multi-sensoric,6 and found that pleasure and arousal reactions during the trial of a
relate to elements of fantasy,7 or escapism.8 Further, digital computer game positively influenced on trial evaluations,
games have the potential to trigger strong emotions9 and lead whereas for a functional grammar checker these emotions
to high levels of enjoyment.10 In other words, choosing, were of minor influence. An earlier study,21 focusing on
buying, and playing a digital game involves consuming a player emotions and preferences for games in arcade halls,
hedonic product. showed that higher feelings of both pleasure and arousal
Within marketing, hedonic consumption has already been resulted in more favorable preferences. This study only in-
widely studied.5,11,12 Research has shown that, for hedonic cluded game preferences measured immediately after a play
products, emotional responses during product trials have a session, however. Besides preferences, which reside on an
significant influence on both subsequent product attitude attitudinal level, actual playing behavior is of additional im-
formation,13 and future consumption duration.14 Despite the portance when studying consumer behavior. Further, playing
massive success of digital games, it remains largely unknown digital games has largely shifted from arcade halls to the
how players actually select games, and how game prefer- home. This makes it particularly relevant to examine how
ences are established. Given the multitude of studies ad- player emotions affect playing behavior at home and how
dressing player emotions as a fundamental part of the digital game preferences develop over time. Further, a crucial issue

1
Department of Communication Studies, University of Antwerp, Antwerpen, Belgium.
2
Human-Technology Interaction Group, Eindhoven University of Technology, Eindhoven, The Netherlands.

1
2 POELS ET AL.

when studying player emotions is the valid measurement of received 15e, the second lasting 20 minutes for which par-
a player’s feelings and experiences.22 The most common ticipants received 5e.
methods are retrospective self-report and real-time physio-
logical measures, which, due to their specific nature, are
Experimental setting, measures, and procedure
complementary to each other.18,23 In the current study, we
looked at how, and to what extent, subjective and physio- We used four PC games: two First Person Shooter games
logical indicators of player emotions during initial game-play (Battlefield 1942, Hitman Contracts) and two race games
are able to predict game preferences and actual playing (Colin McRae, Trackmania), providing variation in genre and
behavior, directly after playing and during a 3-week period realism (realistic: Battlefield and Colin McRae, fictional: Hit-
at home. man Contracts and Trackmania). All four games were highly
rated, relatively cheap, and had reasonable system require-
Hypotheses and Research Questions ments.
After introduction (session one) to the experiment, partic-
Hedonic consumption literature has shown that the ipants were connected to physiological sensors, using a TMSi
amount of pleasure experienced during consumption was Mobi 6 Bluetooth device. The measures included Skin Con-
predictive for later consumption.14 In the context of gaming, ductance Level (SCL), as an indicator of arousal, and zygo-
previous studies indicated the importance of pleasure on later maticus major and currogator supercilii electromyography
game evaluations.13,21 Therefore, we formulate the following (EMG) measures, as indicators of (dis)pleasure. Recordings
hypothesis: were taken at 1,024 Hz. After data collection, values were
filtered following suggestions from Tassinary and Ca-
H1: Pleasure experienced while playing a digital game will be cioppo.28 After filtering, the values were log transformed and
predictive for later playing behavior in a positive way.
baseline corrected to cope with individual differences in skin
In general emotion24,25 and consumer literature,26 arousal conductivity and EMG activation.
is considered as a trigger of behavior. In the context of Participants played each of the four games for 10 minutes
gaming, arousal is one of the main reasons why people play.7 in a counterbalanced order. After each play session, partici-
This leads us to formulate the following hypothesis: pants rated their experiences on the Self-Assessment Manikin
(SAM-scale), a visual self-report scale29 based on the PAD-
H2: Arousal experienced while playing a digital game will be dimensions of Mehrabian and Russell17 and includes three
predictive for later playing behavior in a positive way. nine-point visual scales on which participants have to indi-
cate how much pleasure, dominance, and arousal they felt
Dominance can be considered as the perceived ability to while playing the computer game. The SAM-scale is fre-
manipulate the game flow to one’s goals. This control has quently used to measure emotions in general emotion stud-
been linked to overall game enjoyment27 and, for male ies,24 and in consumer30 and gaming research.31,32
players, related to game preference. We expect that: Next, participants were invited to another lab room in
which each of the four games was installed on different PC’s.
H3: Dominance experienced while playing a digital game will They were instructed to freely choose which game they
be predictive for later playing behavior in a positive way. played, and switch between different games, for 30 minutes.
A camera in the corner of the room allowed us to observe
There are several issues that, although not investigated in playing time for each of the games. Although we aimed to fix
previous studies, are crucial to get a fuller understanding of the time at 30 minutes, this was not always practically feasi-
player emotions and the prediction of playing behavior. ble. The exact playing time varied between participants (Min:
Therefore, we formulate following research questions: 23 minutes 35 seconds; Max: 31 minutes 35 seconds). For the
main analysis we therefore created a relative measure, di-
viding the time played on each game by the total time played
RQ1: What is the unique predictive contribution of the (DV1: short-term playing time). After the 30 minutes free-
different emotional dimensions of pleasure, arousal, and play session, participants ranked all four games according to
dominance? three parameters, asking which game they (1) liked most, (2)
RQ2: Are there differences related to the preference time wanted to keep, and (3) wanted to buy. For every parameter,
frame (short-term vs. long-term playing behavior)? score 1 was given to the most preferred game and then in
RQ3: Are there differences related to the emotion measure ascending order a score of 2, 3, or 4 to the other games ac-
type (retrospective self-report vs. real-time physiological cording to their relative preference on the different parame-
emotion measures)? ters. For our analysis, we averaged the scores on these three
parameters (Chronbach’s a = 0.94) and recoded the scores
Method such that a score of 4 represented maximum preference and a
score of 1 the lowest preference. This led to a general ‘‘game
Participants
preference’’ variable (DV2: short-term game preference).
Nineteen participants (7 females) aged between 18 and 42 Participants were then briefed about the second stage of the
(Mage = 23.47 years, SD = 7.24) took part in the experiment. experiment in which they had to take the four games home
Game-play frequency ranged from ‘‘a couple of times a year’’ and were requested to play these games during 3 weeks and
(n = 4), ‘‘monthly’’ (n = 4), ‘‘weekly’’ (n = 5) to ‘‘daily’’ (n = 6). keep a diary of their playing behavior. Total time spent
The study had a within subject design, and involved two lab playing the four games varied considerably among partici-
sessions: the first lasting 90 minutes for which participants pants (Min: 114 minutes, Max: 4,342 minutes). Therefore, we
PLAYER EMOTIONS AND DIGITAL GAME PREFERENCES 3

employ relative playing time in our analysis, dividing the

2.28 (0.99)

2.4 (1.07)

1.88 (0.94)

3.44 (0.69)
1/3.33
playing time for each game by the total time spent playing the

1/4

1/4

2/4
Game preferenceb

LT
four games under study (DV3: long-term playing time). After
3 weeks, the second lab session took place. Again, we asked
participants to rank which game they liked most, wanted to

2.04 (0.91)

2.32 (1.08)

2.19 (1.03)

3.46 (0.58)
keep, and wanted to buy. These scores and recoded them to

1/3.67

2.33/4
compose a second ‘‘game preference’’ variable (Chronbach’s

1/4

1/4
ST
a = 0.96) (DV4: long-term game preference).

Results

0.11 (0.25)

0.16 (0.28)

0.08 (0.12)

0.65 (0.4)
0/0.33
0/1

0/1

0/1
Table 1. Descriptive Statistics for Criterion and Dependent Variables on the Four Digital Games
Descriptive statistics

LT
Playing timea
We first report a general overview of the main descrip-
tive statistics for all four games on each criterion and de-
pendent variable employed in this study (see Table 1).

0.2 (0.31)

0.15 (0.33)

0.1 (0.23)

0.55 (0.42)
0/0.95

0/0.70
Although Trackmania was—on average—most liked and

0/1

0/1
ST
played, there was considerable variation between partici-
pants. None of the four games was really disliked or left
unplayed.

Skin conductance

- 0.035 (0.054)

- 0.01 (0.057)

0.05 (0.047)
- 0.0089 (0.07)
- 0.12/0.04

- 0.18/0.08

- 0.04/0.14
Core analysis

- 0.1/0.08
We used four different games with the aim of inducing a
wide variety in player emotions. Therefore, in the remainder
of our analysis, we aggregated data across games, with the
goal of maximizing the variance in our data. We first re-
structured our data file such that the scores for the different

Corrugator supercilii
games were treated as separate cases, creating four rows of
Psychophysiology

0.03 (0.11)

0.04 (0.22)

0.08 (0.26)

0.0077 (0.19)
- 0.93/0.24

- 2.25/2.25
data for each participant. Since our data were now ‘‘nested

- 0.42/2.5
- 21/2.25
within participants,’’ we applied a Linear Mixed Model
(LMM) analyses including participant number as a random
factor in our analyses, allowing us to control for differences in
variance that solely reside at the level of participants. We ran
a total of eight LMM analyses divided in two sets. In the first
set, we ran four LMM analyses containing one of the de-
Zygomaticus major

pendent variables (DV1–DV4) as criterion and the self-report

- 0.0016 (0.16)

- 0.07 (0.17)

- 0.017 (0.24)

0.09 (0.22)
- 0.48/0.15

- 0.65/0.56

- 0.18/0.82
measures (pleasure, arousal, and dominance) as predictors.
- 0.28/0.4

In the second set, we ran four LMM analyses containing one

of the dependent variables (DV1–DV4) as criterion and the
physiological measures (EMG: zygomaticus major, corru-
gator supercilli, and SCL) as predictors. Before analyses, the
criterion and dependent variables were standardized. Re-
gression coefficients were calculated indicating the direction
Dominance

5.11 (2.23)

4.58 (1.46)

5.89 (2.05)
6.05 (1.9)

and strength of the relation between the predictor and crite-

1/9

1/9

2/8

1/9

rion variables.

Short-term playing time

4.63 (2.43)

4.88 (2.01)

5.84 (1.92)
Self-report

For the self-reported player emotions, only pleasure had a

4.42 (1.5)
Arousal

1/9

1/8

1/7

2/9

positive effect on short-term playing time (b = 0.31,

t(72) = 2.61; p = 0.011). Physiological measures did reveal a
similar, although marginally significant effect, of zygomati-
After averaging and recoding.

cus major activity (b = 0.20, t(72) = 1.69; p = 0.096) on short-

6 (1.89)

5.26 (2.66)

7.26 (1.41)

term playing time.

5.37 (1.5)
Pleasure

2/9

1/9

3/8

4/9

Relative playing time.

Long-term playing time

Pleasure (b = 0.26, t(64) = 2.13; p = 0.037) showed a positive
Mean (SD)

Mean (SD)

Mean (SD)

Mean (SD)

effect for long-term playing time. The analysis of the physi-

ColinMcRae
Min/Max

Min/Max

Min/Max

Min/Max
Trackmania

ological measures showed a significant effect of zygomaticus

Battlefield

major activity (b = 0.23, t(64) = 2.05; p = 0.045), and a positive,

Hitman

yet only marginally significant, effect of SCL (b = 0.22,

b
a

t(67) = 1.83; p = 0.072) on long-term playing time.

4 POELS ET AL.

Short-term game preferences the game and ability to anticipate and influence the game
flow.27 This is typically the result of frequent practice and
For the self-reported player emotions, pleasure (b = 0.36,
game expertise. In an arcade context, games were relatively
t(72) = 3.16; p = 0.002) had a significant positive effect on short-
easy and could thus induce feelings of dominance quite
term game preference. Results with the physiological mea-
rapidly. In the current experiment, we let people play four
sures showed a similar pattern in which zygomaticus major
more complex PC games for only 10 minutes and previous
activity showed a positive significant effect (b = 0.31,
experience with the games was nonexistent or very low. Our
t(72) = 2.77; p = 0.007) on short-term game preferences.
research design might not have been suited well enough to
cater for dominance as a player emotion. Although, in con-
Long-term game preferences sumer studies, dominance is generally given less attention
For long-term game preference, self-reported pleasure compared to pleasure and arousal,33,34 future studies should
(b = 0.21, t(72) = 1.78; p = 0.079) was only marginally signifi- be able to manipulate dominance levels more explicitly and
cant, yet self-reported arousal (b = 0.27, t(72) = 1.26; p = 0.023) further unravel the predictive role of dominance in playing
revealed a significant effect. The latter was also found in the behavior.
analysis with the physiological measures: SCL had a positive Theoretically, it was to be expected that corrugator su-
effect on long-term game preference (b = 0.35, t(72) = 3.1; percilii activation, as an indicator of displeasure, would yield
p = 0.003), whereas zygomaticus major results were not sig- opposite relations with playing behavior compared to zygo-
nificant. maticus major activation. However, results did not reveal any
Table 2 below summarizes the results found. Self-reported significant relation between corrugator supercilii activation
dominance and corrugator supercilii activity did not show and the playing variables. Playing games is in itself not an
significant predictive relationships with the playing behavior unpleasant activity and the games included in our study were
variables. all highly rated games. As such, the games might not have
been able to generate (extreme enough) negative emotions.
Future studies could focus on (dis)pleasure in relation to
Discussion and Conclusion
specific game events (e.g., failures or successes on specific
The results confirm our expectation that player emotions points in the game) and how these predict later playing be-
during initial game-play have the potential to predict playing havior.
behavior at a later stage. We focused on two parameters of Two, seemingly ambivalent findings, deserve further ex-
playing behavior: game preferences and playing time. For ploration. First, arousal was found to significantly predict
game preferences, a clear pattern emerged. While pleasure long-term game preferences yet contributed little to short-
during initial game-play influenced short-term game prefer- term game preferences. The opposite was found for pleasure.
ence, arousal contributed to long-term game preference. This This finding indicates that the time frame is an important
finding surfaced through both self-report as physiological aspect to take into account when studying game preferences.
measures. For playing time, we found that pleasure experi- The unique contribution of pleasure and arousal clearly dif-
enced at the initial play session was strongly related to fers according to the time frame under consideration. Al-
playing time, both in the lab and during a 3-week period at though only speculative, it could be that on the short-term,
home. Again, this pattern was consistent for both the self- after the lab session, players opted for the most pleasurable
report and the physiological measures. Arousal, on the other games. When playing at home, they further took the time to
hand, did not predict any short-term playing time and added get to know the games that posed more challenges (and were
only marginally to long-term playing time. more arousing), leading to a preference for these games in the
The hypothesis that dominance, as a player emotion, is longer run. Second, results were not always consistent be-
predictive for later playing behavior was not supported. The tween game preferences and actual playing time. Arousal
arcade hall study of Mehrabian and Wixen21 did, however, positively predicted long-term game preferences, but not
reveal that dominance experienced while playing an arcade long-term playing time. Pleasure, on the other hand, posi-
game was positively related to game preferences. Dominance, tively predicted long-term playing time but showed no sig-
as a sense of control, refers to a player’s sense of mastery of nificant relation to long-term game preferences. These

Table 2. Linear Mixed Model Results for Short-Term and Long-Term Playing
Time and Game Preference as a Function of Player Emotions

Short-term Long-term

Pleasure Arousal Pleasure Arousal

Playing time Self-report 0.31* 0.12 0.26* 0.19

Physiology 0.20 + - 0.03 0.23* 0.22 +
Game preference Self-report 0.36** 0.17 0.21 + 0.27*
Physiology 0.31** 0.04 0.04 0.35**

All LMM analyses included 3 predictors (self-report: SAM Pleasure, Arousal, and Dominance; physiology: EMG zygomaticus major, EMG
corrugator supercilii, SCL). Since Dominance, and EMG corrugator supercilii did not show any significant results, we do not report their
values in the table.
*p < 0.05, + p < 0.1, **p < 0.01.
LMM, Linear Mixed Model; SAM, Self Assessment Manikin; SCL, Skin Conductance Level.
PLAYER EMOTIONS AND DIGITAL GAME PREFERENCES 5

findings show the importance of considering both evaluative 2. Malone TW. Toward a theory of intrinsically motivating
and behavioral parameters of gaming. In (social) psychology, instruction. Cognitive Science 1981; 4:333–369.
there is a long line of research on the (in)consistency between 3. Deci EL, Ryan RM. (1985) Intrinsic motivation and self-
attitudes and behavior.35 It is yet unclear what explains the determination in human behavior. New York: Plenum Press.
inconsistencies in predictive relationships of player emotions 4. Holbrook MB, Gardner MP. How motivation moderates the
between the two parameters of playing behavior. We must effects of emotions on the duration of consumption. Journal
acknowledge that the measurement of long-term playing of Business Research 1996; 42:241–52.
time at home was (partly) beyond our control and some 5. Hirschman EC, Holbrook MB. Hedonic consumption:
participants mentioned that other responsibilities like school, emerging concepts, methods, and proposition. Journal of
Marketing 1982; 46:92–101.
work, and family, had interfered with their (preferred) game-
6. Bolter D, Grusin R. (1999) Remediation: understanding new
play intensity. As such, actual long-term playing behavior
media. MA: The MIT Press.
might have been different if it were measured on another
7. Sherry JL, Lukas K, Greenberg B, et al. (2006) Video game
(more convenient) time span. Future studies are needed to uses and gratifications as predictors of use and game pref-
shed a light on the underlying decisional process for and the erence. In Vorderer P, Bryant J, eds. Playing videogames:
(in)consistencies between both parameters of playing be- motives, responses, and consequences. NJ: Lawrence Erlbaum
havior. Associates, pp. 213–224.
The current study already sets a first step toward theoriz- 8. Jansz J, Martens L. Gaming at a LAN event: the social con-
ing about the hedonic consumption of digital games some text of playing video games. New Media & Society 2005;
limitations need to be considered, however. First, the current 7:333–355.
set up does not allow to fully claim causality between player 9. Poels K, de Kort YAW, IJsselsteijn WA. (2007) ‘‘It is always a
emotions and playing behavior. Our stimuli consisted of four lot of fun!’’ Exploring Dimensions of Digital Game Experi-
different games that were not a priori expected to yield spe- ence using Focus Group Methodology, Proceedings of Fu-
cific (intensities in) emotions in our participants. Our aim was tureplay 2007, Toronto [CD Rom], Canada, pp. 83–89.
to induce enough variance in emotions within each partici- 10. Philips CA, Rolls S, Rouse A, et al. Home video game-
pant, so we could link these variables to their later playing playing in school-children: a study of incidence and patterns
behavior. Also, other variables that we could not control for of play. Journal of Adolescence 1995; 18:687–691.
could possibly interfere with the predictive relations found. 11. Babin BJ, Darden WR, Griffin M. Work and/or fun: mea-
For example, game reviews, word-of-mouth, personal inter- suring hedonic and utilitarian shopping value. Journal of
ests, and social setting are expected to have an impact on Consumer Research 1994; 20:644–656.
what kind of games players prefer and play. Further, given 12. Wakefield KL, Barnes JH. Retailing hedonic consumption: a
model of sales promotion of a leisure service. Journal of
the time intensive nature of our study, our sample size was
Retailing 1996; 72:409–427.
rather small. Although we strived for variation in age, gen-
13. Kempf DS. Attitude formation from product trial: distinct
der, and gaming frequencies, we cannot claim representa-
roles for cognition and affect for hedonic and functional
tiveness for the general gaming population. Finally, we only products. Psychology & Marketing 1999; 16:35–50.
investigated the predictive potential of emotional dimensions 14. Menon S, Kahn B. Cross-category effects of induced arousal
as player emotions. We did not include specific game expe- and pleasure on the Internet shopping experience. Journal of
riences that have been identified as contributing to the in- Retailing 2002; 78:31–40.
trinsic motivational qualities of digital games, for example, 15. Mandryk RL, Atkins MS. A fuzzy physiological approach
immersion,36 control,27 and flow.37 Future studies should for continuously modeling emotion during interaction
unravel how these factors interact with player emotions and with play technologies. International Journal of Human-
which unique contribution they have for the study of playing Computer Studies 2007; 65:329–347.
behavior. 16. Ravaja N, Turpeinen M, Saari T, et al. The psychology of
James Bond: phasic emotional responses to violent video
Acknowledgments game events. Emotion 2008; 8:114–120.
17. Russell JA, Mehrabian A. Evidence for a three-factor theory
The authors gratefully acknowledge financial support of emotions. Journal of Research in Personality 1977; 11:
from the European Commission’s Framework 6 IST program. 273–294.
In particular, the work reported here has been supported by 18. van Reekum CM, Johnstone T, Banse R, et al. Physiological
the FUGA project (part of the IST–New and Emerging Science responses to appraisal dimensions in a computer game.
and Technology program) and the Games@Large project Cognition and Emotion 2004; 18:663–688.
(part of the IST–Networked Audio-Visual Systems and Home 19. Ravaja N, Saari T, Salminen M, et al. Phasic emotional re-
Platforms program). action to video game events: a psychophysiological investi-
gation. Media Psychology 2006; 8:343–367.
Disclosure Statement 20. Lim S, Lee JE. When playing together feels different: effect of
task types and social context on physiological arousal in
No competing financial interests exist.
multiplayer online gaming contexts. Cyberpsychology &
Behavior 2009; 12:59–61.
References
21. Mehrabian A, Wixen W. Preferences for individual video
1. Klimmt C, Hartmann T. (2006) Effectance, self-efficacy, and games as a function of their emotional effects on players.
the motivation to play video games. In Vorderer P, Bryant J, Journal of Applied Social Psychology 1986; 16:3–15.
eds. Playing videogames: motives, responses, and consequences. 22. IJsselsteijn WA, de Kort YAW, Poels K, et al. (2007) Char-
NJ: Lawrence Erlbaum Associates, pp. 153–169. acterizing and Measuring User Experiences. ACE 2007
 6 POELS ET AL.

International Conference on Advances in Computer En- 32. van den Hoogen WM, IJsselsteijn WA, de Kort YA. (2009)
tertainment Technology, Workshop ‘‘Methods for Evaluat- Effects of sensory immersion on behavioural indicators of
ing Games—How to measure Usability and User Experience player experience: movement synchrony and controller
in Games’’, Salzburg, Austria [CD-rom]. pressure. In Atkins B, Kennedy H, Krzywinska T, eds.
23. Poels K, Dewitte S. How to capture the heart? Reviewing 20 Breaking new ground: innovation in games, play, practice and
years of emotion measurement in advertising. Journal of theory. Proceedings of the 2009 Digital Games Research Asso-
Advertising Research 2006; 46:18–37. ciation Conference. London: Brunel University, pp. 1–6.
24. Bradley MM, Codispoti M, Cuthbert BN, et al. Emotion and 33. Olney TJ, Holbrook MB, Batra R. Consumer responses to
motivation I: defensive and appetitive reactions in picture advertising: the effect of ad content, emotions, and attitude
processing. Emotion 2001; 1:276–298. towards the ad on viewing time. Journal of Consumer Re-
25. Mehrabian A, Russell JA. (1974) An approach to environmental search 1991; 17:440–453.
psychology. Cambridge, MA: MIT Press. 34. Russell JA, Pratt G. A description of the affective quality
26. Groeppel-Klein A. Arousal and consumer in-store behavior. attributed to environments. Journal of Personality and Social
Brain Research Bulletin 2005; 67:428–437. Psychology 1980; 38:311–322.
27. Klimmt C, Hartmann T, Frey A. Effectance and control as 35. Ajzen L, Fishbein M. Attitude-behavior relations: a theoret-
determinants of video game enjoyment. Cyberpsychology & ical analysis and review of empirical research. Psychological
Behavior 2007; 10:845–874. Bulletin 1977; 84:888–918.
28. Tassinary LG, Cacioppo JT. (2000) The skeletomotor sys- 36. Jennet C, Cox AL, Cairns P, et al. Measuring and defining
tem: surface electromyography. In Cacioppo JT, Tassinary the experience of immersion in games. International Journal
LG, Berntson GB, eds. Handbook of psychophysiology, 2nd of Human Computer Studies 2008; 66:641–661.
edition. New York, NY: Cambridge University Press, pp. 37. Sherry JL. Flow and media enjoyment. Communication
163–199. Theory 2004; 14:328–347.
29. Lang PJ. (1980) Behavioral treatment and bio-behavioral
assessment: computer applications. In Sidowski JB, Johnson
JH, Williams TA, eds. Technology in mental health care delivery Address correspondence to:
systems. Norwood, NJ: Ablex Publishing, pp. 119–137. Dr. Karolien Poels
30. Morris JD. Observations: SAM: the self-assessment manikin. Department of Communication Studies
Journal of Advertising Research 1995; 35:63–68. University of Antwerp
31. Chanel G, Rebetez C, Bétrancourt M, et al. (2008) Boredom, Sint Jacobstraat 2
engagement and anxiety as indicators for adaptation to 2000 Antwerpen
difficulty in games. In Proceedings of the 12th International Belgium
Conference on Entertainment and Media in the Ubiquitous era.
Tampere, Finland, pp. 13–17. E-mail: [email protected]

View publication stats